For most of my childhood – the 80s and 90s, these trains ran via Allentown, where they had to “run around” the train, putting the locomotives on the opposite end, due to the configuration of the tracks there – there was (and is) no direct south-to-west move possible there. These trains were always my favorite, because they would always bring a crazy variety of motive power. In the late 80s it was the NYSW operation of the D&H via the trustee, and then once CP took over it seemed like “anything goes” motive power wise. For a few years in the early CP, the decision was made to route these trains over the Sunbury Line instead of via Allentown. A small group of CP SD40-2s were equipped with the necessary Harmon Ultra Cab cab signal system needed to operate over cab signal territory south of Harrisburg, PA.

That’s not hyperbole. The effort started somewhere around 1997 or 1998, when Todd Clark over at Trainorders.com was doing some slide scanning as a service. I mailed him a batch of slides — actual physical mail, actual slides — and he scanned them and made the files available for download. In the late nineties this felt genuinely exciting, the first time I’d ever seen those Kodachrome frames on a monitor. The quality was what it was, but just having them in digital form seemed like a breakthrough at the time.

It was a start.

The Scanner Era

Around 2000, I got a Minolta Dimage Scan Dual II, which could handle five frames at a time in a strip carrier. For a dedicated slide scanner of that era, the image quality was decent — provided the slides were perfectly positioned and perfectly exposed, which was not always how I had shot them as a teenager. The workflow was also painfully slow in a way that’s hard to remember accurately now; five frames in the scanner, wait, check, repeat, for however many hours it took to work through a box of slides. I got through a bit of my collection this way, but I never got through all of it.

My wife got me an Epson Scan V700 around 2009, which was a meaningful upgrade. The V700 was a flatbed with a backlit film scanning attachment, and it could handle twenty slides at once in the provided holders — a genuine step up in throughput. The image quality was better, the scanning area was larger, and it handled negatives and other film formats too, which made it useful beyond just the Kodachrome. But the same limitation applied: exposure still had to be right in the original frame, and anything that was underexposed in 1991 was still underexposed when you scanned it in 2010.

In 2012 I bought a Pacific Image 5000, which was a genuine step up in capacity — a circular tray that could hold 100 slides at a time, automated transport, and fairly high resolution output for a consumer-grade dedicated scanner. I managed to scan most of what I considered the good ones out of my collection: frames that looked sharp when held up to the light, properly exposed, worth the time. Anything that looked dark or soft I skipped. The quality of what came out the other end was the best I’d gotten to that point, at least in terms of resolution.

But the same problems persisted. The selections I was making — keep this one, skip that one — were based entirely on how the slides looked on a light table with my eyes, which is not the same thing as what information is actually in the emulsion. And I was still fighting the same color temperature issues.

There was also a color problem that ran through all of these early attempts that I couldn’t quite solve. Flatbeds and dedicated film scanners of that era used cool fluorescent light sources for their illumination — fine for documents, not ideal for Kodachrome. When you see a Kodachrome slide on a proper projector, the incandescent lamp throws a warm light through the film, and the color rendering is exactly what Ektachrome and Kodachrome were designed for. That warmth doesn’t come through from a fluorescent backlight the same way. I could adjust color temperature in post, and I did, but there was always something a little off about the results. Anyone who’s sat in a dark room watching Kodachrome slides on a proper projector knows the look I mean — that rich, slightly warm rendering that makes the colors feel saturated and dimensional. Getting that onto a monitor consistently proved harder than I expected.

The Mirrorless Approach

I kept seeing threads on social media over the past year or two about photographers using modern mirrorless cameras to digitize film, with macro lenses and dedicated slide copy adapters. The basic idea is simple: rather than a scanner’s fluorescent lamp, you use a daylight-balanced LED light source behind the slide and photograph it with a high-resolution camera and a macro lens, then you’re working with a RAW file to make adjustments from there. The resolution ceiling of a modern 24 or 40+ megapixel sensor significantly exceeds what consumer flatbeds could deliver, and the workflow is faster once you’re set up.

Years ago, I tried a version of this by photographing projected slides with a Canon DSLR. I projected onto a traditional slide screen using my Minolta slide projector, and then took a photo with the DLSR of the projected image. While the color quality was spot on, there was a huge loss of resolution and I’m sure some mathematical issues with the perspective. I gave up after a few shots.

