SilverHawk

Docker and MQTT (Mosquito)

2026-07-22T13:50:16.844-07:00

My Mosquito implementation is not heavily modified. In fact, it's so simple, I could do a :

docker run -d ---name mosquitto -p 1883:1883 eclipse-mosquitto:2

This would be just fine. However, I want my docker image to restart after reboots, and I'm too lazy to go about it any other way than a docker-compose.yml :

# docker-compose.yml - Mosquitto MQTT broker
services:
  mosquitto:
    image: eclipse-mosquitto:2
    container_name: mqtt
    ports:
      # Standard MQTT protocol
      - "1883:1883"
      # MQTT over WebSocket
      - "9001:9001"
    #volumes:
      # Mount configuration directory
      #- ./config:/mosquitto/config
      # Persist message data
      #- ./data:/mosquitto/data
      # Persist log files
      #- ./log:/mosquitto/log
    restart: unless-stopped

Note the entire "volumes" section has been commented out. I started to add it, and realized I didn't care about persisting message or log data, nor was I running a custom configuration. So, I left it out.

Then, start it up in the usual fashion with a :

docker compose up -d

And off you go!

Tempest Weatherflow - Data Mapping into MySQL

2026-07-22T13:26:09.699-07:00

I'm slowly migrating off of a virtual server at a service provider. Most things have been moving to docker. In this case, it's going onto the docker server, but not going to run in docker itself.

It has to be on the same network as the Weatherflow device to be able to pick up data, and me keeping it on the unrouteable network meant this docker platform needed to be on two separate networks. So, I fired off the command to create a new connection on the same interface.

nmcli connection add con-name unrouteable ifname enp1s0 type ethernet ipv4.method manual ipv4.addresses 192.168.2.10/24 ipv4.never-default yes

Unfortunately, this resulted in the system not being able to boot onto the network. So, it was backed out, and I moved the weather hub off of the unrouteable to the regular network. The code below, I simply could NOT get to arrive in any docker image, or running directly on the docker server (outside of a container). However, after a lot of troubleshooting to verify that the broadcast packets were getting there using :

sudo tcpdump -i enp1s0 udp port 50222 -Xvvv

I finally realized that RedHat had firewall-cmd running by default. Now, I'm trying to make all this work with defaults, which means that instead of disabling it, I'm going to open the port :

firewall-cmd --permanent --add-port=50222/udp
firewall-cmd --reload
firewall-cmd --list-all

You should see the ports: line now containing "50222/udp". And, if you run the server, it should show it's receiving packets. Okay, back to building this "listener". I went down this whole path of making it run an exec() to curl, which made it really easy to send the received JSON to a perl script and customize whatever you needed from there. Yes, at the very bottom of the code I have commented out a "sendJSON" function, and instead I'm using a handleJSON function that will do the parsing, so if you need to flip that switch because you are using a non-docker version of HTTPd and Perl, you can install the package and make it happen.

Since the Perl instances in the Apache/HTTPd 2.4 instances ("httpd 2.4") do NOT include either the Perl JSON module, or the DBD::MySQL module, it ultimately means you can't send this to Apache and deal with it nicely there if you are also using native docker. That leaves one choice - write a UDP "listener" (duh, UDP packages don't "listen" on UDP ports, because there is no connection/handshake).

So, I wrote the dumb thing in perl. I'd already had the UDP side, but since it might as well do the local database insertion, whelp, that's what I did.

Fancy thing is that I can specify which datagram types to accept from the Tempest system. I really only care about lightning strikes and obs_st (standard metrics). But, I slapped SQL for all, and did a fancy ".disabled" extension to hide the ones I wanted. If I don't have a disabled, when it sees a packet, it will dump the hash_ref so that I can easily write one if I need it. I just wrote what I saw, then disabled most of them.

Here's the code for the listener :

#!/usr/bin/perl -w

# written by Joe Lewis <joe@joe-lewis.com> on November 29, 2022
# A UDP listener for the tempest weather system. The API for weatherflow (maker
# of the tempest system) can be found at :
#     https://weatherflow.github.io/Tempest/api/udp.html
# if on RHEL, install :
#   perl-libwww-perl
#   perl-Sys-Syslog.x86_64
# for RHEL, this uses the template tool kit (available in the EPEL repository)
#   sudo subscription-manager repos --enable codeready-builder-for-rhel-9-x86_64-rpms
#   sudo dnf install https://dl.fedoraproject.org/pub/epel/epel-release-latest-9.noarch.rpm
#   sudo yum install perl-Template-Toolkit.x86_64


use strict;
use warnings;
use Data::Dumper;
use DBI;
use Getopt::Long;
use HTTP::Request;
use HTTP::Response;
use IO::Socket;
use JSON;
use LWP::UserAgent;
use POSIX;
use Sys::Syslog qw(:DEFAULT :standard :macros);
use Template;


my $template_directory = 'sql';
my $mysql_hostname = '127.0.0.1'; # if you use "localhost", it will try the unix socket
my $mysql_username = 'user_only_having_insert_or_update';
my $mysql_password = 'users_password';
my $mysql_database = 'weather';

my $verbose = undef;
my $daemonize = undef;

GetOptions(
  'sql-directory=s' => \$template_directory,
  'daemonize' => \$daemonize,
  'verbose' => \$verbose,
);


my($sock, $newmsg, $MAXLEN, $PORTNO);
my $json = JSON->new->allow_nonref;
my $userAgent = LWP::UserAgent->new(timeout => 10);
my $postURL = 'https://protected.silverhawk.net/cgi-bin/weatherflow.pl';
my $postContentType = 'application/json';
my $name = 'WeatherFlow-UDP-conv';
my $nagios_command_file = '/var/lib/nagios3/rw/nagios.cmd';
my $error_log_qfn = '/var/log/'.$name.'.log';
my $pid_file_qfn = '/var/run/'.$name.'.pid';

$MAXLEN = 1024;
$PORTNO  = 50222;

sub sighup_handler {
	exit;
};

sub sigusr1_handler {
	exit;
};

sub logMessage {
	my $msg = shift;
	$msg .= "\n" unless $msg =~ /\n/;
	syslog('info',$msg);
};

sub handleUDPPacket {
	my $weatherstationname = shift;
	my $weatherstationclient = shift;
	my $data = shift;
	my $href = $json->decode($data);
	my $weatherstationip = inet_ntoa($weatherstationclient);
	# just for my sanity, insert the hostname
	$href->{'client_hostname'} = $weatherstationname;
	$href->{'client_ip'} = $weatherstationip;

	my $sqlTemplate = $template_directory.'/'.$href->{type}.'.sql';
	if (-f $sqlTemplate) {
		my $tt = Template->new({INCLUDE_PATH=>$template_directory,INTERPOLATE => 1}) or die "Error launching the template toolkit : $Template::ERROR\n";
		# establish a MySQL connection
		my $dbh = DBI->connect('dbi:mysql:host='.$mysql_hostname.':database='.$mysql_database.':mysql_ssl=1',$mysql_username,$mysql_password);
		$href->{dbh} = $dbh;
		my $SQL;
		$tt->process($href->{type}.'.sql',$href,\$SQL) or warn 'Template Processing Failure: '.$tt->error()."\n";
		if ($SQL) {
			$dbh->do($SQL) or warn "Error running SQL on type $href->{type}: $DBI::errstr, SQL :\n\t$SQL\n";
		}
		$dbh->disconnect();
	} else {
		# if we have a disabled SQL file, do nothing because we're intentionally swallowing it.
		if (!-f $sqlTemplate.'.disabled') {
			# no template defined, warn it
			warn "non-existant SQL template $sqlTemplate containing:\n".Dumper($href);
		}
	};
};

openlog("",'ndelay','daemon');
if (defined($daemonize)) {
	my $pid = fork();
	exit if ($pid > 0);
	setsid();
	$pid = fork();
	exit if ($pid > 0);
};
$SIG{HUP} = \&sighup_handler;
$SIG{USR1} = \&sigusr1_handler;

$sock = IO::Socket::INET->new(
           LocalPort => $PORTNO, 
       Proto => 'udp') or die "socket: $@";
       
logMessage("Awaiting UDP messages on port $PORTNO");

while ($sock->recv($newmsg, $MAXLEN)) {
	my($clientport, $clientaddr) = sockaddr_in($sock->peername);
	my $clienthost = gethostbyaddr($clientaddr, AF_INET);
	handleUDPPacket($clienthost,$clientaddr,$newmsg);
}
my $status = $!;
logMessage("Loop completed, exiting with $status");
die "recv: $status";

Now, that's relatively simple. I created the separate directory for the SQL's specific to the Tempest functionality, and then piled a bunch of files in there :

*** sql/device_status.sql.disabled
INSERT INTO device_status (timestamp,serial_number,client_ip,client_hostname,hub_sn,voltage,rssi,hub_rssi,sensor_status,firmware_revision) VALUES (from_unixtime([% timestamp %]),[% dbh.quote(serial_number) %],[% dbh.quote(client_ip) %],[% dbh.quote(client_hostname) %],[% dbh.quote(hub_sn) %],[% voltage %],[% rssi %],[% hub_rssi %],[% sensor_status %],[% firmware_revision %])

*** sql/evt_precip.sql.disabled
INSERT INTO evt_precip (timestamp,station_id,hub_sn) VALUES (from_unixtime([% evt.0 %]),[% dbh.quote(serial_number) %],[% dbh.quote(hub_sn) %)

*** sql/evt_strike.sql
INSERT INTO lightning_events (epoch,station_id,energy,distance_km,distance_miles) VALUES (from_unixtime([% evt.0 %]),[% dbh.quote(serial_number) %],[% evt.2 %],[% evt.1 %],[% evt.1/1.6093445 %])

*** sql/hub_status.sql.disabled
INSERT INTO hub_status (timestamp,serial_number,frequency,rssi,client_ip,uptime,firmware_revision,hw_version,client_hostname,radio_version,reboot_count,i2c_bus_error_count,radio_status,radio_network_id) VALUES (from_unixtime([% timestamp %]),[% dbh.quote(serial_number) %],[% freq %],[% rssi %],[% dbh.quote(client_ip) %],[% uptime %],[% dbh.quote(firmare_revision) %],[% hw_version %],[% dbh.quote(client_hostname) %],[% radio_stats.0 %],[% radio_stats.1 %],[% radio_stats.2 %],[% radio_stats.3 %],[% radio_stats.4 %])

*** sql/obs_air.sql.disabled
INSERT INTO obs_air (timestamp,serial_number,firmware_revision,hub_sn,station_pressure_mb,station_pressure_inhg,temperature_celcius,temperature_farenheit,rel_humidity,lightning_strike_count,lightning_strike_avg_distance_km,lightning_strike_avg_distance_miles,battery_voltage,report_interval) VALUES (from_unixtime([% obs.0.0 %]),[% dbh.quote(serial_number) %],[% firmware_revision %],[% dbh.quote(hub_sn) %],[% obs.0.1 %],([% obs.0.1 %]*0.029530)+4.7634625,[% obs.0.2 %],([% obs.0.2 %]*1.8)+32,[% obs.0.3 %],[% obs.0.4 %],[% obs.0.5 %],[% obs.0.5 %]/0.621371,[% obs.0.6],[% obs.0.7 %])

*** sql/obs_sky.sql.disabled
INSERT INTO obs_sky (timestamp,serial_number,firmware_revision,hub_sn,illuminance_lux,uv_index,rain_previous_minute_mm,rain_previous_minute_in,wind_lull_mps,wind_lull_mph,wind_avg_mps,wind_avg_mph,wind_gust_mps,wind_gust_mph,wind_direction,battery_voltage,report_interval,solar_radiation,day_rain_accumulation_mm,precipitation_type,wind_sample_interval) VALUES (from_unixtime([% obs.0.0 %]),[% dbh.quote(serial_number) %],[% firmware_revision %],[% dbh.quote(hub_sn) %],[% obs.0.1 %],[% obs.0.2 %],[% obs.0.3 %],[% obs.0.3 %]/25.4,[% obs.0.4 %],[% obs.0.4 %]*2.23694,[% obs.0.5 %],[% obs.0.5 %]*2.23694,[% obs.0.6 %],[% obs.0.6 %]*2.23694,[% obs.0.7 %],[% obs.0.8 %],[% obs.0.9 %],[% obs.0.10 %],[% obs.0.11 %],[% obs.0.12 %],[% obs.0.13 %])

*** sql/obs_st.sql
INSERT INTO twenty_four_hour (epoch,station_id,wind_lull_metric,wind_lull_standard,wind_avg_metric,wind_avg_standard,wind_gust_metric,wind_gust_standard,wind_direction,wind_sample_interval,barometric_pressure_milibars,barometric_pressure_inHg,temperature_celcius,temperature_farenheit,relative_humidity,illuminance,ultraviolet_index,solar_radiation,rain_amount_metric,rain_amount_standard,precipitation_type,lightning_avg_distance_metric,lightning_avg_distance_standard,lightning_strike_count,battery_voltage,report_interval) VALUES (from_unixtime([% obs.0.0 %]),[% dbh.quote(serial_number) %],[% obs.0.1 %],[% obs.0.1 %]*2.23694,[% obs.0.2 %],[% obs.0.2 %]*2.23694,[% obs.0.3 %],[% obs.0.3 %]*2.23694,[% obs.0.4 %],[% obs.0.5 %],[% obs.0.6 %],([% obs.0.6 %]*0.029530)+4.7634625,[% obs.0.7 %],([% obs.0.7 %]*1.8)+32,[% obs.0.8 %],[% obs.0.9 %],[% obs.0.10 %],[% obs.0.11 %],[% obs.0.12 %],[% obs.0.12 %]/25.4,[% obs.0.13 %],[% obs.0.14 %],[% obs.0.14 %]/0.621371,[% obs.0.15 %],[% obs.0.16 %],[% obs.0.17 %])

