Help With Erie Refrigerator Cars

While doing research about suitable reefers for the layout era - 1952 - I came upon several refrigerator cars decorated in dubious paint scheme. Most of them were made by manufacturers know for they "creativity". However, a model among many caught my attention. It was the Accurail #4813 kit, which was re-released with an improved paint scheme (mainly the Erie Diamond logo) as kit #4813-1.

URTX-leased Erie reefer (credit: Accurail)

While Accurail is now for their nice quality kits, we all know they apply the paint scheme on their models if it is fairly similar to a real prototype. This isn't criticism toward Accurail because most major manufacturers do that to make their models more appealing to a large crowd. In fact, Accurail generally try to find prototypes that are close enough to not be foobies and generally, their paint schemes are accurate and based on real-life examples.

This is interesting since from that, I can guess the URTX-Erie paint scheme isn't a fabrication, but probably a real thing. However, finding the real prototype cars specification is quite hard. As far as I can tell, the Accurail car represent a reefer built circa 1936. The big question is did Erie still leased reefers in the early 1950s. If yes, were the cars repainted during that period of time. My guess is yes. I have many reason to believe an updated paint scheme wouldn't have mentionned "Ventilated Refrigerator" as this expression was no longer used by railroads by the 50s. Also, the way the dimension data is spelled is quite archaic.

So basically, my questions are:

Did Erie leased or owned reefers in the early 1950s?

If yes, what type of reefer was it (wood, steel, etc.)? Which prototype?

What was the paint scheme or did Accurail's paint scheme survived until then?

Any reference to specific ORER information would be valuable.

Modelling Early Erie Steel Boxcars

An interesting part of Erie's boxcar fleet was their early steel designs which roamed the rails for many decades. They could be tricky to model, requiring extensive kitbashing, but fortunately, within acceptable means we can build them without loosing insanity. The funny thing is decal companies have made many Erie boxcar decals or dry transfer, but focussing solely on the unavailable oddballs.

Fortunately, a few years ago, Dan Hanley ran a 4-part article in Model Railroad Hobbyist (April 2013) to describe how one could recreate a few of the iconic prototypes. I'm certainly not interesting in going to such lenght as casting my own resin cars, but his articles provide enough data to understand the fleet, its history and particularities.

Among Erie's particular choices were Chicago-Cleveland radial roofs, Buckeye ends and Viking roods. I'm certainly not an Erie fan by choice, but it is essential to capture this diversity to better represent the typical Erie freight consist. The most interesting boxcars are all fall into the 75000-76999 series. All steel, 40ft long, with various heights ranging from 8'-8" to 9'-4", they had character.

While I'm completely open to the idea of scratchbuilding Buckeye ends and alter significantly available models, a few readily available models are already on the market and can serve our goal efficiently.

Atlas ARA 1932 Boxcar (credit: Atlas Model Railroad)

The first model is Atlas ARA 1932 boxcar design. This model sports all the correct details for an Erie prototype and capture the flavour of this railroad. Modification aren't required, except if you buy the version with the yellow Erie Diamond logo, which should have been white on these cars. The paint scheme could also be altered to represent the modern Large Diamong logo instead. However, don't buy any Atlas ARA 1932 since they all vary in details according to specific roads. Stick with the ones clearly identified as Erie.

TLT CPR Minibox (credit: True Line Trains)

The second model is quite interesting since it brings us back to 1929, in the all steel boxcar infancy. Back then, very few railroads embraced the new concept put forward by the proposed ARA 1923 40ft boxcar design. Among the pioneers, Canadian Pacific built its famous home variation called the "Minibox". These cars had radial roof and early Dreadnaught ends. However, the sill had tabs and side steel plates were assembled in such a way it eliminated the ARA 1923 flaw of trapping water and rusting.

Fortunately, Erie had almost similar cars equipped with the same roof and ends. However, Erie built them with the infamous straight sill. It thus could be possible to use True Line Trains own's Minibox  plastic car as a starting point and bashing the straight sill. Also, some minor alterations could improve the likeless of Erie's Dreadnaught ends which were marginally different. While not 100%, this model would be both almost pinpoint on overall dimensions and details.

Red Caboose PRR X29 with Dreadnaught ends (credit: Intermountain Railway)

The third option would be to use the old Trains Miniature (Walthers) 40ft steel boxcar with Dreadnaught ends. This is the famoust PRR X29 design. Starting with a Red Caboose kit could also be a better investment in term of quality and prototypicalness. Only the car with Dreadnaught ends would be suitable. Bashing the radial roof would be the biggest challenge, but the straight sill would be "correct". Think of this option as another way to replicate what could be done with the CPR Minibox. It's a matter of choosing your fight! A radial roof could be made in styrene, then cast in resin for ease of replication. I believe it is somewhat easier to alter a Minibox tabbed sill than tackle a fully kitbashed radial roof. However, it ain't that much hard to do with sheet styrene.