So, back to “today”: I rented a Canon R6 Mark II through LensRentals and paired it with a 100mm f/2.8 Macro lens. For the slide adapter, I picked up a Vello NSD-35, which is a purpose-built attachment that holds 35mm slides in front of the lens at the correct focus distance. I shot tethered to my MacBook Pro, which meant I could see each frame on a large monitor in real time — genuinely useful for verifying that I was focused precisely on the film grain rather than the mount, which is the thing that will ruin an otherwise good copy if you’re not careful.

I also used this project as a reason to finally get serious about Lightroom. I’d been using Bridge for years as my primary image management tool, which works fine but doesn’t give you a lot of creative headroom. Lightroom’s tethered capture, its develop module, and particularly its merge capabilities were all things I’d been meaning to learn. This turned out to be good timing.

The HDR Breakthrough

The thing that changed everything was bracketing.

With a traditional slide scanner, you get one exposure per frame, and whatever the scanner captures is what you have to work with. Underexposed frames stay underexposed. Deep shadows in an original slide either have detail or they don’t. There’s no recovery pass.

With the mirrorless approach, you can bracket — shoot the same slide at multiple exposure settings, just like you would for a high-contrast scene in the field. I was shooting each slide at three or more exposures, then using Lightroom’s auto-merge-to-HDR function to combine them into a single DNG file. Lightroom’s HDR merge is straightforward: select the bracketed exposures, invoke the merge, and it produces a high-dynamic-range digital negative that contains information from all of the source frames.

The results honestly surprised me. Slides that I had written off — frames that looked underexposed and borderline useless when I held them up to the light, or that had scanned poorly on the V700 — came back with real detail in the shadows, proper color in the highlights, and enough resolution to make worthwhile prints or enlargements.

That realization led me to change my approach entirely. With the Pacific Image and every scanner before it, I had been curating as I went — making judgments about which slides were worth scanning based on how they looked on a light table. I decided this time to scan every single slide in my first binder, covering 1991 through 1993, without prejudging anything. Frames I remembered as underexposed, frames that looked soft, frames I would have passed over without a second look using the old workflow — all of them went through the bracket-and-merge process. Almost every one of them came back usable. Some of them came back genuinely good. The HDR merge was pulling detail out of shadow areas that I didn’t know existed in the original emulsion, and the color rendering from the LED light source was holding up in a way the fluorescent-lit scanners never had.

I scanned around 250 slides over the course of a leisurely weekend. The workflow settled into a rhythm quickly: load a slide, shoot the bracket, advance to the next one, merge in Lightroom in batches. The tethered setup and the large monitor made the focus verification step much faster than I expected — when you can see the frame at near-full-screen on a laptop, it’s easy to tell immediately whether you’re on the film plane or not.

Sharing some Slides

The motivation for all of this was getting some of my archived content out in the open, especially among my “Conrail” friends. For example – Slide roll S09 from June 26, 1993 — a day spent chasing Conrail traffic in the Allentown area with my friend Rich Henning — had been scanned before, badly, or not scanned at all. The HDR approach brought those frames back in a way that the earlier scans hadn’t.

I’ve posted a companion piece on the blog that goes through that day chronologically — seventeen trains in about eight hours, including a Soo Line SD40-2 in white scheme on a D&H freight and a pair of ex-Pacific Rail Lines SD40-2 rebuilds at CP-Burn. If you want to see what this scanning workflow actually produces, that post has more examples.

The short version is that I’m now significantly more optimistic about the rest of my slide archive than I’ve been in a while. I’ve got 14 more binders to scan, containing a ton of slides that I’ve never scanned, or scanned once and set aside. The HDR mirrorless approach changes the calculus on what’s worth going back to. Now, just to find the time to do it…

Construction of the new “Lehigh Line in Bethlehem PA” began in earnest in early 2018 after the remodeling project was complete. For whatever reason, however, I haven’t really posted anything about that here. I’ve worked in fits and starts, often with months of other activities (a.k.a. “model railroad nothingness” in between. So, let’s start from the top and introduce you to the layout!