*** sql/obs_st.sql.original
INSERT INTO obs_st (timestamp,serial_number,firmware_revision,hub_sn,wind_lull_mps,wind_lull_mph,wind_avg_mps,wind_avg_mph,wind_gust_mps,wind_gust_mph,wind_direction,wind_sample_interval,station_pressure_mb,station_pressure_inhg,temperature_c,temperature_f,relative_humidity,illuminance_lux,uv,solar_radiation,rain_previous_minute_mm,rain_previous_minute_in,precip_type,lightning_strike_avg_distance_km,lightning_avg_distance_miles,lightning_strike_count,battery,report_interval) VALUES (from_unixtime([% obs.0.0 %]),[% dbh.quote(serial_number) %],[% firmware_revision %],[% dbh.quote(hub_sn) %],[% obs.0.1 %],[% obs.0.1 %]*2.23694,[% obs.0.2 %],[% obs.0.2 %]*2.23694,[% obs.0.3 %],[% obs.0.3 %]*2.23694,[% obs.0.4 %],[% obs.0.5 %],[% obs.0.6 %],([% obs.0.6 %]*0.029530)+4.7634625,[% obs.0.7 %],([% obs.0.7 %]*1.8)+32,[% obs.0.8 %],[% obs.0.9 %],[% obs.0.10 %],[% obs.0.11 %],[% obs.0.12 %],[% obs.0.12 %]/25.4,[% obs.0.13 %],[% obs.0.14 %],[% obs.0.14 %]/0.621371,[% obs.0.15 %],[% obs.0.16 %],[% obs.0.17 %])

*** sql/rapid_wind.sql.disabled
INSERT INTO rapid_wind (epoch,station_id,speed_mps,speed_mph,direction) VALUES (from_unixtime([% ob.0 %]),[% dbh.quote(serial_number) %],[% ob.1 %],[% ob.1 %]*2.236936,[% ob.2 %])

And to ensure it starts on boot (this is outside of docker - meh, I was so frustrated that I'm fine with it), I created an /etc/systemd/system/weatherflow_udp.service file containing :

[Unit]
Description=UDP Listener for Tempest Weatherflow
After=default.target

[Service]
Type=oneshot
ExecStart=/where/ever/you/stored/it/tempest/perl/udp_listener.pl --daemonize --sql-directory /where/ever/you/stored/it/tempest/perl/sql/

[Install]
WantedBy=default.target

And then enabled the service :

systemctl daemon-reload
systemctl enable weatherflow_udp

Viola (or Cello, I don't care), I have data!

One step closer!

Docker and WeeWx

2026-07-22T13:22:40.995-07:00

WeeWX - Weather Station Data Gathering

Note, I eventually gave up on installing the weatherflow-udp extention under docker, and just wrote a UDP-to-CGI bridge tool, documented later. However, I opted to post this anyway because of what I ran into and it could be useful for someone out there (I got fairly close, actually).

This is an interesting tool to accept data for the weather system. I'm using a weatherflow tempest, which means I have a UDP port I have to translate. Plus, it is a little more complicated than the others we've dealt with. First, let's create a pre-run form of docker-connect.yaml :

services:
  weewx:
    image: felddy/weewx:latest
    container_name: tempest
    restart: unless-stopped
    ports:
      - "50222:50222/udp" # Map WeatherFlow UDP port
    environment:
      - TZ=America/Boise
    volumes:
      - type: bind
        source: ./weewx-data
        target: /data

Then, let's create our data directory to grab configuarion files from :

mkdir weewx-data/
chown 1000:1000 weewx-data

Now we can run a "fake" start of the docker image to get the configurations we need :

docker compose run --rm weewx

At this point, the weewx-data/weewx.conf file is ready for manipulation. Copy the weewx.conf into the docker image directory, then modify it by changing the station name, the latitude and longitude, etc. Once ready, change your docker-compose.yml file by adding the bind volume for the weewx.conf file, e.g. :

    volumes:
      - type: bind
        source: ./weewx-data
        target: /data

becomes :

    volumes:
      - type: bind
        source: ./weewx-data
        target: /data

Then, start the docker (usual "docker compose up -d").

It is now possible to run and install extensions. First, just to make sure it's showing, run :

docker exec -it tempest weectl extension list

Then install with :

docker compose run --rm weewx extension install --yes https://github.com/captain-coredump/weatherflow-udp/archive/refs/heads/master.zip

Now, it was right about here that I gave up on getting the extension installed. I'm leaving my notes, just in case. But really, all I need is a UDP translator, anyway. I'm not trying to do anything fancy like serve weather data (that will all happen through Grafana, so, yeah).\

On giving up, I just wrote a UDP listener.

Docker and Grafana

2026-07-22T13:19:46.261-07:00

Managers love pretty pictures, right? I know, this is a personal intranet, but hey, why not?

Grafana was a bit more of a challenge to run in docker. I have issues with it being a default non-SSL port, but we can potentially fix that by throwing Apache in front of it down the road. But, it's the same process as before, but with a major caveat.

We're not doing much special here, but I do need to make a quick statement. My Grafana was an older 9.something version. If you run it in Docker, and top are not starting from scratch (e.g. you are going to copy from a backup), it will fail. In this case, you MUST download a 12.3 package, and do a debian upgrade on that package, and stop/restart so that it can migrate. THEN you can move to the Docker 12.4 image base.

Now, back to the regularly scheduled programming.

Simply copy over your /var/lib/grafana/grafana.db SQLite file. If you have customizations in your grafana.ini file, you can make those changes to a grafana.ini file and map it just like we did with Apache HTTPd above. Note that it will be better to do that old "cat" command we mentioned in Apache to get a copy and then modify that one.

My docker-compose.yml looks like :

 services:
  web:
    image: grafana/grafana
    container_name: grafana
    restart: unless-stopped
    ports:
      - "3000:3000"
    environment:
      - GF_SECURITY_ADMIN_USER: admin
      - GF_SECURITY_ADMIN_PASSWORD: admin
      - GF_SERVER_ROOT_URL=%(protocol)s://%(domain)s:%(http_port)s/grafana/
      - GF_SERVER_ENFORCE_DOMAIN=false
      - GF_AUTH_PROXY_ENABLED=true
      - GF_AUTH_PROXY_HEADER_NAME=X-WEBAUTH-USER
      - GF_AUTH_PROXY_HEADER_PROPERTY=username
      - GF_AUTH_PROXY_AUTO_SIGN_UP=true
    volumes:
      - /opt/docker/storage/grafana:/var/lib/grafana:rw

Note the environment variables.

The SECURITY_ADMIN stuff sets your default username/password. If you are not migrating, and are instead creating a brand-new instance, these will set the default username/password.
The GF_SERVER_ROOT_URL is used to change the default URL (since Grafana is so smart it is going to redirect you to the Grafana port number, and this specifies whoat to hit with).
The GF_SERVER_ENFORCE_DOMAIN - this is where the AI engines kept biting me. They had NO CLUE about this. Grafana LOVES to redirect you, and if the enforce_domain in the grafana.ini uses the default or is set to true, it's doing to redirect you if you didn't hit it with the domain (could be hostname, and in docker, this defaults to "localhost"). So, if you are using docker, and if you are also reverse proxy'ing Grafana, which, in a reverse proxy environment, is not good). It kept redirecting me to http://localhost:3000/ . Setting this value via GF_SERVER_ENFORCE_DOMAIN allowed me to override the enforce_domain and shut that stupid thing off.
The GF_AUTH_PROXY_* environmment variables are to disable forcing a login. If you are reverse proxy'ing Grafana, and the parent process is handling the authentication, this is likely what you will need to do in combination with the ReverseProxy configs I showed earlier that use the rewrite_module under Apache HTTPd.

Then it's a matter of starting it up again (the "docker compose up -d" command).

Docker and MySQL

2026-07-22T13:17:21.915-07:00

In my pursuit of an easier, personal "intranet", my next task was to get MySQL running. It's critical for the Grafana and Nagios instances that run my environmental monitoring. There is a lot of historical (hysterical?) data involved, so we'll make it easier and just migrate it over, right?

First, create your docker directory for MySQL :

mkdir -p /opt/docker/mysql

Then, create only one file in there, the infamous docker-compose.yml :

services:
  mysql:
    image: mysql:latest
    container_name: mysql-container
    environment:
      MYSQL_ROOT_PASSWORD: super-secret-pwd
      # uncomment below if you want to automatically create and select a database
      # MYSQL_DATABASE: database_name
    ports:
      - "3306:3306"
    volumes:
      - /opt/docker/storage/mysql:/var/lib/mysql

Change the MYSQL_ROOT_PASSWORD row in the docker-compose.yml, and you're done.

Now, we are ready to spin that sucker up already - it really is that simple.

docker compose up -d

Give it a minute to spin it up. Then, test the connection (no MySQL Unix-socket will be available, so your MySQL command has to use the "--protocol TCP" option :

[jlewis@docker mysql]$ mysql -u root -p -h localhost --protocol TCP
Enter password: 
Welcome to the MySQL monitor.  Commands end with ; or \g.
Your MySQL connection id is 9
Server version: 9.7.1 MySQL Community Server - GPL

Copyright (c) 2000, 2026, Oracle and/or its affiliates.

Oracle is a registered trademark of Oracle Corporation and/or its
affiliates. Other names may be trademarks of their respective
owners.

Type 'help;' or '\h' for help. Type '\c' to clear the current input statement.

mysql> show databases;
+--------------------+
| Database           |
+--------------------+
| information_schema |
| mysql              |
| performance_schema |
| sys                |
+--------------------+
4 rows in set (0.01 sec)

mysql> exit

Create whatever you need in your database, and you have a modular database that can be quickly moved over to another instance of docker if CPU gets too high. In my case, I'm migrating an existing MySQL database, so I'll run :

mysqldump -u root -p databasename > mysqldump-databasename.sql

Copy that to the location that can write this into the database, create the database ("CREATE DATABASE databasename" in the new MySQL server first) and import it :

mysql -u root -p --protocol TCP databasename > mysqldump-databasename.sql

Doing pretty good so far.

Docker - Apache HTTPd for my tiny Intranet

2026-07-22T13:12:35.416-07:00

I'm rebuilding my network stack for my little intranet, and the first thing out of the gate is the HTTPd web service. So, let's get started.

I'm not using NginX because there will be a lot more CGI, and also because I want to implement my template mechanism, a custom module for the Apache HTTPd serer. Converting from an existing HTTPd configuration is a bit more difficult when it comes to migrating it in to Docker.

Now, RedHat and other Linux vendors/distributors love to part the httpd.conf file out into a whole lot of directories.

Docker doesn't do that. However, we have a little secret. So, let's create our docker image first. Create your docker directory and then the container for this specific image.

mkdir -p /opt/docker/httpd

Because the docker image is actually based on a Debian version, and because it includes an httpd.conf file already (by default), we only need three files in this directory.

For the httpd.conf, I found it easier to grab the httpd.conf file out of the docker image first, then add my custom module. And, since I wanted the image to serve content from the NAS, I needed to get the web content from that mounted up. Let's grab a copy of the docker container httpd.conf file.

docker run --rm httpd:2.4 cat /usr/local/apache2/conf/httpd.conf > httpd.conf

Now, with that, I can add my own module using the LoadModule with any custom directives. Then, we simply add the NFS mounts for the mount the NAS up via NFS, and then use volumes, too. Enter the docker compose.

services:
  web:
    image: httpd:2.4
    container_name: httpd
    restart: unless-stopped
    ports:
      - "80:80"
      - "443:443"
    volumes:
      - /opt/docker/storage/http/html:/usr/local/apache2/htdocs/:ro
      - /opt/docker/storage/http/cgi-bin:/usr/local/apache2/cgi-bin/:ro
      - /opt/docker/storage/http/logs:/usr/local/apache2/logs/:rw
      - ./certs:/etc/apache2/ssl
      - ./httpd.conf:/usr/local/apache2/conf/httpd.conf:ro
      - ./mod_template.so:/usr/local/apache2/modules/mod_template.so:ro

The container name/image, are fairly inconsequential - you can name them anything you want. The default is "apache_service" for this from the Docker image, but I like to get very specific because Apache likes to do a lot of services.

The "restart" line is there in case the EliteDesk hardware loses power and reboots - I want this container to start back up immediately.

The ports are very easy to understand if you are used to TCP. It is formatted as "HOST PORT:CONTAINER PORT".

Also, we need to set up SSL (note we have port 443 in there). Create a certificate :

mkdir certs
openssl req -x509 -nodes -days 365 -newkey rsa:2048 -keyout certs/httpd.key -out certs/httpd.crt -subj "/C=US/ST=Idaho/L=Boise/O=SilverHawk/CN=silverhawk.net"

You will need to set up the standard VirtualHost for it (make sure you uncomment the LoadModule line for the ssl extension while in there) :

<VirtualHost *:443>
        DocumentRoot "/usr/local/apache2/htdocs"
        #ServerName www.example.com
        SSLEngine On
        SSLCertificateFile /etc/apache2/ssl/httpd.crt
        SSLCertificateKeyFile /etc/apache2/ssl/httpd.key
</VirtualHost>

While I was there, I also added proxy and proxypass information for Grafana, (that's coming - note you need to enable xml2enc_module, proxy_http_module, and proxy_module, and make sure you disable proxy for everything except your own proxt so that someone doesn't suddenly use your site as a personal proxy to mask what they are doing on the Internet).