To be noted, another easy bash is using Intermountain AAR 1937 (IH 10') boxcar and replacing the roof with a Viking one. Minimal alterations would make it an excellent car to model without efforts.

As you can see, it is quite feasible to build a decent Erie early steel boxcars without reinventing the wheel. Depending your interested in Erie, feel free to you to follow Dan Hanley and go full kitbashing. However, it is nice to see a large array of interesting cars can be modelled with relative ease.

Unseen Challenges

People following this blog must think I'm just a crazy guy building cars and never operating the layout... Well... They are indeed right! But hey, model railroading has multiple facettes and car building is probably one I enjoy the most.

That said, building a decently prototypical fleet isn't always a walk in the park. Over the last few weeks, while it was expected some cars would be straightforward, I end up facing several road blocks. I could have thought kitbashing would be hard, but at the end is was only time consuming. However, the worst had to come...

Wrong ERIE decals and reefers without ones...

Among them, accuracy of decal sheets was extremely problematic. By example, I ordered a few Tichy cars with decals. At first, I thought to myself: build the car per instruction and apply the relevant decals. So wrong I was. First, I discovered the USRA rebuilt decals for NYC cars were slightly wrong here and there. Not too much but I had to tweak things a little bit. However, the Pennsylvania decals were as bad as can be. The artwork was smudged and the decal film brittle. I could do nothing with them and thus the car is laying on my desk until I get replacement decals.

Slightly incorrect P&LE lettering but decent PRR and LIRR ones

I also bought a Kadee PS-1 ARR boxcar. Erie did have a few ones and since the car was unlettered and painted brown, it was an easy candidate for lettering. One again, I found out the Tichy decals were wrong. In particular, the font used on the Erie Diamond logo was so off I'm still scratching my head. Fortunately, I have left over from CDS Lettering that will do the job perfectly. But now, many more steps have been added to the process, making it slower again. Some other built date will have to be replaced too.

Remark I'm not complaining. This is part of modelling, particularly when you try to be more prototypical. And while I'm not annoyed, I now better understand older modellers complaining about decals and details. If you care about them, you quickly end up stumbling here and there. The only to do is to sit back, assess the situation and take action accordingly.

GARX 37ft Wood Reefers - Part 2

Among the 20-something cars on my benchwork, the first two are now painted and lettered. While they are not pieces of art, it's quite interesting to see how far you can bring such cheap models. It should also be noted the very spartan GARX revised paint scheme of the late 1940s also contrast with the garish billboard cars generally loved by the model railroading crowd. It helps to bring a sense of reality to the layout.

While I'm quite happy with the detailing, paint job and lettering (by Clover House), I must assess the result to learn from my mistakes.

If I had to redo these cars again, I would address two issues that felt minor at first but have a substantial impact on the credibility of the finished models.

The first thing would be to make the fascia board a little bit narrower. Thought I made sure to measure the fascia on a quality picture with little distorsion, it seems it is to wide. Hard to tell why, by when I painted the roof, it became quickly apparent. A comparison with a Rapido GARX meat ree shell confirmed that. To be honest, I believe the fascia were originally quite large. It was common on old wood car. But visually speaking, it doesn't look right.

A second issue is the roof pitch. I kept the Varney/Life-Like steep pitch since it was the normal thing to do. However, at the end of the day, the roof looks a little bit weird, particularly when compared to similar wood reefer models. I'm not sure if Varney was wrong, but I've seen many older model that got the roof pitch inexact. I'm pretty sure it wasn't. I feel I shouldn't have been lazy and could have fixed that earlier. Oh well, lesson learned!

That aside, the models now look decent. Not great, but OK. With some fading and weathering later, they should blend into the crowd. And while it took me countless hours to get that result, it isn't wasted. I've learned a lot and improved my skills, which was useful when I had to work on other cars. Better, scratchbuilding several parts and trying to figure out the underframe helped me to better understand how real cars are built. This is something you can't learn by simply aping kit instructions.

Kit Manufacturers Comparison or Where's The Fun?

Over the last few weeks, I’ve been working on several freight car kits to build up Harlem Station roster. I had to deal with kits from various manufacturers and thought it would be neat to share my impressions of each ones. It must be noted I’ve built models from these manufacturers for a long time. However, it was the first time I built some much detailed cars in such a short amount of time and with a focus on small details (brake gear, grabirons, roofwalk, etc.).

Life-Like: not a contest winner, but definitely an enjoyable process.