Background

Many model railroaders look to recreate childhood memories, and I’m no different: I was an avid Conrail fan ever since I was a little kid in the early 1980s, growing up in the Allentown, PA area. My mother would frequently take my brother and I down to Allentown Yard after church on Sundays, and as I got older, we’d ride our bikes there to take crappy Instamatic photos of whatever we could find. Into high school I got more serious about rail photography, started taking slides, and after I left for college I would document what I could when I visited home for breaks.

I decided to model the year 1994, specifically the “blah” of winter a.k.a. “the dead season” – with no snow on the ground. 1994 was the beginning of my senior year of high school, and represented the intersection of a few key traffic movements in Allentown, most notably being the last full year of nearby Bethlehem Steel in full operation.

Prototype Territory

The layout represents Conrail’s Lehigh Line beginning at the eastern end of Allentown Yard, continuing east through CP-JU (and connection with the former L&NE Cement Secondary), across the Lehigh River, and then east from there between the river and Bethlehem Steel’s sprawling mill. The Conrail Reading Line is represented mostly in staging, except at the diamond at-grade crossing at CP-Bethlehem. Total miles modeled isn’t going to be more than 3 or 4 miles, but this territory – especially then – was operationally dense, enough to support a sufficiently-entertaining variety of operations.

Pre-Conrail, the former Lehigh Valley line along the river was an east-west artery running from New Jersey to Buffalo. The Reading’s Philadelphia Branch terminated at Bethlehem, and the route continued across the LV and over the Lehigh River to allow Reading trains to access the Central of New Jersey’s Allentown Yard. A 1967 joint facilities agreement led to a connection between the CNJ and LV at Bethlehem as well. Starting with the 1972 retrenchment of the CNJ from PA, the Lehigh Valley took over Allentown Yard. The CNJ east of Allentown was later abandoned, and the LV west of Allentown (East Penn Jct.) largely abandoned or downgraded.

By 1994, the former LV main line was Conrail’s primary artery through the area, with some traffic flowing through Allentown Yard on the CNJ side, and some traffic staying on the Reading Line single track (“the hole”) out to Basin Street, where both routes re-converged onto former Reading territory west to Reading and from there to either Philadelphia or Harrisburg.

Traffic

The epicenter of the layout is the convergence of Conrail traffic on CP-Bethlehem, coming from 4
directions:

• Former LV single main from the west – trailer trains, coal & ore, some freight.
• Former CNJ main from Allentown Yard – freight and local
• Former LV main to the east – trailer trains, some freight & local traffic, and a bit of coal
• Former Reading Phila Branch to Saucon Yard/Bethlehem Steel – coal & ore trains

Overall, the big emphasis is on steel traffic: coal, coke, ore, and finished steel products. Manifest freight traffic will be prominent, and we will feature a few Trail-Van (TOFC) trains and a RoadRailer movement. Bethlehem Steel’s in-plant Class III shortline, the grandiosely-named Philadelphia, Bethlehem & New England, will be featured.

CP Rail had purchased the Delaware & Hudson in 1990 including the trackage rights south and east from Allentown. D&H had rights to serve Bethlehem Steel directly via Florence, D&H manifests & trailers from the north & west often ran around on the Layoff track between Bethlehem and Florence, and CP instituted trailer trains 265 & 266 to and from Oak Island (Newark, NJ).

The Layout

The layout is a triple-level shelf layout that primarily occupies a finished family room in an L configuration, with a helix at either end that connects the levels together. The bottom and middle levels are operational, and construction of the top level is limited thus far to only benchwork. Diagrams below illustrate the overall floorplan, with an emphasis on the primary viewing area in the family room. The Employee Timetable is available for download.

That’s it for now, we will share more soon!

Ever since first seeing the Conrail “bible” – the Conrail Freight Schedules binder in 1987 or so – I’d always been fascinated by the logic, the data, the planning, etc. I did a version of this presentation with Ed Kapucinski a couple years ago, that is viewable in the YouTube below.

The PDF of the slide deck is available here.

For scanned copies of the Conrail Freight Schedules, check out Multimodalways.