A grafana note :

I really strugged with proxy pass in this situation. all of the AI engines kept telling me things that were not accurate. So, here's what I have. In my httpd.conf for docker, I have the following excerpt :

ProxyRequests Off
<VirtualHost *:443>
        DocumentRoot "/usr/local/apache2/htdocs"
        #ServerName www.example.com
        SSLEngine On
        SSLCertificateFile /etc/apache2/ssl/httpd.crt
        SSLCertificateKeyFile /etc/apache2/ssl/httpd.key

        RewriteEngine On
        RewriteRule .* - [E=PROXY_USER:%{LA-U:REMOTE_USER},NS]
        RequestHeader set X-WEBAUTH-USER "%{REMOTE_USER}e"
        ProxyPreserveHost On
        ProxyPass /grafana/ http://172.17.0.1:3000/
        ProxyPassReverse /grafana/ http://172.17.0.1:3000/
        RequestHeader unset Accept-Encoding
        <Location /charts>
                TemplateEnabled off
        </Location>
</VirtualHost>

Then, in my Grafana docker-compose.yml file, I had to add the GF_SERVER_ROOT_URL and GF_SERVER_ENFORCE_DOMAIN environment variables (there are some others, to set up anonymous browsing since this will be front-ended by an authentication later) :

    environment:
      - GF_SERVER_ROOT_URL=%(protocol)s://%(domain)s:%(http_port)s/grafana/
      - GF_SERVER_ENFORCE_DOMAIN=false
      - GF_AUTH_PROXY_ENABLED=true
      - GF_AUTH_PROXY_HEADER_NAME=X-WEBAUTH-USER
      - GF_AUTH_PROXY_HEADER_PROPERTY=username
      - GF_AUTH_PROXY_AUTO_SIGN_UP=true

Volumes are where the magic happens for the custom module and configuration as well as the web root. Each row represents a specific file/directory that is mapped from the docker image. For example, I have my mod_template.so in there so it gets mapped into the container, plus I am mapping an httpd.conf file, as well as two NFS directories so I don't have to restart the docker every time I change the website.

Then, it's simply a matter of launching it up.

sudo docker compose up -d

Note, if you run into errors, you can use (and there is a -f option to tail -f them) :

docker compose logs

This will dump the containers log files (I suppose you COULD re-map log files if you want those pulled out of the image naturally, but this was easier).

Flushing The Radiator

2026-05-05T19:37:00.000-07:00

While preparing to wire in dual electric fans on a switch, I knew I needed to give the cooling system a clean bill of health. I mean, what's an evening compared to peace of mind? (Or is it supposed to be a piece of mind because I may have lost it?)

I slipped by the big box home improvement store, bought two 6-foot garden hoses, two 1-1/4" PVC reducers (to 1/2" female pipe thread), three short 1/2" pipes (2" in length, I think), one PVC ball valve, and two garden hose adapters (one male and one female, both with a female 1/2" pipe thread on the other end).

I screwed it all together, and had a solid system, ready to flush the radiator. I cracked the thermostat housing, and... realized a previous engine owner had mixed orange and green antifreeze (I'd filled with water at one point, but never added antifreeze). Orange and green antifreeze do NOT mix well!

Here's the thermostat.

I think it's pretty safe to say, don't mix orange and green antifreeze. Any way, it took some time to get all of that cleaned out, including removing the thermostat (so I can flush the engine side, of course - if that's still in there, it won't open, and you won't get any flush). Then, it was time to flush the system. I started with the radiator side, and put the ball valve host between the house garden hose connection and the top of the radiator (connected from the engine side), and the other hose at the other end, with the hose in a bucket.

Turn on the house connection, then turn on the ball value, and it pushed water through. Initially, it came out a thick-ish (it wasn't really thick - it was still very much liquid), but definitely not transparent.

I emptied one bucket, filled the next, and the next one came out a little cleaner.

I switched sides to flush from the other direction (still on the radiator), and ran two more buckets, until I was satisfied with clear water.

I disconnected my lines from the radiator hosts and re-attached them to the engine, then took off the radiator side in order to flush the engine. Again, I did this from both ends, flushing coolant from both directions.

Again, repeating until I had clear water.

Once I was satisfied that water was clear from both directions, I stopped.

I detached from from the hoses, put a new thermostat (95-degree) with switch and sensor threaded holes, and now I'm ready to run this engine once I get the fuel pump in.

Hammo Can Antenna

2026-05-02T14:27:00.000-07:00

I needed a better way to store my Hammo Can antenna. I ran to the local hardware store and found some 1-1/2" PVC schedule 40 pipe, because that's larger than the diameter of the antenna base. I cut a 48" length of the pipe.

Next, I glued up some fittings so that I could thread things on. I'm going all-external to keep the 1.5" interior dimension to fit the antenna.

After that, I needed to machine some pieces that I could set inside those end caps. I took some UHMW round, bored it through 0.800" (to clear a male PL-259 thread, which I had put a male-to-male connector in the bottom of the antenna). I then bored out only 1.25" deep to a 1.038" inside diameter (to grip the antenna's straight knurling). The outside was turned down to 1.480" (a slightly slip-fit for the inside of the pipe end).

With the part parted from the rest of the stock, and inserted into the pipe, I cross drilled with a 13/64" (a clearance drill size for a 5mm screw) both the outside pipe and the part itself.

I used the hole in the part to line it up with a drill bit, clamp a vise around it to hold it securely, then drilled the part itself for a threaded insert with the size specific for my inserts (letter "J"). I installed one heat-set insert, and bolted the part in place. Since this is my first of four holes, I added a witness mark via sharpie so that I knew which hole went where.

The part was chucked back up into the lathe to turn it round again just on that surface, and then re-installed to the pipe with one screw. It was then taken back to the drill press to pop the three remaining holes with the 13/64" drill bit. The same technique was used to then clamp on the 13/64", switch to the "J" drill bit, and add clearance for the inserts. Then, install the last three inserts, and turned back down to dimension on that insert band on the lathe again.

Next up, I heated the part up until I could force the antenna/pl-259 adapter into the part, and locked it down with a washer. Back into the pipe went the part/antenna.

This means I can unthread this from a pipe, flip it around, and thread it back into the pipe, and I have an outside antenna that can be staked to the ground via the pipe, or return it internally to the pipe to protect it in storage.

Now I have a functional storage for the antenna that also doubles as a riser!

You Light Up My Life

2026-04-20T09:01:00.000-07:00

Whoever designs Honda Accord's probably ought to find a new job.

One headlight went out. I decided I need to just replace them all at the same time (Rock Auto had all four for the price of one light at O'Reilly's Auto).

I managed to replace two lights, on the passengers side.

H11LL are the low-beams for a 2017 Honda Accord.
9005LL (HB3) are high beams.

The outside light on the passengers side was the easiest. It was simply disconnecting it, rotating it to release, put the new one in and lock it in place, and re-connect the wiring.

The inside light was an irritant. It took me a half hour to realize I could pop the plastic tank blocking it off (there is a tab on the back holding it in place, a long screw driver will release it) and lift it out of position without disconnecting any tubing. Then it, too, because fairly easy.

The drivers side was a whole different ball of wax. I did not want to remove the batter, so tried working around it. I broke the new bulbs trying to do that.

Simply put, just remove the battery. If the drivers side is one you have to swap, just remove the battery. It will make your life a WHOLE lot easier.

Dock(er) Wok

2026-02-12T14:57:00.000-08:00

Docker is a simple way to create virtualization without having full systems running. I took a moment to figure out how to run docker in relation to Postgres.

Note you need at least three servers for this for it to have any sort of redundancy. One is the "master". You can probably get away with two (run the manager on one of the nodes), but I do not recommend this. Really, you could get away with one server that runs the manager and two instances (I did), but know that if you don't run the docker instances on separate hosts, you've now lost all high availability, because if that server fails, the whole stack will cease to exist.

First, install Docker :

sudo apt-get install docker.io
sudo systemctl enable docker
sudo systemctl start docker

Then, create your swarm (anything recent should have swarm built in, you just need to set it up).


sudo docker swarm init --advertise-addr 192.168.x.x

Next, join a docker worker to the swarm

username@server1:~$ sudo docker swarm init --advertise-addr 192.168.0.3
Swarm initialized: current node (NODE IDENTIFIER) is now a manager.

To add a worker to this swarm, run the following command:

    docker swarm join --token TOKEN_STRING 192.168.0.3:2377

To add a manager to this swarm, run 'docker swarm join-token manager' and follow the instructions.
username@server1:~$

Then, on each worker node that isn't the manager, run the command you were provided (use sudo) :

sudo docker swarm join --token TOKEN_STRING 192.168.1.237:2377

Next is your basic configuration. Each host is relatively identical, with just a few modifications. Each host has 3 different configuration files, pg_hba.conf, postgresql.conf, and pg_ident.conf. Let's start by creating our directory structure.

I created a directory to house all of this so I can clear it out quickly after my learning curve.

mkdir cluster-postgres
cd cluster-postgres
mkdir -p {master,slave-1,slave-2}/config

Next, create our three files for each host :

touch {master,slave-1,slave-2}/config/{pg_hba.conf,postgresql.conf,pg_ident.conf}

With the files created, let's populate them. Here's the gist for each host, but at the end, I'll identify differences as needed.

postgresql.conf

This file contains :

# -----------------------------
# PostgreSQL configuration file
# -----------------------------
#

data_directory = '/data'
hba_file = '/config/pg_hba.conf'
ident_file = '/config/pg_ident.conf'

port = 5432
listen_addresses = '*'
max_connections = 100
shared_buffers = 128MB
dynamic_shared_memory_type = posix
max_wal_size = 1GB
min_wal_size = 80MB
log_timezone = 'Etc/UTC'
datestyle = 'iso, mdy'
timezone = 'Etc/UTC'

#locale settings
lc_messages = 'en_US.utf8'   # locale for system error message
lc_monetary = 'en_US.utf8'   # locale for monetary formatting
lc_numeric = 'en_US.utf8'    # locale for number formatting
lc_time = 'en_US.utf8'       # locale for time formatting

default_text_search_config = 'pg_catalog.english'

#replication
wal_level = replica
wal_keep_size = 512MB  # Adjust this value as needed
archive_mode = on
archive_command = 'test ! -f /mnt/server/archive/%f && cp %p /mnt/server/archive/%f'
max_wal_senders = 3

However, on the slave nodes, lines 28-34, which contain :

#replication
wal_level = replica
wal_keep_size = 512MB  # Adjust this value as needed
archive_mode = on
archive_command = 'test ! -f /mnt/server/archive/%f && cp %p /mnt/server/archive/%f'

are removed.

pg_hba.conf

This file contains :

# TYPE  DATABASE        USER            ADDRESS                 METHOD

host     replication     replicationUser         0.0.0.0/0        md5

# "local" is for Unix domain socket connections only
local   all             all                                     trust
# IPv4 local connections:
host    all             all             127.0.0.1/32            trust
# IPv6 local connections:
host    all             all             ::1/128                 trust
# Allow replication connections from localhost, by a user with the
# replication privilege.
local   replication     all                                     trust
host    replication     all             127.0.0.1/32            trust
host    replication     all             ::1/128                 trust

host all all all scram-sha-256

However, on the slave nodes, line 3 :

host     replication     replicationUser         0.0.0.0/0        md5

is removed.

pg_ident.conf

This file contains :

# PostgreSQL User Name Maps
# =========================
#
# Refer to the PostgreSQL documentation, chapter "Client
# Authentication" for a complete description.  A short synopsis
# follows.
#
# This file controls PostgreSQL user name mapping.  It maps external
# user names to their corresponding PostgreSQL user names.  Records
# are of the form:
#
# MAPNAME  SYSTEM-USERNAME  PG-USERNAME
#
# (The uppercase quantities must be replaced by actual values.)
#
# MAPNAME is the (otherwise freely chosen) map name that was used in
# pg_hba.conf.  SYSTEM-USERNAME is the detected user name of the
# client.  PG-USERNAME is the requested PostgreSQL user name.  The
# existence of a record specifies that SYSTEM-USERNAME may connect as
# PG-USERNAME.
#
# If SYSTEM-USERNAME starts with a slash (/), it will be treated as a
# regular expression.  Optionally this can contain a capture (a
# parenthesized subexpression).  The substring matching the capture
# will be substituted for \1 (backslash-one) if present in
# PG-USERNAME.
#
# Multiple maps may be specified in this file and used by pg_hba.conf.
#
# No map names are defined in the default configuration.  If all
# system user names and PostgreSQL user names are the same, you don't
# need anything in this file.
#
# This file is read on server startup and when the postmaster receives
# a SIGHUP signal.  If you edit the file on a running system, you have
# to SIGHUP the postmaster for the changes to take effect.  You can
# use "pg_ctl reload" to do that.

# Put your actual configuration here
# ----------------------------------

# MAPNAME       SYSTEM-USERNAME         PG-USERNAME

There are no differences here for each node.

Last Configurations

Create the docker "network" :

sudo docker network create postgres-cluster-network

This will print a fairly large alphanumeric ID.

Starting the "master"

At this point in time, start the master node using the following command :

sudo docker run -d --name postgres-master  --net postgres-cluster-network \
-e POSTGRES_USER=postgresadmin -e POSTGRES_PASSWORD=admin123 \
-e POSTGRES_DB=postgresdb -e PGDATA="/data" -v ${PWD}/master/pgdata:/data \
-v ${PWD}/master/config:/config -v ${PWD}/master/archive:/mnt/server/archive \
-p 5000:5432 postgres:latest -c 'config_file=/config/postgresql.conf'

If you haven't downloaded the postgres docker image yet, this will actually cause it to try and install. If you get an error about it already existing because you tried to start it one and it gave you an error :

docker: Error response from daemon: Conflict. The container name "/postgres-master" is already in use by container "6362ca473d3f29b7fe1bf02558f5a871fb6cf6827eff23fe01714f34cb386951". You have to remove (or rename) that container to be able to reuse that name.