My endeavour started with a pair of Life-Like/Varney refrigerator cars. While absolutely crude, the sturdy and thick plastic used back then proved to be a blessing during the process. It is a sound and solid foundation upon which to one can add details to its heart content. Once again, Tichy details proved to be great to work with while Intermountain brittle plastic was less fun. Accurail AB brake assemblies saved also a lot of time and money. They can be modified too, though I’m not a fan of delrin plastic since I had to use epoxy glue to keep them permanently attached to the car. As much as possible, I prefer to use plastic that can be cemented with solvent. I also used Accurail Bettendorf trucks which are finely casted and affordable. The fine lettering moulded on is a plus to me. I certainly wish Accurail made more trucks. Finally, Kadee brake wheel were a bliss to install. Probably among the best on the market and ones must admit they are incredibly sturdy while having the finest cross section you can expect in HO scale. Overall, kitbashing Life-Like and Varney old train set took a lot of time, but it was a fun process and never tiresome or frustrating. You could almost consider it scratchbuilding since only the basic shell was kept and could have been easily built from scratch with less effort.

The next model I worked on were four Life-Like Proto 2000 Greenville Mill Gondolas. The last time I built this specific car, I was in high school back in 1999 and used a dull X-acto blade and CA glue. I recall the process to have been tedious. This time, sharp blades, solvent cement and tweezers made it much easier. However, the notoriously flimsy stirrups and annoying brake gear are certainly irritating aspects of P2K cars. I decided to order photo etched replacements from Pierre Oliver. However, I must admit the P2K cars are great to build and prototypical to booth. When you know their weaknesses, you can work things around so they are sturdier.

C&BT: Lots of remedial work required and far to be top notch

Then, I decided to tackle a pair of C&BT Santa Fe ice steel reefer. Tony Thompson and Andy Sperandeo once wrote about these cars back in the early 1990s. I thought Thompson was quite harsh when he concluded on his blog they weren’t worth the trouble. But after having spent hours on them, I must admit he was right. The plastic C&BT used is annoying at best. It is brittle and for this reason, many parts had to be oversized. This limitation also made many parts to look coarse approximation of the real thing, even from Average Joe’s standards. I ended up partially trashing the B-end details, the brake gear, ladders, stirrups, hinges, etc… I think you understand the pattern, only the shell was salvageable. But even then, dimensional errors plague the model. The roof doesn’t sit at the proper height and has oversized overhangs. The roofwalk is a joke, many platform brackets not molded.  I also had to alter the roof platform so they be prototypically correct (SFRD platform had long slots so the hatch latches could be dropped into it and not be a hazard for workers).

C&BT: This is an unintentional parody of a brake apparatus

At the end of the day, you end up with a model you need to extensively rebuild, alter and modify. It seems C&BT worked with flawed and insufficient data which made the model an approximation of the car. I can understand the limitation of the early 1990s, but it is quite frustrating to trash almost every details and waste time rectifying another’s mistakes only to get a passable result. When I buy a kit, I generally understand I can build it straightforwardly without having to extensively rebuild it (except if it is the stated goal from the start). I certainly don’t recommend this car, particularly when a vastly superior model is now available from Intermountain.

Intermountain: Flimsy stirrups, but great looking car

I then built two Intermountain PFE R-40-23 reefers. They did a great job capturing the prototype and the building process is quite straigtforward.  However, I’m not a fan of the plastic type favoured by Intermountain. It is quite flimsy for small parts and can easily break during handling. They plastic stirrups are a lost cause. They broke the moment I installed them and I can assure you I used paranoiac care to make sure they would survive. Thus, I ordered photo-etched replacement parts from Pierre Oliver too! But that said, Intermountain cars are generally well-designed and easy to assemble. The instructions are easy to follow too, which is a necessity when dealing with hundreds of parts.

Athearn: With some love, you can bring forward the potential

The Intermountain models then prompted me to update four Athearn 40ft ice steel reefer. Knowing the car was based on the PFE R-40-23 design, I ordered several Intermountain detail parts. I also extensively kitbashed the roof and hatches so they would be closer to the prototype. Using the Intermountain cars as a reference made it quite easier to do. While it required a lot of work, it was a great and rewarding challenge. I don’t regret any minute working on these cars and think they will be a great addition to my layout when completed.

Tichy: Top notch and extremely affordable

Finally, I continued my endeavour by tackling a bunch of Tichy USRA boxcars (single sheathed and rebuilt). Must I say Tichy product have the detail quality of resin kits and the ease of assembly of styrene kits for a fraction of the cost (even competing with Accurail). It is the best of both worlds. The level of detail is incredible and parts are generally free of flashes. Assembly is straight forward, instructions are generally well written and they use a sturdy type of plastic that is both durable and easy to work with.

That wraps up my impression with working with various types of kits available on the market. Meanwhile, I’ll soon have to work with Rapido kits (which are devoid of any instruction), Westerfield and Funaro & Carmerlengo resin kits. While I’m not new to resin kits – having built my share of Japanese garage kit and a Sylvan CN caboose back in my college days – I’m curious to see how I’ll find the experience.