You should know that there is a Murphy’s Law of Model Railroading – starting a custom project guarantees that a manufacturer will come out with a factory model of your project. So naturally, of course, Jason Shron **had** to announce Rapido’s forthcoming B36-7’s over the summer, but, I took that as a good thing…

I’m modeling 1994 Conrail in Eastern PA, and easily need 4 of these things to cover the multiple Mail & TV trains that ran thru the area; not to mention that the B36’s also found themselves leading RoadRailer trains or as the cab-signal leader for numerous (and interesting) western road lashups that worked their way into the Keystone State. Needless to say, I’ve preordered two Rapido B36-7’s as well!

Anyway, these two units started out as Atlas B30-7A Phase II undecorated units. I replaced the cabs with 2-window cabs from Atlas. I had to swap out the trucks with GE Floating Bolster trucks, order later model handrails, and I installed Loksound decoders and LED’s (including working marker lights. That was a pain.) I relied on Hi-Tech Details for new turbo stacks, anticlimbers, and the various grills necessary to make this into a B36-7. Also, (and with many thanks to RGDave here for tips & reference photos), I made the necessary modifications to the conductor’s side battery box latches, hinges, and I added the early’90’s Conrail touch of battery box door locks at the top of each door. I also put brass horns up top, and bits of brass wire to represent the details on the cab roof. Let me think… I added the Shapeways Conrail GE marker lights, and had to do a lot of sanding and contouring of the nose to remove the old nose headlight pad and give it the more modern look. (I hate sanding). The nose handrails were modified to accommodate the anticlimber; the safety chain is on separate stanchions on these. I used a leftover pair of stanchions from a Conrail SD60M for this. Other details include hand-bent uncoupling levers for the front (a giant pain), MU connectors, hoses, spare coupler knuckle, radiator lift tabs (these will surely break off in the first operating session), etc.

Decals were from various Microscale sets, and I used Scalecoat II Conrail Blue in a spray can (love the stuff) and Scalecoat’s flat overcoat. Weathering was a combination of MIG pigments for the first layer, PanPastels, some acrylics (the burn marks on the 5040 roof) and more PanPastels.

The 5013 has some sort of crap on the inside of the cab windows. Could have been drillings from the pin vise, or some sort of contamination from not-quite-dry paint. Either way I’m going to have to live with that as it’s too tricky to try to disassemble all that. Also, I never fully understood the “gloss coat before decals” step until these 2 units… Meaning, I got some silvering on some of the older decals that I dealt with using small dabs of paint. Next time around, I’ll gloss coat first especially if I’m using older decals.

Other than those small items, I’m glad I took my time & I’m very happy how these came out. Looking forward to seeing Rapido’s in person, too!

• Hi Tech 6004 Anticlimber C30-7
• Hi Tech 6005 C30-7 Rear Intake Grills Late Model C30-7’s
• DA 1803 Sinclair Antenna
• Hi Tech 6001 GE Exhaust Stack Dash 7 thru Dash 8
• DW AH-190 Leslie RSL-3L-R3
• Shapeways CR GE Marker Lights
• DW CB-170 ATC Box
• DA 1024 Dual Headlight Pyle Vertical w/ Shield (front HL)
• DW PL-155 Snowplow 2nd Gen Hood Units
• Soundtraxx 810113 35mm x 16mm speaker – fits perfect (almost) 
• ESU 75413 decoder
• DW DL-229 ditch lights
• DA MU-1509 MU Receptacle Set
• DA LR1108 lift rings (radiator section-roof)
• DW RM-357 Rear View Mirror Set
• DW SK-196 Spare knuckle & bracket
• DA 1101 Lift Ring – for custom front uncoupling levers
• Miniatronics SMD LED 12-621-04 Red

It has been awhile since I’ve updated everyone… First, about the layout. The Aspinwall Belt was dismantled and put into storage in March, 2016 in preparation for waterproofing and finishing the basement. So far, that hasn’t happened yet. But, I was smart enough to leave my modeling work area intact, and so I’ve been using the time since then to continue working on weathering and detailing locomotives and rolling stock. With that in mind, I’ve (finally) completed a project that’s been on the shelf for over three years!