Then list the dockers, and delete it :

username@server1:~/postgres-cluster$ sudo docker ps -a
CONTAINER ID   IMAGE             COMMAND                  CREATED          STATUS    PORTS     NAMES
6362ca473d3f   postgres:latest   "docker-entrypoint.s…"   10 minutes ago   Created             postgres-master
username@server1:~/postgres-cluster$ sudo docker rm 6362ca473d3f
6362ca473d3f
username@server1:~/postgres-cluster$ sudo docker ps -a
CONTAINER ID   IMAGE     COMMAND   CREATED   STATUS    PORTS     NAMES
username@server1:~/postgres-cluster$

Now try to start it if it had previous failed again, and it should simply give you a large alphanumeric ID again.

Next, create a replication user :

username@server1:~$ sudo docker exec -it postgres-master bash
[sudo] password for username:          
root@63919c16e356:/# createuser -U postgresadmin -P -c 5 --replication replicationUser
Enter password for new role: 
Enter it again: 
root@63919c16e356:/# exit
exit
username@server1:~$

Just type exit and get back to the prompt.

Now, we can move on to starting the slaves.

Starting the slaves

For each node, run :

username@server1:~$ sudo docker run -it --name postgres-slave1 --rm \
> --net postgres-cluster-network \
> -v ${PWD}/slave-1/pgdata:/data \
> --entrypoint /bin/bash postgres:latest
root@447571e3e11e:/#

This will put you in a bash prompt as "interactive", where you then run :

pg_basebackup -h postgres-master -p 5432 -U replicationUser -D /data/ -Fp -Xs -R

The command initiates replication, and will use the password you specified on the master node when you started that up with the "createuser" command.

Repeat for slave-2 (of course, replacing postgres-slave1 with postgres-slave2, and the slave-1 folder names with slave-2).

Create standby instances

Run the following :

sudo docker run -d --name postgres-slave1 --net postgres-cluster-network \
-e POSTGRES_USER=postgresadmin -e POSTGRES_PASSWORD=admin123 \
-e POSTGRES_DB=postgresdb -e PGDATA="/data"  \
-v ${PWD}/slave-2/pgdata:/data -v ${PWD}/slave-2/config:/config \
-v ${PWD}/slave-2/archive:/mnt/server/archive -p 5002:5432 \
postgres:latest -c 'config_file=/config/postgresql.conf'

Repeat for slave-2 (of course, replacing postgres-slave1 with postgres-slave2, and the slave-1 folder names with slave-2).

Test It

Connect to the master node using :

sudo docker exec -it postgres-master bash

From in here, you can run your psql commands to create databases and manipulate whatever you need.

psql --username=postgresadmin postgresdb

Exit, and then check the other slave nodes by connecting to them (sudo docker exec) and running psql to query any tables you've created and populated with data.

Just Pulley-ing My Leg

2026-01-19T09:21:00.000-08:00

A year or so back, I had heavily modified am early 1940's tiny lathe with the intent of using it to turn pens. There are a number of small lathes that are likely cheaper than the cost put into that one, but for some odd reason, I love the old iron, and did it anyway.

The results were, less than stellar. Indeed, it did work, but I had to take my time because the stepper motor did not have enough torque. And that is with a small pen.

So, enter the next chapter. Someone over on the hobby machinist forum sent an old sewing machine motor to me (think Pfaff 130, and you have the footprint of it), just for the cost of shipping. With the pulley that was on it, I found myself in a bit of a quandary - I needed more modifications to see if this would work. So, I set about creating a new belt tensioning system. I patterned this variation off of a cars alternator - where it is installed and then tightened over an arc-bracket.

I made the bar using my drill press, because, I didn't have a rotary table large enough for the arc (mine is 4", the arc is 8" - but that is due to only having a scrap bar with an 8" arc that was an off cut for the 127-tooth compound gear), and also because I was too lazy to drag out the milling machine onto a cold back porch and cut it. Really, it was because I was way too lazy.

But, it was close enough for what I needed.

So, I drilled a series of holes, then used an endill in the drill press to smooth out the arcs. I drilled a few extra holes in some bar stock, and used a 1/4"-20 bolt to put it all together.

Next, I needed a pulley. So I grabbed another off-cut/drop (2" diameter aluminum), chucked it up, and made a 2L pulley that was as large as this lathe could accept, but small enough to fit (1.950" with the belt). This was faced, drilled, then reamed. It was then turned on the boss, then the flat part of the pulley, then the angles.

The result is a usable 2L pulley to match the one on the sewing machine motor.

I wired the motor to test it, and once that checked out okay, it was time to put it all together.

This immediately feels like more torque than the stepper. But, the proof is always in the pudding. What is the end result? Let's give it a try. I ran through the normal steps preparing a pen blank (this time, I chose a gun-metal "mini" bolt action pen kit from Penn State Industries - I have a lot of family that love the regular ones, so I thought I'd give this more "unique" variant a whack.)

Usually, mark the blank lengthwise to keep grain lined up. In this case, it's small, and I was fixing another pen while I was at it, so this step didn't matter.
Cut off two chunks to fit your pen tubes, about 1/4" longer than the tubes themselves. Again, this was small enough, and I was using a scrap blank I had from another project.
Drill the centering hole for the blank(s).
Glue in the tubes. Many people just use CA glue. I have always loved 5-minute epoxy for this.
Barrel-trim the ends to get it flush with the brass tubes (don't take off brass or it might not fit).
Put the tubes on the appropriate bushings, then on the mandrel, and turn.
Once to slightly larger than size, use the lathe to sand the blanks. Get as high a grit as possible for the best finishes.
Remove from the lathe and assemble according to the instructions.

So, once I got to the lathe part, I grabbed a photo.

They do look fantastic (I had two on there - I was repairing another pen while I was at it). Then, assembly prep :

Then, the final assembly.

While the motor was slightly "under-powered" (It IS a 1/5 HP sewing machine motor), it still did the trick. I do have to take things a little slower with this, but it does mean I don't have to drag out the larger twin for this little thing. I can simply pick it up, clamp it to a work bench, plug it in, and turn a pen.

Finding God in Failure

2026-01-11T12:04:00.000-08:00

Many religions believe we are being punished for Adam's transgression. We do not.

"And now, behold, if Adam had not transgressed he would not have fallen, but he would have remained in the garden of Eden. And all things which were created must have remained in the same state in which they were after they were created; and they must have remained forever, and had no end.

"And they would have had no children; wherefore they would have remained in a state of innocence, having no joy, for they knew no misery; doing no good, for they knew no sin.

"But behold, all things have been done in the wisdom of him who knoweth all things.

"Adam fell that men might be; and men are, that they might have joy.

"And the Messiah cometh in the fulness of time, that he may redeem the children of men from the fall. And because that they are redeemed from the fall they have become free forever, knowing good from evil; to act for themselves and not to be acted upon, save it be by the punishment of the law at the great and last day, according to the commandments which God hath given." - 2 Nephi 2:22-26

Man is fallen. The earth is Telestial - imperfect.

It is this imperfect world that allows us to be imperfect.

Were it not for the imperfection of the world in which we live, we could not know how to hope for a better one. Two verses before the famous "faith" scripture in Ether, we have this little gem.

"Wherefore, whoso believeth in God might with surety hope for a better world, yea, even a place at the right hand of God, which hope cometh of faith, maketh an anchor to the souls of men, which would make them sure and steadfast, always abounding in good works, being led to glorify God." - Ether 12:4

All of us find themselves in times of "hoping for a better world".

Sermon on the mount - after stating that we should pray for those who curse us, states :

"for he maketh his sun to rise on the evil and on the good, and sendeth rain on the just and on the unjust." - Matthew 5:45

Do I mention cancer? Heartbreak? Heart Failure?

By Elder Gong tells the following story.

A Chinese story begins as a man’s son finds a beautiful horse.

“How fortunate,” the neighbors say.

“We’ll see,” says the man.

Then the son falls off the horse and is permanently injured.

“How unfortunate,” the neighbors say.

“We’ll see,” says the man.

A conscripting army comes but doesn’t take the injured son.

“How fortunate,” the neighbors say.

“We’ll see,” says the man.

This fickle world often feels tempest tossed, uncertain, sometimes fortunate, and—too often—unfortunate. Yet, in this world of tribulation, “we know that all things work together for good to them that love God.” Indeed, as we walk uprightly and remember our covenants, “all things shall work together for your good.” - Elder Gerrit W. Gong, Apr 2024, "All Things for Our Good"

Do we want to be "tempest tossed", or to wonder "why me?" How do we avoid this?

"We become our truest, best selves only as we put off the natural man or woman and become a child before God." - Elder Gerrit W. Gong, Apr 2024, "All Things for Our Good"

How do we put off the natural man?

For the natural man is an enemy to God, and has been from the fall of Adam, and will be, forever and ever, unless he yields to the enticings of the Holy Spirit, and putteth off the natural man and becometh a saint through the atonement of Christ the Lord, and becometh as a child, submissive, meek, humble, patient, full of love, willing to submit to all things which the Lord seeth fit to inflict upon him, even as a child doth submit to his father. - Mosiah 3:19

Yield to the enticings of the Holy Spirit.

Create an environment where the world can be quiet. Temple, at home in your personal room, or sitting in your car while waiting.
Commit to follow the direction you are about to receive.
Pray for the enticing of the Spirit of God to show how you can serve someone.
Remain in the quiet environment until you feel that you have something to do.
Obey immediately.
Repeat.

As you put this effort in to interact with the Spirit of God and to follow through, you are becoming a saint through the atonement of Christ the Lord. You become meek, humble, patient, full of love, and willing to submit to all things the Lord seeth fit to inflict.

You are building a trusting relationship with Christ.

"When we trust God and His love for us, even our greatest heartbreaks can, in the end, work together for our good." - Elder Gerrit W. Gong, Apr 2024, "All Things for Our Good"

This trusting relationship does not obliterate our enemies, or push obstacles out of our way.

Shadrach, Meshach, and Abed-nego, would not cede their covenants to a King, and were to be thrown into a furnace of fire. When the King questioned the power of their God, their response was :

"If it be so, our God whom we serve is able to deliver us from the burning fiery furnace, and he will deliver us out of thine hand, O king.

"But if not, be it known unto thee, O king, that we will not serve thy gods, nor worship the golden image which thou hast set up." - Daniel 3:17-18

The second half of that, "But if not", is the crucial component. Whether something happens or not does not cause a break in what our commitment should be.

Remember, we are mortal, living in a mortal world. But when we get lost in that "forest for the trees", pause, to also remember this.

"Even if you cannot always see that silver lining on your clouds, God can, for He is the very source of the light you seek. He does love you, and He knows your fears. He hears your prayers. He is your Heavenly Father, and surely He matches with His own the tears His children shed." - Jeff Holland, Oct 1999, "An High Priest of Good Things to Come"

God knows you, deeply and fully. If he is "endless", pause and realize that his love for you is also endless.

"He is the light and the life of the world; yea, a light that is endless, that can never be darkened; yea, and also a life which is endless, that there can be no more death." - Mosiah 16:9

We must recognize that our timing is never God's timing.

"Some misunderstand the promises of God to mean that obedience to Him yields specific outcomes on a fixed schedule. They might think, “If I diligently serve a full-time mission, God will bless me with a happy marriage and children” or “If I refrain from doing schoolwork on the Sabbath, God will bless me with good grades” or “If I pay tithing, God will bless me with that job I’ve been wanting.” If life doesn’t fall out precisely this way or according to an expected timetable, they may feel betrayed by God. But things are not so mechanical in the divine economy. We ought not to think of God’s plan as a cosmic vending machine where we (1) select a desired blessing, (2) insert the required sum of good works, and (3) the order is promptly delivered.

"God will indeed honor His covenants and promises to each of us. We need not worry about that. The atoning power of Jesus Christ—who descended below all things and then ascended on high and who possesses all power in heaven and in earth—ensures that God can and will fulfill His promises. It is essential that we honor and obey His laws, but not every blessing predicated on obedience to law is shaped, designed, and timed according to our expectations. We do our best but must leave to Him the management of blessings, both temporal and spiritual." - Elder D. Todd Christofferson, April 2022, "Our Relationship with God"

And once you recognize this, you will also recognize that, while we are imperfect and mortal, there is a full regalia of blessings waiting.

"Some blessings come soon, some come late, and some don’t come until heaven; but for those who embrace the gospel of Jesus Christ, they come." - Jeffrey Holland - Oct 1999, "An High Priest of Good Things to Come"

Recommend Jan 2009, Thomas Monson's speech to graduating students called "Great Expectations".

Rudyard Kipling, "If" :

If you can keep your head when all about you
Are losing theirs and blaming it on you,
If you can trust yourself when all men doubt you,
But make allowance for their doubting too;
If you can wait and not be tired by waiting,
Or being lied about, don’t deal in lies,
Or being hated, don’t give way to hating,
And yet don’t look too good, nor talk too wise:

If you can dream—and not make dreams your master;
If you can think—and not make thoughts your aim;
If you can meet with Triumph and Disaster
And treat those two impostors just the same;
If you can bear to hear the truth you’ve spoken
Twisted by knaves to make a trap for fools,
Or watch the things you gave your life to, broken,
And stoop and build ’em up with worn-out tools:

If you can make one heap of all your winnings
And risk it on one turn of pitch-and-toss,
And lose, and start again at your beginnings
And never breathe a word about your loss;
If you can force your heart and nerve and sinew
To serve your turn long after they are gone,
And so hold on when there is nothing in you
Except the Will which says to them: ‘Hold on!’

If you can talk with crowds and keep your virtue,
Or walk with Kings—nor lose the common touch,
If neither foes nor loving friends can hurt you,
If all men count with you, but none too much;
If you can fill the unforgiving minute
With sixty seconds’ worth of distance run,
Yours is the Earth and everything that’s in it,
And—which is more—you’ll be a Man, my son!