GARX 37ft Wood Reefers

Some of our decisions seem stupid when rated in term of efficiency. In this case, rebuilding Varney and Life-Like wood reefers as GARX refrigerator cars may sound foolish, and to some extent it is. However, simply judging a project value – particularly in the case of a hobby – from this perspective is brush over many layers of relevance.

Styrene, paper and brass...

The project started when I found out the general proportions of these models were a good match for 37ft meat reefers - a common lenght for decades in the packing industry - an idea found on Tony Thompson's blog. I had two cars, plenty of trucks and freight car parts that could be used to bring them to life. Dry transfer lettering is also a cheap and convenient commodity. Thus, my mind sought to find a way to make decent models out of them. My guess was that following prototype pictures and Rapido’s own meat reefer official artwork would yield acceptable results. Little did I know that I would venture in quite a quest.

Individually carved wood planks with bolts and wood grain.

Certainly, I can affirm a lot of work was involved. At the end of the day, only the car sides and ends were kept. Grabirons, stirrups, roof and hatches, brake gear and underframe were removed altogether, including door hinges. Someone could argue I would have been better to scratchbuild the damn thing from scratch and save myself a lot of tedious and unrewarding work. That’s probably true… However, I had to use several different techniques and skills to get the results I wanted. This kind of learning will be useful on other more worthy models. Practice makes perfect they say, and you can’t skip that step – even if it may look petty – if you wish to be a better craftman.

I didn't count the exact time it took to upgrade the models, but it certainly occupied me with quality leisure time for a few weeks. At each day, I looked at the models and decided which little detail I wanted to build or improve, taking my time to do the job right.

Correct underframe and brake detail, improved door hinges

This is probably becoming a leitmotiv here on Harlem Station, but I certainly believe working on a small and manageable prototype gives you the opportunity to progressively improve your skills and redo things better as required. On a large layout or project, the imperative of simply trying to complete the project makes it impossible to go back and redo or improve things. The show must go on…

If my explanation isn’t clear enough, think about it in terms of digital video encoding. If you only use a one-pass algorithm, you’ll get a somewhat acceptable result. But as you make a second and a third pass, fluidity and sharpness greatly improve. Depending of your goal, it can or cannot fit your aspiration. But if you want to learn skill and reach excellence in your modelling, this could be a good way.

In my case, I’ll continue to upgrade older freight cars, as they are an excellent way to improve oneself and spend some quality leisure time.

Reefer Colors in the Acrylic Era

Finding the right color for freight car can be a real pain, and I'm not talking about the intricacies about boxcar red.

One of the worst color for me so far as been to find a correct orange reefer in acrylics. I've looked at many conversion charts and I'm certainly full of doubts, particularly for FGE and SFRD refrigerators cars of my era (late 40s-early 50s). While I've seen paint labelled as Orange Reefer, many modellers aren't convinced of their accuracy.

Actual pictures show us FGE reefers were kind of light orange closer to a dark harvest yellow than pumpkin orange. The same could be said about SFRD cars, which always have a kind of yellowish color though it seems FGE and SFRF had their own specific paint color.

Among suggested color, Microscale suggest Gunze Sangyo #109 Character Yellow as a substitute for Floquil PFE LightOrange. They even suggest Humbrol #18 Orange, which is quite a saturated orange shade a little bit over the top for SFRD reefers.

While many manufacturers now offer "Reefer Orange", I'm not that much inclined in testing a few of them only to be disappointed. Not all reefers were born equal and I'm pretty sure painting these cars is much more than just spraying some random stuff out of the bottle.

Thus, if anybody have suggestion on acrylic paints that can be airbrushed and suitable for FGE and SFRD car, feel free to comment. In the past, I wouldn't have minded and mixed my own custom color. However, as I'm building several cars at different moment, I'm not eager to have huge discrepancies among the fleet.

Edit: Reading again articles by Tony Thompson and others, it seems SP Daylight Orange would be a close match for PFE Light Orange reefers. Only need to find a correct color for the SFRD cars.

Kitbashing Erie 37500 Series Gondola - Part 2

Everyday we learn a new thing – at least we should – and modelling is probably a good teacher in that regard. While working on my pair of Erie 37500 series gondola I got a lesson where I expected it the less.

As a starting point, I’m using the old Roundhouse 40ft gondola. Some will object it’s a crude model by today’s standards but details are crisp and the number of panel fit the Erie prototype I’m trying to replicate. Certainly, there are some discrepancies in length, but this high side gondola fits decently the bill. My approach is to make it more credible by installing wire iron grabs, A-line metal stirrups, improving the braking gear, replacing the trucks, thinning down some section profiles and redetailing the bare interior. In short, it’s a typical rolling stock improvement project that should take a few evenings to complete at a leisure pace.