Back when I first considered re-entering HO Scale, with the obvious choice of modeling the Allentown & Bethlehem (PA) area in the mid-1990s, Lou Capwell immediately suggested that I needed to model the Outer Limits caboose. The “Outer Limits” caboose, Conrail 21248, was one of 113 N-21 class cabeese build in about 1984, by FGE at Alexandria, Va. These were the only cabooses bought new by Conrail, numbered 21201-21313. (CRHS) An employee stenciled the phrase “OUTER LIMITS” on one side of the caboose, which adds to its distinctive appearance.

I had a Walthers “International” caboose in CR paint, that seemed a close match, and so about 3 years ago I set it aside and started to work on it. My attention to detail and weathering skills have improved since then, which helps me feel less bad about not finishing this project sooner. I started out by disassembling the caboose, removing the windows, etc. I had to blank out a couple of windows, putty and sanding (blech) them smooth. I didn’t get too particular about replicating every detail of an N-21 caboose, but I wanted to give the impression of the N-21. For example, I relocated some of the brake appliances to approximate their locations as observed in prototype photos. I also fabricated an equipment/battery box on the stove-pipe side of the car, using strip styrene.

I found it easier to paint and weather the underframe of the car as a separate item from the caboose body itself. Ultimately, I repainted the caboose using Scalecoat II Conrail Blue spray can, and added decals from a variety of decal sets (Herald King, Champ, and of course Microscale) to get as close as I could to the prototype. After paint and basic decals, I set out to do more detailing: The stove pipe is a Bowser detail part from an N8 caboose, with a bit of brass wire added as a support. On this model, the end sills/ladder assemblies inset into the caboose ends on the body, leaving visible seams. I use a neat product called Mr. Dissolved Putty which is just what it sounds like – drop it in and it wicks into the cracks. It may take a couple of applications to get it right, but overall it was pretty effective at making the top/end area ready for marker lights and the reporting marks/number. Cal-Scale marker lights were used – I approximated by using two different types to match the abused marker lights from the mid-1990s photos I could find online.

Other finishing details included an equipment shelf on the end walls, little bits of styrene along the frame to more closely match the N21; and I made a fuel vent and fill with a bit of styrene, the threaded end of a tiny screw, brass wire, and bits of insulation. Finally, I added brake wheel chains, air hoses, and fashioned my own uncoupling rods out of brass wire.

For the graffiti and weathering – I used a variety of decals to get the OUTER LIMITS lettering to look somewhat correct. I modified some of the letters to get closer to the correct font. A combination of diluted acrylic rust brown and rust PanPastels were used for the roof and ends. I drybrushed white paint, and added white PanPastel to represent the overspray from the OUTER LIMITS stencils (look closely!). I masked and painted the “patched out window” using acrylics and rust PanPastel powder. A little grimy gray powders, rust Acrylic drips, and lots of Testors Dullcote to finish it off.

Overall, I’m pretty happy with the look and feel of this car – it will be right at home with my CR GP38-2 fleet, SW1500 switcher, and GP15-1 which will all rotate in assignments to the East Penn Drill on my planned new layout.

John Z had just burst into the room explaining that a train was stopped outside The Station Inn, with a rare SD45-2 being towed behind the newer, more typical Norfolk Southern locomotives that pass by here daily.

I reasoned that breakfast was more important.  I was still paying a price for the past two days’ festivities. Besides, while the SD45-2’s are rare, it’s not all that atypical for them to travel back-and-forth between northern New Jersey (where they spend most of their time today) and Norfolk Southern’s Conway Yard for routine maintenance.  I’ve seen them before, and figured I’d see them again – and besides, it’s not like these things are leading the train, right?

Only once before, to my recollection, did an SD45-2 lead a train over the Pittsburgh Division.  In 2015, Norfolk Southern decided to paint #1700 into the same Erie-Lackawanna livery that it wore when it was built in November, 1972.  On October 18, 2015, Norfolk Southern’s company photographer, Casey Thomason, arranged for the 1700 to lead the Conway-Allentown 10N mixed freight:

OK, so that WAS pretty awesome. That day, PittsburghMike and I were able to photograph the 10N at a couple of locations on the Conemaugh Line northeast of Pittsburgh.