Elder Holland spoke about one of his own "travails" - he and his young family had car troubles. When he drove past that spot later in life, how he wished he could tell himself :

"Don't you quit. You keep walking, you keep trying, there is help and happiness ahead. Some blessings come soon. Some come late. Some don't come until heaven. But for those who embrace the gospel of Jesus Christ, they come. It will be alright in the end. Trust God and believe in Good Things to Come." - Jeff Holland, Oct 1999, "An High Priest of Good Things to Come"

And in the same talk, he continued :

To any who may be struggling to see that light and find that hope, I say: Hold on. Keep trying. God loves you. Things will improve. Christ comes to you in His “more excellent ministry” with a future of “better promises.” He is your “high priest of good things to come.” - Jeffrey Holland - Oct 1999, "An High Priest of Good Things to Come"

The Spindle Depth Stop

2025-12-18T14:51:00.000-08:00

I had a small run of parts on the lathe that needed to be the same length. There are a couple of ways to do this, such as :

After creating the part (and making sure both ends are faced, but a wee bit long), measure the length of the part.
Calculate the amount you need to remove by subtracting the final length of the part from your previous measurement.
Re-insert the part into the chuck (or collet), and measure the length of the stick-out from a known reference point.
Subtract the distance you need to remove from this stick out measurement.
Face the part until you reach that distance calculated in step 4.

However, I'm lazy, and that's a little too much effort. I'd rather do that only once, and on a single-run part, I'd have to do that, anyway. The micrometer stop does a bang up job of getting it right on once you've gone through that. But if I have four parts? Uh-huh. I'm too lazy, since I could insert the parts into a depth, while leaving the carriage/cross-slide in the one position.

It's time to make a spindle depth stop - something that locks into the spindle so you have a consistent reference point on all the parts.

Mind you, my spindle has either a collet closer attachment installed, or it has the big fat gear used by the collet, so I can't just throw a standard one in place. But, just in case, my South Bend Heavy 10L (large spindle bore) regular, non-gear depth-stop was also designed. This spindle depth stop can fit all three, just by making the locking spider for all three. That's only two extra parts, allowing this to fit three different configurations. I also wanted to use as many parts that I didn't have to make. Because, again, I'm lazy.

Okay, okay, enough blab. Let's get to it. Grab the materials (I'm going to use CR1018/cold rolled steel).

First, let's get the stainless-steel 3/8"-16 all thread cut to length. I used a 36" bar, and cut it at the 24" mark. No precision necessary here, I just have two variations in case I don't like too much sticking out the back for an extra long part.

Next, I already had the nuts in hand (that's not what I meant, and you know it) for the 3/8"-16 all thread, and a 5/8"-11 nut for the locking lug.

First part to "really" make was the locking lug. Really, it was a simple turning job for the most part. Turning to 0.625" for the threaded end, threading to 11 TPI to match the nut, and then getting the main body set, the internals bored for clearance for the threaded rod (and one end threaded to 3/8"-16 of that internal bore. The hardest part was the "cone" on the end. What I had drawn up in the plans had a 17.615 degree angle, and to be honest, you just need it "close". Using the compound, you can set the angle, then lock that in. Do not change that angle until you have completed your locking spider(s)!

This was my first taper turning, and I went with a standard 17 degree. I don't care how close I am to 17 - it's not moving until I've made the entire thing.

Next up was the washer. It was coming from the same large stock as the spiders, so it's easy to slap in there and turn, face, and bore.

You just might notice that the proportions aren't lining up with the drawing. I actually opted to make this 1/2" thick instead of 1/4" thick, and I bored out 1/4" of the internal bore to match the larger diameter on the locking lug I've finished. This will give me a little more expansion room, and allow me to use less material on the spiders.

Now I just need at least one spider. I'll do the gear-based spider first :

Then, I'll do the collet attachment spider :

Now, I'm not making this next spider, because I have that fat gear that sticks out for the spindle attachment. It means I can't really use it. However, I drew the plans up anyway, just in case someone needed them.

So, I started out creating my two spider blanks. I chucked up the material, faced one end and the diameter in order to get the base set up. I also rough-turned one to get it not quite to dimension (I'm 0.050" oversized on the outside). I did not go all the way to dimension yet because I needed the bore set up so that as stresses are relieved, the outside will then be truly concentric to the bore.

Once I had a close-but-no-cigar dimension on the outside, I turned (pun is not intended) to the inside for the boring process. First, I used drill bits to get it to 0.5" (needed to get to 0.810"), and then I could use the boring bar.

Once the inside bore was complete, it was time to also turn the taper. This is the exact reason why the compound was not moved - it already has the matching taper from the internal part, so I knew it would be a match made in heaven (or at least on the lathe).

With the spider's internal bore/taper and the locking lug fitting as they should, I went back and turned the outside down to the final dimension, and then turned the 0.125" circular relief groove using a rounded insert (could only find metric, and it was close enough). Yes, in the following photo, the lathe is OFF. I refuse to lose a hand. But, I did it to provide a visual on the groove.

After the groove, I removed the chuck from the lathe and ensured I had a snug slip fit into the targeted location. At that point, I could pull it out of the chuck and do a test fit up. It definitely locks in there, but without the expansion slots, it won't truly lock.

I do know that I need to remove more on the outer end (largest diameter face), because I didn't get it short enough to fit on the locking lug. So, into the chuck it will go with some brass shim stock to protect the finish, and then I can face it down to size until I have a perfect arrangement.

It's not often you hit the target width so closely. I thought it was close enough, and assembled everything (not usable until the slots are cut).

Now, the remaining task is to cut the slots into the surface. There are six slots, but using a band saw (my slitting saws are not large enough), it's only three cuts. So, next I need to build a jig to keep it perpendicular to the saw blade, mark the three cuts, and then start slicing. This had my anxiety running high, because there is no way I'm willing to lose a finger. And, with two separate spiders for different dimensions, I need something adjustable for that smaller diameter. I need it to be at least close to parallel to the axis. So, after thinking on it for a month, one morning, I needed to get out of the house, and I ran out and grabbed some 2"x1/2" aluminum flat bar, and simply marked out a few parts. Cut, drilled, tapped for M3 screws (because I didn't want to wake my wife early in the morning), and I had this little thing.

I can thread those two screws in or out, depending on the diameter of the spider end, so there is the adjustability that I needed for the two spiders. It allows me to adjust those two screws until I have a parallel setup, and I can then feed that into the bandsaw blade without fear of something catastrophic. It does fit like I hope.

The next task was the actual cutting of the slots. The jig held up quite nicely, though I should have adjusted the screws part way through. While there, I also drilled relief holes where those slots ended so that there wasn't a hard corner or edge to start cracking on.

I used a triangular (bastard) needle file to deburr the slots (outside and inside), and a round needle file to remove the burr on the inside where the drill bit broke through. But, once that was done, I had a relatively decent part. A quick test fit of everything together....

Now, it's time to try it out.

It locked in there without much of a tightening on that large nut (it could still slip, so I'd likely use a wrench when I really need it). Functioned pretty well!

The second one can now be chucked up and replicated to the dimensions of the first (with the smaller diameter for the collet attachment tube).

Finally Got My Pulse

2025-11-26T20:12:00.000-08:00

Well, we've been dealing with side effects of the cancer for a while. The symptoms were dizzy spells, and not just when changing positions (like standing up). These came while sitting down and staying there.

Officially, we finally got it half-way pinned down to a heart failure diagnosis with a "reduced ejection fraction", meaning the heart wasn't pumping enough blood. When we had tests ran, the numbers came in at 41% of what it should be.

[sigh].

We next did an ablation on the heart to knock off some arrhythmia that had crept into the picture while the chemo was happening. Things were not perfect, but much better. After watching things a bit, we found that the procedure had a definite impact :

It went from about 10-20 spells per day down to about 2-5, with some days not having any at all.

Gotta get the ol' ticker running properly, and this was a good step!

I Drove a Porch

2025-10-24T14:42:00.000-07:00

No, it's not misspelled.

As you age, so do things around you. I knew of a couple who was getting long in the years, and their back porch was becoming rickety. The "porch" was simply a set of welded metal steps. And if I stood on them without moving, they'd rock back and forth. Seriously, it was the only way for me to look like I had any dance moves. So, their daughter volunteered her and her siblings to build them something new. Somehow, I ended up becoming a de facto decision maker.

I started out by making "dad" mark out where it was going to go. While they headed off to take care of some legal things, one of the siblings hubby and I (and their kids) ran over to get started.

We fired up an auger to drill footings - but by the time we got it running, the young lad had already dug three of the holes. Ah, youth and vigor.

We dropped some "Sonotubes" in place, shoved some rocks in the bottom to stabilize it a bit more, and filled them with concrete followed by some 4"x4" post base brackets, and let that set up overnight.

The next morning, we started installing vertical posts. I made the young lady of the group do the bolt hole drilling - she did fantastic. Truth is she did a lot more than that. She worked on the actual decking, the stairs, etc. She ran the drills, the impact drivers, etc.

Part way through, another siblings kids showed up. They just cranked through the project, and did fantastic. Apparently, I was simply there to be beautiful.

But, I was the idiot who climbed onto the roof panels before they were fully anchored in place. Those sheets had to be screwed down from above, so someone needed to, and I've had a good life. I figured I had less life to lose, eh?

Anyway, we put up the railing, and built a "roof line" that was close to the house roof line. Unfortunately, it needed to be "separate" so that we didn't require a permit. We kept it close, but still separate. In one weekend, we got the foundation, the frame, the deck, and the base of the roof built out.

We still need to install the moisture barrier and the roofing itself, but the basic shape is there, and it is much more stable.

Rebuilding my Roller Saw Horse Stand

2025-09-26T13:34:00.000-07:00

My workshop is fairly limited in size. While that can be changing, in the immediate time, I don't have a lot of space. It means I have to be able to set up in a mobile fashion, e.g. I take over the back patio when I need to.

I've used saw horses to set up a large work bench before. And about 6 years ago, I did what I shouldn't do, and I used roller stands (the kind you'd use to allow long boards to roll off of a table saw, jointer/joiner, or a planer). Now, they "roll", which they are supposed to, so the idea 6 years ago was tested to keep them from rolling when needed. I used some 2" PVC pipe, which (when cut lengthwise to allow the arms supporting the rollers to slide into the "pipe") seemed to fit absolutely perfectly on the rollers.

The problem is that over time, the wood split, warped, and encountered some heavy water damage. They became unusable (not quickly, but just over time). I needed something that would last a little longer. So, I made a few purchases, and set to work on "revision 2".

Here's what you'll need :

Two HDPE sheets measuring 3/4" thick x 12" wide x 48" long
Four aluminum angle iron 1-1/2" x 1-1/2", with legs that are 1/4" thick (thick enough to act as a "nut")
48" of 2" PVC pipe, cut into 12" lengths
Eight 2" wide x 1-3/4" thick x 3" long HDPE bars
Sixteen heat-set threaded inserts for assembly screws of your choice
Two thread inserts for "locking pins"
At least two transfer punch screws (4 is preferable)
One countersink screw bit
One drill bit for tapping screws of your choice
One drill bit for clearance of the same screws
One drill bit for the thread "locking pin" inserts
One drill bit for the heat-set threaded inserts
Eight spacers (I used drops from other projects) that need to be the same length - about 1" in length, maybe 1" in diameter
One (or more if you break them) tap
A soldering iron with a heat-set insert tip
A screwdriver

How to :

Identify (and mark) which side of the angle iron is to be the "surface" side.
Drill the tapping-size holes for the screws of your choice into the aluminum angle iron. I used 5 on each surface-side, with spacing at uneven (check that, my CDO/OCD had issues, but I wanted to decrease linear distance the surfaces would be screwed down with). Two pieces of angle-iron are used per "saw horse", and the downward legs will face each other. Orientation is key here.
Tap the holes.
Drill four holes along the downward legs of ONE of those angle iron bars per horse (if building two horses, you'll do two).
Tap these holes.
Use the transfer screws in the downward legs and bring the two pieces of angle together to mark the other set.
Drill the clearance size holes in the second piece of angle aluminum.
Put your spacers in between, and screw the two pieces of angle together into a structural arm.
Center the arm on one piece of HDPE sheet, and mark ONE hole on the HDPE sheet.
Drill that hole for the clearance size.
Bolt the HDPE sheet and the structural arm together with that one bolt.
Mark the opposite end of the structural arm to the HDPE sheet.
Drill with the clearance size, and connect the two together. The two parts should not move at this point from each other.
Use the remaining holes and either use transfer screws, or use transfer punches.
Drill the remaining HDPE holes for structural arm attachment.
Countersink the holes from the top-side of the saw horse table top.
Attach the two together (note, they might come off the table later in this process, see step 27).
Use a hole-saw in a jig (I just built a square "box" I could clamp them to on the drill press, then clamped that box to the drill press table) to cut an arc on one side lengthwise. This should be about 2-3/8" in diameter, but I just used a 2" hole saw and made two different runs so that they matched.
Drill two holes (threaded insert sized)that will be perpendicular to the arc all the way through the bars. These will allow threaded inserts to be set in from each end, and allow the pipe to be screwed to these blocks.
Use threaded inserts on the back side of the HDPE blocks as well (allowing the blocks to be attached to the HDPE table top surface).
Heat-set inserts into the blocks.
Once cooled, use transfer screws on the non-arc side of the blocks to mark them onto the HDPE table top.
Drill (clearance sized) holes for each block.
Countersink the remaining holes.
Attach the blocks (permanently - we shouldn't have to remove them again) to the table tops.
Use the PVC pipe as a guide and mark the arc pattern into the angle aluminum structural bar. If the pipe clears, you don't need to do anything.
If the pipes do not clear the structural arms, remove one piece of aluminum angle at a time, and grind or cut the clearance so that the pipe will not touch the aluminum. I gave between 1/16" to 1/8" of clearance.
Once clearanced, re-install the piece of angle aluminum and tighten the screws.
Repeat for the other one as necessary.
Use a table saw to cut the pipe lengthwise. A word of caution - PVC pipe is under tension, and when it is fully cut through, will snap together and pinch your saw blade. This is extremely dangerous. There are other options, but if you go this direction, find a good way to hold the pipe securely without having your fingers close, go slowly, and if it pinches the saw blade and brings it to a halt, do NOT panic. Just turn off the saw, and work things free.
Use the table saw to cut the pipe to the width to clear the roller arms.
Use the transfer screws to mark the pipe in conjunction with the HDPE bars.
Drill the holes in the pipe and countersink so that you can attach the pipe to the HDPE bars.
Drill the CENTER of the table top to take the threaded inserts.
Countersink the remaining clearance holes (PVC pipe, table top, etc).
Screw everything together.
Heat the inserts up and screw them into the table. These allow you to put in a pin and prevent the workbench from sliding on the HDPE.