The problem arose when I made a new steel deck with hand-punched rivet details. I was quite proud of my work and thus decided to glue it on the bottom of the gondola. My first idea was to simply secure the bottom with a few drops of CA glue, however, the rivet punch process slightly warped the thin styrene sheet and I was afraid this could impeded the bond. Thus, I decided to simply brush a generous amount of solvent-based glue on the gondola and cement permanently the new deck. At this point, all was fine.

However, on the next evening, I found out the car sides had badly warped toward the interior and by a large margin. Using my hand, I carefully pressed the warped parts outward by applying a certain amount of force. I thought the sides were affected by the glue bond at the perimeter of the new deck… But imagine my surprise when the next day, the car sides were once again badly curving inside. Not only that, but now the deck itself was sagging the middle. It didn’t take long to find out the problem. The gondola bottom had warped significantly due to the solvent used to cement the new deck. When I installed the metal underframe, it no longer sat on the plastic shell but rocked on the curved bottom impeding the trucks movements. Big problem!!!

What happened was a simple and well-known phenomenon. When laminating two different materials, they have a tendency to warp when both have different characteristics. Generally, the process takes some time and is gradual. In my case, the generous use of solvent softened the plastic to the point that when materials hardened and cured in the following hours, their different rates of contraction created a disaster. The result was predictable to some extent, but little did I know a 1 mm bulge would manifest itself in such a spectacular way.

At this point, I was able to straighten the gondola sides. However, the only practical solution to repair the warped bottom - which should be better described as a belly - was to shave down the bulge with a chisel. It surprisingly went well, but it certainly didn’t address how badly warped the steel deck look. If it was a wooden gondola, I wouldn’t care since these cars were generally badly warped and sagging in old pictures. Unfortunately, steel cars generally keep their shape better.

From there, I had two options: try to fix up the deformation or simply cutoff the car bottom and replace it with a brand new one. In both case, it was a question of releasing the built up stress in the plastic. While at first I was leaning toward a full reconstruction, my long experience with resin figure garage kits kicked in. I plugged the hairdryer, set it to maximum heat, and applied evenly the hot hair on the car. I made sure to soften the plastic until it was malleable but was carafully enough so details start to molten. I didn't keep track of time, but it was less than a minute. When the car was soft enough to be reshape, I put it bottom up on a flat surface and gently pressed down the belly uniformally. When the shape was restored, I kept pressure until the plastic hardened again. As expected, the gondola sides straighened and the bulge disappeared once for all.

24 hours later, both cars were still keeping their shape thus I resumed the redetailing process. I was also a good occasion to address will laziness toward molded on ladders. At first, I'd planned to remove them, then decided to keep the ladder and save myself a tedious task. But after all the effort put on the cars, it was silly to still keep clunky ladders. I heard the voice of reason, took my chisel blade and scrapped off the unwanted details.

While I was able to repair my horrible mistake, I've learned my lesson. Next time I face a similar challenge, I will superficially glue the new deck with non-solvent based cement to eliminate altogether the risk of warping. I’m certainly not proud of myself for what initially seemed to be a well done job. Once again the old rule applies: use as little glue as needed. 

For some reason, we have a tendency to believe styrene is a stable material compared to wood or cardboard, but this is far to be true. Styrene (and other plastics) is surprisingly  vulnerable to warping and I've seen more than my share of projects going awry in a matter of days or weeks.

But enough about mistakes... The rest of the project was relatively straight forward, involving a new Accurail gondola brake apparatus, retainer valve and details from Tichy and a Kadee brakewheel. I went as far as adding the fulcrum and installing a bracket for a cut lever. Also, riveted steel straps were added inside the wall of the car as it did exist on some gondola. It was to reinforce the impression this is a real steel car and not a one piece injection-molded shell.

The goal is simple, these cars, even if they are prototypically accurate should at least have the same level of detail as a Proto 2000 or an Intermountain car. My reasoning being discrepancies in details are much more noticeable than overall accuracy. The model should look like a real car and have the basic features. From that point on, sky is the limit.

Kitbashing Erie 37500 Series Gondola

An so the endless waltz of car roster improvements starts with two old gondola to be bashed as close as possible to Erie prototypes. The idea is not to make exact replica, but to put enough efforts so the models can be credible. I've bought sets of dry transfer from Clover House suitable for Erie 37500 series. These cars were initially built as drop bottom gondolas until they were later rebuilt with fixed floor. While I've been lucky enough to find as built picture of Erie cars, my search for rebuilt car wasn't rewarded by any significant results. For more information, Ronald Dukarm wrote a short but informative story about these cars on railfan.net.