Years ago, it was a bit easier to find and photograph the SD45-2 fleet in Pennsylvania. Originally built for long-haul trains between New York and Chicago, the thirteen E-L SD45-2’s were built with a distinctive 5000-gallon fuel tank, so that the units could make the entire trip without refueling. Also, these units were most powerful of the mainline diesels of their time – packing 3600 horsepower with a 20-cylinder 645E3 diesel engine. In 1976, Erie-Lackawanna became part of the Conrail merger, and the thirteen-unit fleet eventually wore Conrail’s signature blue.

By the 1990’s – when I made my first trips to the Altoona & Cresson (Pennsylvania) areas – the SD45-2’s were often found in helper service, pushing heavy coal, ore, and freight trains over the summit of the Alleghenies. By this time, very few 20-cylinder engines were still in service, some having been converted to a more-common 16-cylinder type. One summer day in 1997, I photographed a gaggle of them moving around the Cresson engine terminal:

Of course, CSXT and Norfolk Southern purchased and dismembered Conrail in 1999, dividing rail lines, freight cars, and locomotives between the two companies. Today, three of the SD45-2’s survive on CSX, and five on Norfolk Southern. Many of them continue working for “today’s” Conrail – the Shared Assets Area in Northern New Jersey. As noted earlier, the NS units in the fleet occasionally make a round-trip to Conway Yard for routine maintenance.

—

Breakfast was tasty – blueberry pancakes – but not as good as my favorites, the Sausage Gravy biscuit, or even better… the Pork Roll sandwich. As I finished the last of my coffee, I could see through the window that this train was still stopped at the signals at “MO”. A few guys excitedly ran inside to announce that the power was turned… and the SD45-2 was leading!

[box]This eastbound train, later revealed to be the same Conway-Allentown 10N, had experienced mechanical problems with the original leading locomotive.  While I munched my pancakes, the crew “spun” the power by turning it around (reports differ as to where – they could have “wyed” the power consist at Cresson or at the “loop” at Gallitzin), placing the NS 1703 into the lead position.[/box]

I hopped into the Pathfinder with a few of the guys, and motored quickly down to the Railfan Viewing Platform in Cresson.  Everyone’s pretty excited to see this unfold:

I swapped to my long telephoto lens and chatted with a father-and-son team that was videoing the event. (It’s always great to see the hobby passed down to the younger generation!) After a few minutes, a double-stack train overtook the 10N on track 2, and a bit later, the 10N started east:

We decided that we needed to try to get to AR Tower. With some creative driving, we barely made it. I made quick use of a hole in the fence to capture the 1703 starting downgrade, past the disused tower at AR:

We watched and listened as the three “ponies” started down the steep grade. Nick egged us on to try to get to Benny. We thought that to be crazy, but a good idea nonetheless. I was grateful for the 4×4 and high ground clearance of the Pathfinder as we went down the cemetery access road, and ambled our way back to the trail. [Apologies here to the same father-and-son video team seen earlier, as we enveloped them in road dust, not realizing they’d be there!] We could hear the dynamic brakes whine on the three locomotives as we lept out of the car and ran like crazy for the overlook. Panting and out-of-breath, we made it to the “spot” as a westbound started its way uphill on Track Three. The 10N slowly made its way down Track One, putting on an excellent show for us!

With memory cards full, and emotions running high, we decided we’d return to The Station Inn and get on with the rest of our day – which included seeing CSX’s “Johnstown Jet” local make a rare Sunday move, and seeing a few trains on the NS main in Johnstown as well. All in all, a great day – and I didn’t have to skip breakfast, either.

Christmas music is playing softly, I’m sipping a beer, and the kids are asleep (well, the baby is resisting, but she’ll be asleep soon enough…) so I will take a moment to update on things on the Aspinwall Belt.  We had a great operating session in late September, with a group of newer folks.  We ran with conductor/engineer pairs and ran at a 2:1 fast clock in order to keep things fun – it’s worth spending extra time with the new operators, as they benefit from learning some of the knowledge that I take for granted – let’s face it, normal people don’t know all of the esoteric Conrail stuff that some of us do!

We “only” ran through half of the scheduled trains that evening, so an upcoming session will be held to run the rest of them, again at a slower pace to help everyone “catch on” and provide more time for conversation.