The screws themselves won't be holding the weight; they just keep things from moving around and shifting. It means that you don't need shear strength in picking your hardware.

The end result can just clip together when you need a "saw horse" mode from your roller stands. When you need roller stands, just take them back apart.

They are quite nifty for the "mobile" or "temporary" workshop.

Hammo Can - Radiated Nerdism

2025-09-11T11:41:00.000-07:00

I'm a HAM. Not that I'm funny, seriously, I'm a HAM. An amateur radio operator. I function as an emergency contact for neighborhood issues when infrastructure is down, as well as a point of contact and support for the city (I work with both data networks as well as voice networks in the radio environment). I also work for a local healthcare company on their radio connection network in my spare time. It's not what I do for a career, this is what I do for kicks and giggles. My current portable radios are poor, my wife doesn't want an awful antenna sticking up from the house, and that leaves either a permanently mounted in the vehicle (it's been broken into once or twice, so things are not left in there) or a portable base station with some power.

Old ammunition storage boxes are often referred to as "ammo cans", so taking it into the world of radio and adding an "h" on the front is not a long stretch. I am NOT the coiner of the phrase "hammo can". I refuse to take credit on coining the terminology because, frankly, I'm not that smart.

Under the circumstances outlined above, a "hammo can" is the perfect option. It will be light weight, portable, and not tied to a vehicle or property. I can easily set one up under a canopy in the yard if it's needed. So, I finally decided (11 years after first obtaining my radio license) to venture into the "hammo can" world. So, that proverbial bullet was bitten, and I started ordering parts :

The radio is a TYT's "TH9800D". It's tried and true, and a lot of HAM operators swear by it. Theoretically, it will fit a Yaesu FT-8900, but I cannot confirm that without a radio. This one also requires three m4-0.7 flat head machine screws, one being 16mm long and the other two being 12mm long.
The container is a 50 caliber ammo can off of Amazon. I ordered it in purple (they have other colors, and this one visually separated itself from my grab-and-go case of hard drives).
The battery is a 12v LiFePo battery to be able to run this station for an extended period of time.
A power manager (AtemPower AP100BM) is used in this build (and some switches so I can completely disconnect the battery for storage).
Battery terminals (2 negative [black], one positive [red])
Random switches and battery connectors for easier charging.
Heat-set inserts for m3 screws - while it started out with some m2 screws threading directly into the plastic, it was eventually changed to all-m3 (except for the radio itself) for consistency.
27 m3x10mm machine screws (add four if you print this with the two-part top plate).
2 m3x10mm pan head screws (for the radio face-plate-to-battery box).
Heat shrink tubing
1/4"-hole circular connectors
1/4" female spade connectors for the battery
A PL-259 90-degree elbow
4" Cable with PL-259 ends (male and female)
Mounting connector for a PL-259 for the top plate
Wiring (I used a radio adapter that had long cables, and I just used the extra cable for all my power needs)

So, I have the hardware to have a "hammo can", but it is kinda ugly piled into a chaotic mess, and I don't want to re-organize everything before I can use it in an emergency. I need this thing to be ready to run with only setting up an antenna. The simpler it is in an emergency, the better.

Design

That lead me on a course to print an insert for the "hammo can". I mapped out a lattice, and started gluing things together in FreeCAD. Please note - by the end of this write-up, the structure will change. There is no lattice for rigidity, and it is all handled completely by brackets. This will give more parts to print, but reduce the over all time it takes to print it. But, let's skip this digression and just move to the original design.

There are compartments to slide the battery into, the radio into (faceplate goes on top), and the power management adapter. So, I also loaded the "thing" to thingiverse if anyone else wants to go down this course of action.

I think this will work sufficiently, but it looks like it's just about 2 weeks to print (17 days). Plus, without a brim, it looks like it just barely fits on a Creality CR-10s build plate (with a brim, it did not fit). I do want to make one change - I'd like to make the radio area a bit more "modular" so you could potentially use other radios, so this could become a more viable option to build one-off hammo cans. We'll see if I do.

I got into gear by printing the power module bezel (I was using it as the test platform for interrupted filament prints because the chassis is going to be 2.5 reels of filament (yes, that much). I don't want to be so attentive for two weeks while this prints, so an empty filament reel should result in a pause. That's two prints that I can use to make sure it is all running properly. The power bezel print did abort, while I figured out the issues with the filament sensor housing, and I printed a new housing and had it working okay, so I could run some more tests.

I managed to get the bezel printed with interrupted prints, so I think I can test this once I get the time to do it (mid-to-late August at this point).

The bezel plate came out great.

The bezel plate will just screw in to the chassis. There is one section I'm going to have to modify for a power shunt, so I might as well get the radio caddy a little more modular, too. I spent a few hours getting things set up for a modular setup.

I then built out the shunt plane, as well as a modular radio, and also printed out the switch plane while I was at it.

The switch panel looks good.

There are two "generics" in the thingiverse file - one radio module and one shunt plane. But I also have the shunt plane for the AtemPower monitor and the TH-9800D radio loaded in there. The "generics" are in case you want to try something a wee bit different (e.g. fit in a different radio, for example).

I then started to print the shunt plane for the AtemPower (not the generic one - I don't need one of those). Of course, the shunt plane print failed four times. I re-leveled the bed, changed the nozzle, and then finally gave up and ordered a new PEI printing bed plate (I ordered a few, just because this could use up more consumables like the print bed). Then I started printing it yet again. I'm getting close to running out of filament on one spool, so this is perfect - it's going to give me a real test of the filament sensor break.

The print was good - it finished right on time. The shunt plane finally finished, and did not warp. Now, if this should look funny, it's because I wanted to finish using up my silver filament. I knew during one of these parts I would have an opportunity to prove out the filament sensor interruption, and it came about 2/3's of the second to last layer. So, the base is "black", but the last layer is "silver". It's going to be buried in the can, so I'm not at all worried about a discrepancy. This print took about 4.5 hours with the filament reload.

Now for the radio module. Since I had the silver installed, I simply just went for that. It will all be hidden on the back end, so I'm not terribly worried about presentation here.

Again, I created a "generic" in the thingiverse file for this, but also a specific TH-9800D variation for my implementation. If someone wants me to do a non-generic module for a different radio (if they fit in the same basic size), they can always donate a radio and I'll try to design it to make it happen. Again, in theory, this may fit a Yaesu FT-8900 or 8800 radio (since that is what the TYT TH-9800 is based on), but no guarantees. Send me one, and I'll make sure it does work. I probably won't send the radio back when I'm done.

I only printed the TH-9800D radio bracket, not the generic, so I cannot vouch for the generic one's functionality, but the specific one came out well. I assume the generic one is fine, since the TYT-specific one was based on it. This print took about 16 hours.

I realized that my measurements were 0.039" off on the screw spacing, so I adjusted the module, but rather than take 16 hours to reprint, I just modified it using a drill. The radio body fits in using one m4-0.7-16mm screw on the top and two m4-0.7-12mm screws on the bottom.

Next came the print that would take a few weeks, the one I've been testing the filament change in anticipation of - the actual chassis. Seeing that my previous prints have come out slightly undersized, I did a quick google search, only to find out that an STL doesn't have real dimensions in it, so your slicer can only guess so much. This means that you have to adjust the size in your slicer before you slice it. That leads me to the following. The chassis itself is 6.102" (154.9908mm) wide, 11.217" (284.9118mm) long, and 6.315" (160.4010mm) tall. Okay, on to the printing. My slicer seemed to have it pegged on dimensions well, so, off to the printer. I know I should adjust for shrinkage, but meh. We'll see.

This should start out in silver filament (because, again, that's what is on there), but should switch to black before the top layers when the silver filament runs out (it takes 2.5 rolls of filament, so it will definitely run out) - and that means it should look just fine on the outside. My CDO (OCD in the right order?) will know the bottom is silver, but it will still function.

Yeah, that's a few layers of silver below the black. Sure, it's going to look whacky, but I don't care. (The fact that I am yet again referencing this "issue" probably means a second print later.)

This is going to take a while. The slicer said 13.4 days, but the print isn't coming across at the same rate. After 33 hours of printing time (started on the 11th at 8:54 AM), I've calculated each percentage point as taking 249 minutes, or 4.15 hours. The stats claim that it will take a grand total of 17.3 days (on the dot, which puts it at finishing on the 28th at 2:54 PM) - which is what 415 hours equates to.

At the 20% mark (8:10 AM this morning on the 14th, 72 hours and 9 minutes in, I revisited the stats. The target date shifted from the 28th at 2:54 PM to the 26th at 8:46 AM. The per-percentage time went from 249 up to 277, which makes me question my math in one of these two check points. We'll see what happens. It does look good, and taking random photographs will help me post-process the supports, clearing this up. As I work through it, the photos will definitely come in handy.

The next check point, 7:37 AM on the 15th, was 27% complete, with about 1.888" of height, 94 hours and 36 minutes into this (5,676 total minutes so far). On this one, I took the time to also measure the chain guard distance to the stabilizer, because I know the stabilizing rods get in the way of the x chain. That came out as 6.146" of height remaining, and the calculations are to 5.664" of height left (which falls a lot longer than the 6.315" of total height).

This 21,022.2222 minutes in total, or 14 days, 14 hours, and 22 minutes. That brings the new estimate of completion to the 25th at 11:16 PM. Of course, there are some filament changes that will have to happen in there somehow.

My next check point came in a few times on the 18th, while I started paying closer attention to the filament running out for a second time. I clocked it at 171 hours and 51 minutes at 12:51 PM, and had the filament changed at 172 hours and 15 minutes at 1:39 PM, approximately 24 minutes in print time, but 48 minutes apart. That means that the print timer itself is total time, and does not include time between filament being out to the change time - or print pauses. That's good to know.

So, the estimate for the next filament change is the 24th at 8:46 AM. At that point, we should be about 97% to 98% done with this, and we'd only have 1 day and 7 hours, and 53 minutes remaining (25th at 4:39 PM is the new completion target, depending on how much time I waste forgetting I need to change the filament).

It is now 62% done, but at 59% done, it was 203 hours and 45 minutes into the print, and 4.374" tall. If this was fully linear, it would mean that the print would be 7.414" tall, but we know that the print should only be 6.315", so we have nearly a full inch shorter. This is a solid indicator that the print percentage is NOT in layer count, it is either in over-all time, or in total distance of filament printed. At this rate, it should be 5 days, 21 hours, and 35 minutes remaining, a target completion of the 25th at around 9:50 PM.

Now, if only I could wait patiently (and hope we don't have a power outage in the next 5 days, 13 hours, 22 minutes, and 46 seconds).

Bugger.

9 days in, at 66.3% complete, I walk into the room (I'm not living in the same room as this thing for a week) and hear odd noises.

The nozzle is being scraped across the top of the print!

Apparently, I printed a 6.35"x6.1"x4.6" pile of a roll of filament. I aborted the print job, and grudgingly went back to the drawing board. [sigh].

No fancy lattice work, no huge set of prints, and all of the modular parts will still fit. This also includes other improvements that were re-developed int6o the new structure (I thought I would see interference on the corners with the curves in the can itself, so I added some fillets to work around that). So, off to print a few parts for the printer itself, then to start with the four new parts :

The bracket that ties the shunt plane to the radio, the battery, and to the top plate
The extra corner bracket (where I have some space that could be used in the future if I wanted to add networking components or Raspberry Pi's)
The new top plate with mounting tabs
Battery box

All said and done, these four new parts should take approximately 5 days to print - a massive difference in time down from the 15 days it was going to take. Additionally, these new brackets can also be adjusted to make this even more customizable. The top did not fit with a raft - so I created a 2-part top plate. However, I really don't like the visual, but it does still function, so if you are making one of these without a large printer, it will still work.

The top parts did get plastic-welded together after being bolted with screws. Here it is without the plastic-weld, which is still functional.

And the visual... really isn't so bad! The top panel pieces took 40 hours and 40 minutes on the CR-10s with 100% infill and a raft, which is why the surface is a bit sub-par. Here it is with the top panels all combined, and with no wiring.

Next was the battery box. Started at 9:44 Aug 31, followed by the brackets. And, after all that, it finally started to come together!