For this reason, I ask any reader that can point me decent sources of information about these particular cars to contact me. Many books and publications about Erie cars have been released over the years, but sometimes, it can be quite hard to find out particular prototypes when you have very little information about them except length, road numbers and sparse details.

In fact, I'll have to build (or rebuild) many boxcars and gondolas following Erie practices. Thus, I'm quite open to acquire relevant books/magazines on these subjects. However, I'm not that much familiar with Erie and would prefer to purchase books based on educated suggestions. Factual data over nostalgic books would be better. For the boxcars, Don Hanley's excellent series of articles in Model Railroad Hobbyist back in 2013 is an excellent starting point. I certainly would love to have better information on gondolas.

JMRI Operations & Car Roster

I finally succeeded in tweaking JMRI Operation until I got what I wanted. Until now, I got LCL cars playing ping pong all over the layout and unable to go back to staging. As expected, it quickly clogged layout spots. The answer what rather simple. Since I wanted to receive LCL but also send LCL, I originally created a single commodity called "LCL". Sounds logical... Many spurs were scheduled to receive and ship such commodity. But as expected, when a car was loaded it was routed to the nearest available spot, thus on the layout instead of the far away destination in the staging area. It was a disaster in the making, the endless waltz of cars shuffling around the place forever started.

To address this issue, I simply created a LCL-IN load for inbound traffic and LCL-OUT load for outbound stuff. The car swapping ended since going to staging was the most logical (and short) way to handle the loads and empties. A human brain is able to tell LCL is going in and going out, but that must be explained flat out to the software

Another problem emerged as a result of saturating the layout with incoming cars. This was due to the inbound traffic being set using the Walthers carfloat maximum capacity which is well over sixteen 40ft cars. That's a lot of traffic for a small crowded terminal and soon enough, the place overflowed with cars. JMRI - even if it was against it's own scheduling rules - had to route cars back to staging. Since the staging spurs weren't designed to accept such loads, the cars stayed stranded forever on the carfloat (acting as a yard) searching forever for a spot that didn't even exist and slowly but surely reducing the carfloat capacity to bring new cars in and get others out. Just like a Tetris game going wrong, the layout was destined to be game over in a matter of time.

However, the solution was once again very simple. I found out Harlem Station was served - most of the time - by platform carfloats. These special carfloats have a platform in place of the central track. This implement is generally used to unload cars directly on the platform which is linked by a sliding ramp to a pier or a warehouse (look at "Pier Platform" in the glossary). This is particularly useful when cars are bound to a non rail-served marine terminal. Why Erie did use the platform carfloats to serve Harlem Station is a mystery, but I guess they had them and it was handy at that time.

But that type of carfloat plays to my advantage: the number of incoming car is reduced to 12 cars maximum. Given most pictures and statistics show an average of 6 to 8 car was typical, I changed JMRI so the longest train is made out of 12 cars. Also, I made sure to set a random value in such a way the number of incoming cars varies and is generally around 8 cars.

Finally, I better assigned customers' and storage spurs following what I could understand from old pictures and historic sources. It certainly isn't 100% accurate, but it is a fair enough approximation of the real thing. Moreover, with less but better routed cars, I got rid of the loaded cars stranded on the carfloat. Now, every car is routed correctly to its scheduled destination and goes back to staging as it should be.

Given this is my first serious effort using JMRI, I'm quite happy to have reached a decent level of proficiency in less than 4 days. Certainly, this is enough for a small switching layout, but I'm well aware I could go further with it, taking in account the days of the week and cargo-specific conditions. This will be done in due time though I already made some adjustments that brings life and unexpected events.

On another hand, JMRI made me understand how much a small car fleet (about 60 cars) can get repetitive with a 55 cars layout capacity. As I said in my previous post, this is getting quite boring fast and it disrupts completely the suspension of disbelief. For this reason, I tracked down parts and decals to complete or kitbash cars in my collection that had no purpose. I'll probably be able to add about 11 new reefers and I still have decals to build more Erie boxcars. In itself it doesn't bother me since bashing and improving rolling stock is always a nice way to spend time in a creative way.

Until then, the next step is to clean out the rails and wheels for flawless operation and experiment first hand with the new switchlist. Jérôme is already working out the mechanical aspect of the fleet including trucks, wheels and coupler height. It seems I did a poor job protecting the wheelsets from over spray and it quickly gunked the rail leading to poor electrical pickup. Lesson learned!

Setting Up JMRI & New Discoveries

Work on Harlem Station recently resumed. I’m mainly motivated in finally implementing JMRI Operations on this layout to create automated switch lists. While the learning curve is quite steep, the software itself has a lot of potential, particularly for such a terminal.

Setting up a credible operating plan isn’t as easy as one could think even if this layout is rather small. It is quite a busy place and space is at premium. It quickly becomes a real issue when you run a few passes with JMRI and see how the car moves on the property.