Regular readers will know that my Conrail modeling is constantly inspired by Dave Abeles’s excellent Onondaga Cutoff – and this blog post is no exception.  Running trains at Dave’s layout got me inspired to start upgrading my locomotives with sound decoders, and he also steered me to the eBay listing for the Conrail 6150 featured in this post.

In September, I bought the CR 6150, and installed a SoundTraxx AT-1000 “GE FLD-16 Modern” decoder, along with a 30mm 8-ohm speaker designed to fit inside the Atlas speaker mount.  Thinking this was a quick project, I did a bit of weathering, and put it back together, and found a couple problems: First, one of the trucks’ wiring disconnected; next, the factory lighting was blue-ish and dim.  This was No Good.

So, back we went, sanding and soldering the electrical feeders from the trucks, and installing upgraded LED lighting.  The Atlas units have LED’s in all the right places (unless I really wanted independent numberboard lighting), and so I dropped Miniatronics 3mm Yelo-Glo LED’s in place of the factory; and for the ditchlights I used Micro Krystal Kleer to glue in a pair of Richmond Controls surface mount LED’s, and I soldered those to the existing PC board pads Atlas provides to power the ditch lights.

A couple notes about that – the magnet wire on the Richmond Controls lights is insulated by a plastic “dip” – I use a lighter to burn it off and then a bit of sanding to polish.  Don’t get too close with the lighter, otherwise you’ll melt the wire, too!  Also, the soldering iron will melt the plastic around the shell (when soldering the feeders to the copper “tabs”) – watch yourself, and use a drop of CA to secure the tab when done.

The speaker is caulked in with clear silicone caulk – a good seal helps the acoustics of the speaker enclosure.

The 6150 would have been pretty new in 1994, so I went with heavy weathering on the roof but minimal along the sides of the unit.

The 6150 was a hit at the September operating session, and so I decided to go ahead and do the 703 during October.  LMS 703 was built new by GE in the spring of 1994, for a joint Conrail-GE venture called “Locomotive Management Services” – and it needed to look factory fresh.  This was a used unit I picked up from a buddy a couple years ago.  A few interesting details make this a bit different – the nose grab irons are in 4 pieces instead of 2, so I fabricated new ones out of brass wire.  Also, for some reason, online photos show the 703 as having a rather tall turbo stack.  So I replaced the Atlas part with a Hi-Tech Details Dash 8 stack, which I shimmed up a bit with scrap styrene.

Most of the wiring was done fairly quickly, but one of the LED’s failed (cold solder joint, perhaps) so that was annoying; but this unit also had a broken handrail assembly on the conductors side – so I ordered a new one, then found after finding the time to work on it that it was the wrong part.  More waiting, for the correct replacement part, but a big kudos to Atlas for making good on this, by the way – they included some other small parts that I needed as well!

As far as weathering – I tried something new (for me) – I applied Testors Glosscote to the blue paint, and Dullcote to the black.  This required a lot of masking, more than I’d typically do for an engine I’m not even painting.  I used PanPastels and some enamel paint for the weathering – my typical weathering routine is to use thinned black for the grilles, then apply Pastels going from light to dark, then to clear coat and repeat if needed.

On this unit, I brush-weathered the turbo stack to make it look pretty burned up (they don’t call these GE’s “toasters” for nothing) and was a bit lighter on the exhaust.  I applied a light spray of dullcoat on the top of the unit to protect the pastels.  The look of the anti-reflective black nose works pretty well compared to glossy Conrail blue.

The one change I haven’t made to these units (haven’t quite figured out how) is that Atlas uses the wrong font on the numberboards.  A minor detail, to be sure, but I have a couple sheets of the correct on Microscale’s various Conrail decal sets.  But the Atlas numberboards are a bit short to fit the decals (or perhaps the decals are a bit tall).  As I really like making these units look as close to prototype as possible, this detail is going to bug me.  (Less so are the Atlas oversize windshield wipers.  $10 a pop for wipers means these will be be good enough for quite some time.)

So, while I didn’t get the 703 finished until the end of November, I now have a pair of Conrail GE’s that I’m happy with – and they gurgle away on the ready track both fully-sound-equipped.  My 3rd Atlas GE, 6061, is on the shop shelf awaiting this treatment sometime in 2016.  In the meantime, I have a few other items on the workbench that I’ll be sharing in the coming months!