Assembly

Install the heat-set inserts in ALL the parts. If you used the two-part top plate, you will use an extra four of the inserts, and an extra four of the m3x10mm machine screws.
Next, install the radio itself to the radio module.
If you printed the two-part top plate, combine the two top plates together with four screws. Plastic-weld the two parts together (*this is purely optional*).
Fit the switch plate to the top plate, and screw together.
Install wiring on the radio body side of the switch (you won't be able to get to that after you install the radio).
Temporarily install the radio face plate (the part that comes with your radio) to the adapter block, but do not tighten.
Install the radio face plate to the top plate.
Measure where the radio adapter bracket (what came with your radio, not the block itself) sits, and remove the block and radio face plate.
Re-install the radio adapter bracket without the radio face plate itself to the adapter block.
Install the radio face plate adapter block and bracket to the top chassis plate again.
Shift the radio face plate adapter until it matches what was measured above in step 7.
Tighten the radio face plate adapter bracket (what came with the radio) to the adapter block.
Remove the adapter block, and add the cables required for the radio face plate, and install the face plate to the adapter block/bracket combination.
Install the battery box to the top plate through the radio face plate connector.
Install the radio face plate to the top chassis, running cables into the cavity underneath the box where they can be connected to the radio.
Fit the power module into the bezel, then attach it to the top plate.
Install the battery connectors for the top plate.
Install the radio module itself to the top chassis.
Install wiring to the battery, and to the power module, the remaining switch terminals, the battery terminals, the battery box (add the connectors for the battery, too) and the radio itself. In my case, the negative battery was wired to the switch, then to one battery terminal. I essentially used the battery posts on the top chassis plate for the interconnection of wires.
Install the extra bracket.
Install the shunt plane bracket (not the adapter).
Insert the battery to the battery box, and strap it in using a velcro strap (not really required, but recommended to prevent the battery from bouncing around) to secure it in position.
Route wires around the shunt plane bracket.
Install the radio module shunt plane adapter and shunt plane.
Program the radio using your computer (again, while optional, I highly recommend it).
Drop the block into the hammo can.
Weather strip around the top edge to lock the panel into position and secure the chassis.

I did use a strap to hold the battery when I started putting it together. It rattled around and went flying through the air and landed on toes when pulling the chassis in and out of the can while assemling.

I finally got the last few parts to finish putting this together (the one remaining battery terminal, and the antenna connector).

It does fire up, and things appear to be operational!

Your radio should now be ready for an antenna and an emergency. A good source of information is the "Radio Reference" database (https://www.radioreference.com/db/aid/7771) where you can look up frequencies specific to your location. I would recommend (if your radio supports these frequencies) :

27.065 (CB Channel 9) - Emergency
52.525 (6-meter) - FM emergency band for ham radio operators
138.225 - FEMA disaster relief (primary)
146.52 (2-meter) - non-repeater, HAM frequency
Local Fire, Police, and Search and Rescue (frequencies will vary by location)
156.75 - maritime weather alerts
156.80 - maritime distress
163.4875 - national guard emergency channel
163.5125 - armed forces disaster preparedness
168.55 - federal civilian agency for emergencies and disasters
243.00 - military aviation emergency
462.675 (70cm) - GMRS emergency channel 20

Once set up, connect your antenna and give it a test run.

Summary

Now, while I don't like the finish of the top surface, nor the fit, I think this will be operational for what I need, and I secretly think this is awesome. I'm hoping to surround the top plate with weather stripping, and put some custom gaskets beneath the switch and power bezels for a bit more moisture-sealing, but with the gaps under the radio faceplate itself, this is definitely NOT water tight. I'd like to revisit an antenna (if I can find a purple pipe wide enough to protect the antenna as well as act as a bit of a stand) to see if I can get that a bit more appealing.

With all of these prints together, it was time to assemble the whole thing.

And with that, we have a "hammo can" (portable base station) for emergency use!

Old Hard Drives

2025-08-22T10:40:00.000-07:00

When you visit family, and someone hands you an old hard drive from 22 years ago and asks for a data recovery because it could have your moms letters on it, you don't question it. Grab that hard drive and take it.

So I started a little journey. I popped it into my PATA USB case, and plugged it in. It did not register with the system (I'm using Linux so I should at least see a USB interface get logged, even if the hardware has failed).

It spun up. There were no clicks (so the hardware had not failed). I dug into the logs - nothing, indicating the interface board was likely the problem.

So, I took the interface board off. It was a Western Digital WD400 from 2003 (I told you it was 22 years old). I grabbed my Western Digital WD400 from my case and started to swap the interface boards temporarily to commence a recovery.

One screw changed between them. There are associated with the WD400 product line, but actually have different model numbers. I'd wager that if it wasn't for that screw being in the wrong place and not being able to be anchored down properly, it would have been perfect, yet this would not be.

I could have 3D printed a bracket that fit across and held the interface circuit board down, but I just put the original back in, and looked at the jumpers. Old PATA/IDE hard drives had a jumper on them for "Master", "Slave", and sometimes one or two other options (such as "CS" or "Chip Select"). There were no more than two IDE drives on the same IDE cable because of the electronic design. One would be designated the "master", and the other a "slave". In later years, they added a "chip select" option to try and make it a little more plug-and-play.

I had the thought I should move the jumper from Master to CS, and suddenly the old interface loaded when it was reconnected. Ah hah!

It loaded no files (an empty filesystem). The next step is always to run a recovery, just in case a new filesystem was put over the top of an old one. So, I ran :

photorec /d Documents/recovery /dev/sdc

This dumped a few web files (CSS, XML, and HTML), a few executables (EXE and DLL), a few TXT files (that turned out to be nothing more than configuration files, registry files and some text extracted from executables). Two files, a WAB and an EDB, both stood out to me, because a WAB is supposedly a Web Address Book, and EDB is an Exchange Database. Turns out, they were simply mislabeled data files for the executables.

No, the drives did not have any letters from my mother. But it's better to know than not to.

Doh! A Delicious Dish of Dumb Dumb!

2025-08-05T12:21:00.000-07:00

Alright. I'm only posting this because of the blatant error in my own reverse engineering. I had a Coleman pop-up canopy, with fancy "push button" locks and unlocks. It's really not bad - it's survived for 10 years with only minimal issues. (Two years ago, one of the locking mechanisms on one leg wouldn't release.) Here it is in all it's glory (still in it's bag).

This year, that same locking mechanism released (perhaps a better term is it fell apart). Since I had silver filament and these were gray, I thought, what could I lose? I'll see if I can draft and model up a replacement part, and print it.

I started out by taking the few remaining parts off to see what was broken. This is a pretty ingenious device. I'm impressed with the design (but not so much the implementation).

The broken part is circled in red above. A few more photos, just to make sure :

It is time to create the model. I whipped this up in FreeCAD :

About 50% done (I'm doing 100% in fill so it is solid), I picked up one of the parts, and realized - I didn't look deep enough. There are two locking tabs on each end (I got both of those), and... recesses in the mating part where there were supposed to be four more locking tabs on the sides.

Really? I play the part of an idiot VERY well, sometimes. I'm missing the ones in green here. Notice, I had them listed as an alignment pin. Looking closer to the original, and these locking tabs had broken off (three of the four were down to the plate, one was half height, and I built all four off of that one).

So, back to the drawing board. I simply copied one of the two tabs and positioned the four new instances.

Okay, we can try printing this once more. Here we go. 8 hours and 38 minutes.

Alright. The overall length should be 2.904", but the final print came out at 2.890. In Cura, it shows 73.6, which is about 2.898", so we're losing about 0.008" in the print, and Cura is also not getting sizes from the STL file with much accuracy from FreeCAD. That should be just fine, really, as long as I remember to adjust the size by 100.48% before slicing.

It looks good, the clean up was good (the aborted print gave me an idea of how this was going to be cleaned up), and with some fine adjustments (hobby knife and a Dremel for some inside work), it works exactly as I'd hoped.

I went to install it, and it simply didn't lock together. So, that led me on a re-do, and I added the toggle cover with a new base that screwed together (no tabs). This particular variation uses screws to bring the cover and the main housing together, but uses the same toggle switch itself, the same spring, and the same spring assembly.

Note, be careful how you print this - if you print side-to-side instead of top-to-bottom, the locking tab that holds the housing to the actual metal upper leg will break, and then you have to superglue it into place (ask me how I know).

I have my canopy back!

Moving Badge Access to your Finger

2025-07-24T10:00:00.000-07:00

I work for an entity that implements physical security for pretty much everything. Workstations in the office can even have their own badge reader (through "Imprivata") to "login". One group of employees we cater to absolutely uses these systems to keep notes, and to be able to move from one office to another quickly and effectively. The badge access systems make "authentication" quicker and slicker.

I wanted to make it even easier than reaching for my badge to tap in and out of the computer, so I bought a T5577 RFID writable ring off of Amazon, and an RFID reader/writer. At first, I got fairly frustrated that neither of those tools did what they were supposed to do.

Then I made one adjustment, and things were suddenly magic.

So, what did I do?

Use the mode key to select "HID 125KHz".

Once the HID 125KHz is selected, insert the badge, and press "read".

The device will then say to put the T5577 card in and press "Write". This will take some trial and error to find the right position to hold the ring for writing.

Once you have a successful position and press write, it will tell you it was successful.

With that, it was time to test it. At a workstation, I used my primary badge, and locked the session. Then I tried with my hand (ring) to unlock, and the light indicator went to "green" from "red", showing success.

The workstation unlocked, angels sang from the skies above, and I was suddenly being served pina coladas on a beach in Tahiti. (Well, one of the three was true.)

Now, it did NOT work at the office doors - those apparently require encryption on the badges that this simply cannot do. I'll have to think around that.

Folding Desk

2025-05-27T07:21:00.000-07:00

There have been a few times where I've had to temporarily set up "shop" (my workstation) in different places around the house. For example, when we have a visitor and the bedroom immediately next to the workstation is occupied and the time they wake up doesn't match.

These situations have usually had me splitting the expanding table and attaching a dual monitor stand in between the two haves, then pushing back together and setting up in the kitchen.

I have a folding drafting table, and I wanted something similar I could use for a workstation. I was severely tempted to just purchase a butcher block top along with a folding stand, but if you know me, if I'm going to put one together, I might as well go all the way. I had some aluminum extrusion laying about for a different project that ended up not being used, so my brain said.... why not?

I set about designing the desk in my head, and then just threw it together. Here's what you'll need (with modifications from what I did because you likely will make changes). Since I was doing this with 2020 series extrusion, my bolts and nuts were all M5.

3 fir boards 2x6 that are 10 feet long
Biscuits
Wood glue
2x 48" horizontal extrusion rails for the main body (mine were shorter, and you'll see overlap on the folding ends later, which messes with my CDO, er, OCD)
2x 43" slanted extrusion rail to stabilize the main body (these form the "x" in the main body and stabilize the whole thing - one will need to be cut into two pieces, but do that after the first one is in place)
4x 23.5" horizontal extrusion rails for the folding ends
2x 27" slanted extrusion rails to stabilize the folding ends
6x 26" vertical extrusion rails for the height of the desk
8x 45-degree brackets for the extrusion rail
8x 135-degree brackets for the extrusion rail
16x corner hardware brackets
76 bolts (I had 40 for the piano hinges, so your mileage may vary depending on your hinges)
76 washers
76 extrusion nuts
2x Piano hinges about 20" long
12x extrusion end caps

Tools you'll need :

A square
Allen keys for the bolts
A biscuit cutter
A chop saw for the boards
A band saw to cut the extrusion to length (and some angled cuts)
Panel clamps (I used quick clamps along with off-cut boards to keep flat and Saranwrap to keep those off cuts from sticking to the glued up surface)
Sanding pads
Planes (optional)

First, cut the extrusion to the lengths you need.

Second, for the slanted pieces, cut the folding ends to 45 degree angles (angles are NOT parallel for these, but opposite of each other). The main body angles ARE parallel.

Third, start bolting each frame it all (except for the second cross bar in the main body) together.

Once it is all together, you can mark where to cut the second cross bar in the back, cut it, and install that one half. Then you can line up the other half and mark where that one needs to be cut as well, cut it, and then install it.

With that done, you can then mark where the top bar of the main body connects to the cross bars. Then, remove the top bar on the main body, and cut those two pieces off, then reinstall the two pieces. This should keep the whole thing somewhat stable. That middle section shouldn't be discarded yet - in fact, you will cut a little more off and re-install it about 6" below where it originally was. This should give you clearance. (Pictures don't show it offset until you get to the end of this post, and they don't show the second cross bar installed, so be aware).

Fourth, install the hinges (but don't tighten quite yet). With the folding area in place, open one end perpendicular to the main body and get it square and vertical to the floor. While in that position, tighten the open ends' hinges to secure it in place. Repeat for the other folding end.

You should now have a folding rack and desk base.

Yes, notice the folded ends overlapping. This is because my desk was less than the 48" I've now recommended above. But it's still "functional". Anyway, back to the build.

Next, cut the fir 2x6 boards in half. This should fit the span of the desk. Line them up in position on the desk surface, and mark and cut for biscuits.

Then, glue it all up and wait.

With the surface glued up, it's time to finish it. Granted, you don't have to, but you won't be using the desk to write with the way 2x6's have curved corners. I'd recommend you use the hand planer to get it flat, and then sand and finish. However, you may want to simply pour epoxy in (sealing the ends, of course) until you build it up far enough to have a flat surface. My unfinished, but usable workstation :

I used lacquer spray to somewhat seal the bottom (I did not spray the top), and then taped the bottom and edges so I didn't have epoxy run all over. While there, I also hammered some small gears, screws, etc into the surface. All of that was filled with a metallic bronze epoxy mixture, then sanded. I went to 220 grit top, bottom, and edges, then went 800 grit on the top and edges. Then I lacquered the whole thing (yes, even the bottom again). Once done, I hit it briefly with 2000 grit sand paper to knock off the little spikes that form (this spray can lacquer isn't the greatest and smoothest finish after spraying).

That gave me a phenomenal surface that won't poke holes in paper, and won't snag on clothes. It also has a great look to it. The epoxy will look black or dark unless the light hits it at a specific angle, so everything blends together in a fantastic way.

I will have to build the nipples for the bottom of the table, the threaded tubes to make it a standing desk, and the fixtures for the top of the stand itself. I whipped up the specs for machining the 12 remaining parts required (for what I'm doing at least - you COULD adjust with a top hinge and make it only a fully folding sitting desk. I want the option of standing, so I'm manufacturing a few pieces yet.