As a matter of fact, most of you know I set the layout date to 1952. It is mainly based on information found in an article about Harlem Station that ran in Erie Railroad Magazine that specific year. It provides a list of customers and commodities handled by the terminal, which is premium information to start building a decent switch list.

But nothing is ever simple or straight forward with Harlem Station and I’m still struggling with the track plan even if I’m studying this terminal since 2010 (it’s been 7 years!). Erie Railroad Magazine mention there is 10 sidings for a total of 55 cars that can be handled. But on my layout, I’ve got about 13 sidings that can hold more than that. What’s the problem? Where did I mess up?

The answer is simple, but finding it required to study a lot of old pictures, most of them bad reproductions of printed material. To booth, some parts of the yard are seldom photographed, probably due to accessibility and the fact most photographers prefer to shoot locomotives. Add to that the fact I’ve never been in New York of my life, the prototype no longer exist and I’m stuck with limited access to primary and secondary sources… no wonder it’s so hard to piece up the puzzle.

Let’s identify every siding on the layout with a number in order of succession from the car float pontoon to the yard lead:

01 - Short coal siding near the pier and parallel to East 149^th Street

02 - Short siding to the pier on Harlem River

03 - Long siding parallel to the car float

04 - Long siding to the concrete loading ramp

05 - Long siding superposed to the freight shed/thawing shed up to the concrete loading ramp

06 - Siding #1 to gantry crane

07 - Siding #2 to gantry crane

08 - Siding #3 to gantry crane

09 - Very short siding parallel to Exterior Street

10 - Long siding parallel to Exterior Street

11 - Siding #1 to freight house

12 - Siding #2 to freight house

13 - Long siding between Freight House and gantry crane

Then let’s clarify some things about the different structures at Harlem Station. From now on I’ll call “Freight House” the large brick and concrete structure standing near Exterior Street. This is the office and it acts as the defacto “station”. The gantry crane was standing on the property, on the same spot, from day one until the terminal closed down under Erie Lackawanna tenure. The concrete loading ramp also seems to have been there since the beginning too.

Now, let’s tackle the freight shed/thawing shed issue. This is the long wood structure built by the concrete ramp and a lot of speculation exists about it. Without much fanfare, here are my findings. Reminder: please bear in mind most historic Harlem Station pictures aren’t in the public domain and by respect to their owner, I won’t publish anything. This freight shed is not an original building. On pictures shot in 1929 and 1932, even if not clear, the shed doesn’t show up. In the 1932 one, it is particularly clear two sidings run parallel up to the concrete ramp (sidings #04 and #05. My layout is based on that design and while not very accurate, the 1928 Sanborn Map do show the big freight house, the pontoon and what seem to be a structure over siding #09 (I wouldn’t be surprised this short siding used to be an engine house similar to the one standing at the adjacent CNJ Bronx Terminal.) However, no trace of the long freight shed which would have caught attention of people surveying the area.

The next piece of evidence is the 1942 Bromley Map which clearly depict the long freight shed built over what used to be siding #05. In fact, that siding is still – partially? – in place and run up to the shed wall. The next photo evidences are from circa 1951. At that point, we have relatively good shots of the shed.  While we can clearly see the turnout leading to siding #05 is still in place, there is not enough space to spot a car in front of the shed. Interestingly, a large pair of door exists on the end wall has if cars could have been shoved into the building. This is probably what lead peoples to propose – as an unverified hypothesis – it could have been a thawing shed for coal hoppers. While this idea would be great and logical, this is not my conclusion after carefully analyzing the architectural feature of that structure.

It must be noted the end doors, while sufficiently tall to clear a freight car, aren’t large enough. Also, some 1950s pictures clearly show a wooden platform serving that door with delivery trucks. This is further confirmed by the presence of a floor inside the structure. If I could describe it better, I’d say the freight shed was built over a wood platform erected on concrete bases. The presence of that platform or floor can be attested on every picture showing the freight shed. Thus, it would have been impossible to shove cars inside the structure. Finally, on some picture, it is clear siding #05 is buried in dirt and used by trucks to access the freight shed loading dock. All these evidences – combined with the fact the shed bear very little architectural resemblance to real thawing shed – proves it was a freight shed probably built to supplement lack of storage space at Harlem Station. It seems sacrificing a siding was less important than getting additional warehousing capacity. Thus, it is not a mystery this structure disappeared circa 1957 when the freight house was substantially enlarged. Since many sidings were removed at that time and the wooden freight shed was redundant, siding #05 was put back in service again.

Now, some people will ask what happened with siding #05 when the freight shed was built. My educated guess is the siding was left there and the structure built over it. There was little incentive to remove the track and maps and photos all agree the turnout and rails were still in place when the freight shed was still standing. This is interesting because it could indicate the freight shed was a temporary measure and Erie didn’t thought it was required to remove the track since they could be used later. It should also be noted the freight shed was quite a hastily-built structure clearly not made to last a long time.