I built a fixture plate that bolted to my lathe faceplate (so that I could screw the next part onto it and just make it match), and then did some bandsaw work so I could reduce the amount of turning work I'd have to do, and started cranking through.

I needed four nipples for the bottom of the desktop, four threaded mounts for the frame, and four pipes with one end threaded to the same thing as the frame mount. Plus, I had to shorten some M5 screws.

Once these were created, I could install the frame-to-table parts.

Then it to flip the desktop over, line it up, then install the nipples to the bottom.

A final test before I install the standing spacers :

And the final spacers :

I did slap the laptop on it and work in standing position for an afternoon. This will work well for me when I need to be on the move.

Next, I didn't want to lose the aluminum "standing" rods, so I printed a storage case that simply connects in and locks them when not in use.

Now, I went to use this for a little bit, and there have been some issues. When it was in "sitting" mode, the height just wasn't quite right. So, I trimmed off lengths 3" (I already updated the measurements above). When I tried to use a dual monitor stand that snugs onto the desktop itself, the rear, top bar was in the way. I added a second cross bar in the main body (pic below), and I cut out the top cross bar and moved that down 6" so I could have clearance for the monitors. When I threaded the standing-mode bars on, some wouldn't go on all the way. I ran a threaded die onto them and cleaned up the threads (I had to use an adjustable die so that I could get it perfect, and that meant running it over the threads a few times while adjusting it tighter in between until it fit just right.

I'm loving this thing! It is now perfectly level, it has excellent stability (though it is a bit heavier), and it all just works as it was designed.

Now, if I have a mobile Internet connection, we could go just about anywhere and work, and my family can wander off to see sights while I work. This would be good, as long as the family doesn't get eaten by bears without me. That would be bad. Perhaps if I'm eaten by bears along with them, but not without them.

Chess - Ah, Nuts!

2025-05-27T06:33:00.000-07:00

A group of chess enthusiasts returned to their hotel room after a tournament, and were still standing in the lobby while bragging about the defeats of their opponents. After an hour of loud nerds, the manager told them they needed to disperse.

"But, why?"

"Because," he replied, "I can't stand chess nuts boasting in an open foyer."

Perhaps, I had better apologize for the pun. I'm not sorry for it, but I will apologize.

I'm one of "those" kinds of people who loves chess, cars, and I do spend my fair share of time on Pinterest. I spend little precious time playing video games, and a lot of time just exploring. And, while in Pinterest, I have seen numerous instances of chess sets made out of fasteners (you know, nuts-and-bolts kinds of things).

Now, a little insight into me in general - I have chess sets. I have chess sets my dad gave to me when I was 8 or 9 years old. I have marble chess sets, glass chess sets, wood chess sets (a folding one, and a wall hanging one), and I had a friend make a set for my dad in a machining class out of aluminum (it's anodized blue and clear - it is a beauty). I'd even made my own by using an existing set and creating molds for the pieces, and casing my own out of green and gold resin (I made a folding chess board back in 8th grade wood shop class - it is definitely rough, but I'm keeping it as is). I have sets for Lord of the Rings, and I'd like some other resin-cast variations. Simply put, I LOVE chess sets!

But I do not have my own metal set, and these looked intriguing. I wanted one. The materials are not a cheap way to get into the game - trust me, you can find MUCH better options out there (that are even regulation sizes and a bit more portable for things like chess in the park) as cheap as $21.

So, I decided to try my hand at it, just because they look pretty darn cool. Here's what I ordered :

10 brass M10-1.5 20mm bolts (for 8 pawns and 2 knights)
2 brass M10-1.5 25mm bolts (for 2 rooks)
4 brass M10-1.5 40mm bolts (for 2 bishops, a queen and a king)
10 brass acorn nuts for the pawns and bishops
4 brass M10-1.5 nuts (simply to lock some other parts to the bolts)
3 "castle" M10-1.5 nuts - if you find these in brass, I'd be surprised, so you are looking for actual castle nuts with a gold tint - it's close enough for the boys my wife dates
2 brass lamp rings (these were M10-1.0, because you will not find these with a 1.5 thread - but that's okay, because you're just going to tighten them on anyway (without the nuts to lock them) as far as they'll go.
1 brass flange nut (for the queen)

Repeat the purchase for the other color (I went with black to give it a good contrast - for machining, you basically have brass/gold, black oxide, polished stainless, or just a steel). I've thrown it all into an amazon list if you are extra lazy. Just know this is not exactly cheap.

Now, it is time for assembly.

For a pawn, just take 8 of the 20mm bolts, and tighten the acorn nut on as tight as you can by hand. Maybe a little tighter, but this isn't structural, so you don't need to worry about it being really tight.

For a rook, spin onto the 30mm bolt a regular nut, followed by one of the "castle" nuts (they are officially called "castellated" nuts, and they are used with cotter pins to keep a nut from spinning off of a thread for shafts in things like cars). Put the peaks upward, and when you have all rooks top nuts in the right position, bring the other nut against it to lock it in place.

For the knights, just thread the lamp rings on. The improper thread size will lock them in place fairly quickly.

Bishops, treat JUST like the pawns, but with the 40mm bolts.

For the king, just like the rook but with a 40mm bolt.

And for the queen, use a flange nut upside down and a lock nut.

The set doesn't look bad at all in the above photo, (it is incomplete, e.g. the queen is not done). Additionally, I assembled some black pieces, because contrast between black and gold is just... beautiful.

If you look closely enough, there are still some parts missing (black rooks and king are both missing the "crown", brass queen is missing the flange nut). Once those arrived, it was relatively easy to complete the set.

Still, I'm debating in my head around making a chess board using gold and black without machining one from stock and going through the black oxide process, etc. Still working on that thought, but it might ultimately come to that.

A Dimensional Awakening - An Old 3D Printer (Creality CR-10s)

2025-04-10T15:49:00.000-07:00

Since the cancer scare of 2024, a host of old projects are just itching to be scratched [off the list]. One of those projects was my old alarm system needing new hardware around the garage door (the wood it was made out of seems to disintegrate over time). So, it was time to break out my old Folgertech Prusa i3.

I couldn't get it upgraded to handle ABS, which is where I wanted to go for these parts. I ran through updates, I adjusted beds, I even bought an enclosure for my Prusa. ABS was no bueno.

So, I hit the local classifieds and bought an old Creality CR-10 for $75 with hopes of getting it to be my ABS-dedicated machine. It was extremely dirty, gross, and misaligned. Plus it was a CR-10s, not a CR-10 (note, not a CR-10 4S - that extra digit and the space make the "s" mean something totally different). This "s" means it's supposed to be "smart", and has a filament sensor on it which allows it to detect when filament is running out. There were a few broken parts on it, and a few customizations (e.g. the fan shroud was a dual 5015 fan that was a simple, but really elegant design).

First, filament sensor housing had snapped and was loose. There was a spare piece of filament jammed into it to bypass this, which was okay to get this thing back online.
Second, the extruder cooling fan shroud had cracked. The previous owner had glued it together using superglue glue, but this had failed under temperature.

The housings for both failed parts were after custom, and printed with PLA, which doesn't hold up to heat very well and considering sits around the extruder with high temperatures, is not an ideal material for a fan housing. But, I think I can deal with a fan that is slightly loose while I use the printer to replace some of it's own parts.

So, I cleaned it up, bought an enclosure, bought a new build surface (the gold PEI plates work best - the black one stuck a little too well and shredded after a few uses because I couldn't get the parts off), and leveled it out. I started playing with settings and ended up at 255°C on the nozzle and 95°C on the bed. That worked very well for the first few prints (I used real, usable and needed parts as my test prints, such as x-chain links for small parts).

I then backed down to 250°C on the nozzle and 90°C on the bed. I'm using a brim to keep things as glued to the PEI as possible (again, the gold magnetic PEI build plates seems to be the best as they release when the temperature cools enough, but don't hold onto it like a snake locking jaws in a bite).

I have successful prints. Now that I know it's ready to repair itself and it functions for what I need, I bought some 6mm extrusion cover in black, and changed this thing to start blacking this thing out.

I have a full x-axis cable support chain ready for a fan shroud housing. It's time to do the filament detector housing, as well as the fan shroud to connect the x-axis support chain up to.

Bugger. I can't find the original custom parts online. I was going to use those because I KNEW they worked.

So, I had to design and build my own. Out came the calipers, micrometers, and gauge pins to map out and chart the holes. This was a few weeks of printing and trial and error while I got everything mapped. First was the filament detector housing. I started with someone elses STL, loaded into FreeCAD, and converted it to a solid to start playing with it. I angled it up a bit to reduce tension on the filament inside the housing, and ended up with something I liked.

Since I stole someone else's design, I could re-use the same lid/cap they had designed. I printed it out, and... it warped. Another bugger! But, it was still usable enough to get that part online. So, onto the machine it went. And it was good... for a little bit. When I started installing the z-chain cable management, not so much. So, back to the drawing board. I came up with this thing posted to thingiverse.

Next, I designed the replacement fan shroud. I used the same tools (FreeCAD, calipers, micrometers, gauge pins) to map out the existing, known-to-have-been-working design, plus I wanted to add the x-axis chain cable support. I got to where I liked it.

I tried printing the fan shroud, and... it warped so bad it was unusable. [sigh].

So, I turned off the cooling fans, and reprinted.

There was MUCH less warping (but there was still warping around). I cleaned it up, and tried to get it installed....

Fantastic! The third time was the charm! To finish my x-axis chain, I just need to print the vertical bracket and install it! This printer is essentially better than when I got it. The broken parts have been repaired, and additional parts are being completed. The final fan shroud is available on thingiverse under the name of "Creality CR-10s Dual 5015 Fan with X-Chain Interface" :

https://www.thingiverse.com/thing:7006624

Started working on the cable management (I'm doing all three axis). One thing I don't like about the x-chain is that if you are using tie-rods to stiffen it up, you DO lost z-axis. I'll have to address that later.

So, once the cable management was all complete, it is a pretty solid printer. I really do like this thing.

Now, about that filament sensor.... the next itch that needs scratching is the upcoming "hammo can". It's an insert for a .50 caliber ammo can that sets up a mobile HAM radio station. It doesn't fit on this Creality CR-10s with a brim, but the bottom plate is large enough that it shouldn't need one, and without the brim, it just barely fits. That's going to take 2 weeks to print on this server (and 2.5 roles of 1kg filament), but it's now doable.

Pfaff! A New Machine for Me!

2025-04-08T15:36:00.000-07:00

305.076.08

I have an amazing aunt (well, my fathers aunt, but I claim her). Every once in a while, she contacts me about a family heirloom. Well, she knows exactly what I do for kicks and giggles, and when she realized her family had quite a few of these sitting around, she immediately contacted me and asked if I wanted one.

My first thought was "it's the wrong family", but it didn't take much to twist my arm. Truthfully, I was twisting my own arm.

I ran and grabbed it.

It is a Pfaff 130. The machine itself is in surprisingly excellent shape. It doesn't rotate very well right now completely around, which indicates that something is likely bent inside, or it just needs a good cleaning. The belt has disintegrated, but other than that, it appears to be well taken care of as there's not a spot of rust on it. Kudos to that side of the family for taking such good care of their machines.

First things first, I need to try to identify this a little more than just using the model number off the front, because they were made for quite a while. I found the serial number on the inside, "5 123 825" :

A not-as-easy-as-expected search (took a few manipulations of key words, but that is all), revealed a page out there where Brenda Dean of ISMACS' locale released a basic chart

https://ismacs.net/pfaff/pfaff_manufacture_dates.html

And that tells me precisely that my new toy was manufactured in 1954.

I went through this thing, and I ordered a belt (some websites linked to this 14-3/4" belt) #305.076.08. It did NOT fit. Instead, I opted for a 15-1/2" belt, and with some adjustment, it works beautifully. The motor and electrical are good to go, and everything has been oiled!

This will let me sew mild leathers for car seats, canvas for carrying cases, etc. I'm excited for this new toy!

Port-A-Lign - Misappropriated For Another Use

2025-03-01T11:34:00.000-08:00

I have a need. There were a number of un-identified reamers that I need to officially know their sizes. My milling machine doesn't have the reach, and I don't want to chuck up long bars of stock sideways to be drilled. The best option I came up with was the old Craftsman Port-A-Lign. It's led to some newer variants, but those were mostly plastic, and I struggled with those thoughts - the Craftsman has the drill chuck centered between the rods, and that makes me think it will be a better option to abuse.

Starting out, I need more length (some of these reamers are much longer than jobber drill bits). I grabbed the diameter of the rods for this as 0.496" :

Next up was to identify the spindle that the chuck sits on. It's a 3/8-24 thread, but I'm missing part, and also the chuck is a 3/8" chuck. I needed to be able to handle a 3/4" diameter reamer, so a new chuck was in order, and none of those have a 3/8-24" thread mount. That means I need to make a new spindle, with a JT3 mount on the end for the 3/4" chuck.

I pulled the spindle out. Two clips hold that spindle in place :

I found a 9mm socket fit perfectly on the end to drive that spindle out. I got my measurements :

(Not written down is the JT3 taper end, which is 1.2188" long, and the two ends are 0.7461" and 0.8110", or a taper-per-foot of 0.639", and an angle off of center of 1.5251 degrees.)

I chucked the new drill attachment into the lathe (1/2" straight shank, with a JT33 end for a 3/4" drill chuck), and cut a groove down to 0.467" (inside diameter for the new e-clip). Next the shaft went into the freezer, and the chuck went onto a heater vent to get a good temperature differential for assembly.

So, I have a fully functional Craftsman Port-A-Lign with two interchangeable chucks, the original chuck, and a 3/4" banger for use in first drilling, then reaming a hole. It's only to test reamer sizes, remember? Sheesh, my memory has really taken a hit with this chemo thing.