But how all that new information impact the layout is interesting because it helps to understand the discrepancies between Erie Railroad Magazine description and my track plan. So let’s see things through new lenses.

Clearly, the sidings referenced by the 1952 article could be #01, #02, #03, #04, #06, #07, #08, #10, #11, #12 and #13. But that brings the count up to 11. Certainly #05 didn’t count since it was buried under the freight shed and unserviceable at that time. #09 was very short and all photographic evidences (including aerial photos) show this track was not accessible by road vehicle (no concrete pad, no dirt road and no grade crossing). It seems it was only used as an extra off spot car storage (some pictures show a car spotted there), maybe still used as an engine track from time to time. Unfortunately, is it hard to tell what was the last siding omitted. All of the remaining ones were actives and used to load and unload cars. My first idea was that one of the very short sidings at the pier (#01 and #02) was omitted. But both had a road access and were in use. Even the shortest one did hold two hopper cars for Miranda Coal. Could siding #01 have been only a storage track for Miranda since it would have been hard to unload a car there into a truck? I can’t tell, but on aerial photographs, there is a hint of a small concrete pad build against 149^th Street retaining wall. Anyway, there is no trace of coal handling at siding #02, confirming that commodity was unloaded on siding #01.

Then could it be that siding #02 was used only to hold extra cars even if it was perfectly accessible by truck? It could be. After all, a 1957 photo shows a loaded PRR coal hopper left there but in a spot that couldn’t be used for unloading purpose. Also, cars are seldom seen spotted on siding #02. A set of picture shot there in 1951 show a boxcar on the pier while the locomotive is picking up and setting out cars on the car float while a color picture from 1955 show a flatcar hastily left on the siding in similar conditions. This brings me to think this track could have been used as extra space when sorting out cars. It is perfectly located for the task and can be used to store extra cars the rest of the time. It would have been really handy and if required, it could have been used as an extra spot on demand.

Thus, I can now affirm with a decent level of certainty the ten sidings used by customers (this precision is critical) back in 1952 were #01, #03, #04, #06, #07, #08, #10, #11, #12 and #13. Erie Railroad Magazine mentioned the terminal could handle 55 cars at once which, incidentally, is exactly the capacity JMRI computed for my layout. This, I feel, is a great achievement.

Thus, I reach the conclusion of my research. Certainly I could go further, but the hours invested wouldn’t probably pay off in term of enjoyment. The layout is as accurate as can be with the data I can reasonable gather. At this point, continuing the search would divert resources, time and money that would better be invested directly on the layout. Such efforts would be to scratchbuild the freight shed and modify (or replace) the Walthers car float so it has only 2 tracks and a center platform as it used to be the rule on Harlem River. This would bring the maximum of cars brought to the terminal at 12.

According to 1951 Erie statistics, about 1800-1900 cars were handled that year. With a 6-day per week schedule, it would be an average of 6 cars per day, which is consistent with period photographs which show about 6 to 7 cars on the float. Doing so would reduce the exhausting workload generated by JMRI (16 cars per carfloat) and help alleviate the layout saturation with cars. This little detail as caused many car routing issues due to the impossibility to find open spot. And we must keep in mind that with an average of 32 moves per sessions, at about 3 minutes per move, operating the layout can take over 90 minutes which seems to be a little bit over the top if it happens to often. On the other hand, an average of 6 cars brings the session length to a comfortable 35 to 45 minutes, which I consider optimal to ensure the crew isn’t exhausted and isn’t bothered taking time to operate as prototypically as possible.

Finally, another aspect that JMRI brought forward was how often cars come back on the layout even with more than 60 cars on the roster. It became almost a joke after building several virtual trains. At some point, many cars were in and out at every session to the point suspension of disbelief was no longer working. It means three things: bringing the number of car per train lower will reduce the need to feed cars onto the layout, fine tuning the schedules will be required so cars won’t magically appear the next session when they should be traveling thousands of miles  and the roster will need to be buffed up. This last item isn’t truly an issue since I have many cars in my stash that could make great modelling projects. The diversity of rolling stock handled by terminal such as Harlem Station is also a great opportunity to learn better about railroading all over North America.

By the way, I’d like to point out I’ve been using OPSIG customers lists [ML1] to bring life to my layout by using real life customers for the layout. It certainly isn’t required, but felt it brought a sense of purpose to the layout. Isn’t it great to know that specific reefer is loaded with Carnation Milk from that plant or this Bronx based waste paper broker sell is stuff to a large paper mill in upstate NY? No need to go in great details, but knowing a few real customers of Harlem Station, the commodities handled and the roadname of cars seen on photographs really helped to shape a credible and immersing world.