CNR 3737 :: Smokebox front

On Friday, Andy Malette hosted me for another work session on the CNR 2-8-2 project – and we tackled a major modification: the smokebox front.

New smokebox front

Our donor engines are USRA-designed light Mikados from Overland Models, which feature a smokebox front held in place with 20 dogs. Depending on the class of locomotive (and possibly the builder) Canadian National Railway steam engines featured either 10 or 12 dogs on the smokebox front.

We tend to notice these things, so it was obvious that the old dogs would have to come off and be replaced with new ones in the proper pattern. In this case, I would need 12 dogs – which means I couldn’t even cheat and simply grind away every other dog on the USRA smokebox front.

The Overland Models smokebox front is removable, but the dogs are cast as part of the face. So the easiest way to remove them was to turn them off on a lathe. I worked with very light passes, checking my progress frequently until I was happy with the results. After using the cutter, I finished up with progressively finer grits of sanding paper until all evidence of the dogs had disappeared. (This process also removed a portion of the hinges, but they will be relatively easy to replace with brass strip.)

Once the old dogs were gone, it was time to install new ones. Andy has some dogs cast by another friend of ours, but I would have to drill the smokebox face to accept them, then solder them in place. We moved the smokebox front from the lathe to the mill and added a 12-position indexer to evenly space the holes:

Index and centre drill

I started with a centre drill, then carefully drilled for the dogs, adding cutting oil for each hole. The last thing I wanted was a broken drill – Andy said he’s ruined a couple of smokebox fronts that way – and I’m pleased to report I drilled 12 perfect holes with no incidents.

Andy then locked the smokebox front in the vise and demonstrated how to solder these tiny dogs into place. He did the first one – I did the next 11. A minimal amount of flux, heating near – but not on – the dog, and a light touch with a thin length of solder on the side away from the heat was the key: Done right, the solder would melt onto the face of the smokebox than draw itself under the dog.

Here’s a composite photo showing the smokebox front, before and after the alterations:

Smokebox before and after

I think it was definitely worth the effort.

As part of this work session, I also added a drain pipe to the feed water heater, which I’d somehow missed during last week’s piping session. It’s the smaller pipe in this next photo, just ahead of the large exhaust pipe:

CNR 3737 Piping progress. Engineer's side front.

There are still a few details to address – including the various appliances that mount on the smokebox front. But I now feel confident that CNR 3737 is heading towards the paint shop this year. I’m really pleased with the progress made so far this month and look forward to the next session…

CNR 3737 :: The engineer’s side

After a hiatus for the holidays – and a month of avoiding the outside as much as possible due to heavy snow and bone-chilling cold in our part of the world – Andy Malette and I emerged like Wiarton Willie (a month late), saw our shadows, and decided it was time to continue working on the CNR 2-8-2 project.

At yesterday’s session, I turned my attention to the engineer’s side of the engine:

Piping - Engineer's Side

Here, I added some new piping, including the assembly in front of the cab and the angled pipe leading to the Hancock check valve on the top of the boiler, to the right (ahead) of the steam dome. I also reinstalled some piping under the cab and below the running board. And, I fabricated and installed the raised running board section that spans the power reverse unit.

For some reason – still a mystery to me – my soldering was much better today than it has been in the past. Maybe I’m finally getting the hang of this? I won’t inquire too closely – I’ll just roll with it, and keep doing what I’m doing. We’re having another session in a week, when I’ll tackle a project that will make major change to the appearance…

CNR 7456 in HO

I haven’t been doing much on Port Rowan this year for various reasons. Truth be told, I haven’t done too much in the hobby this year, period. But I have been trying to keep my hand in – primarily with some projects for others.

This locomotive is one of them:

CNR 7456 - Weathered
(CNR 7456 in HO scale)

A while back, my friend Stephen Gardiner and his wife Heather bought a townhouse – and in the summer, a bunch of us descended on his place to build benchwork for Stephen’s HO scale layout based on Toronto’s Liberty Village district. (You can read more about the benchwork party on my Achievable Layouts blog, and more about Stephen’s Liberty Village layout on his blog.)

Even before Stephen moved into his new place, I knew that I wanted to have a locomotive to take out to operating sessions. And when I happened to stumble across a “like-new” example of the brass CNR O-18-a imported many years ago by Van Hobbies, the die was cast. I picked up this model earlier this year, and started working on it back in May.

If I’m counting correctly, this is the fourth example of the VH O-18-a that I’ve owned, and I’ve regretted selling on every previous model, so I was excited to find this one. And it was indeed in great condition. Every one of these that I’ve owned has enjoyed a super smooth mechanism ideal for slow speed running, and this model continued in that tradition. However, the models are quite venerable now – they were imported a couple of decades before anybody had even heard of DCC – so they do need their motor upgraded. I also needed to drill the headlight and back up light and provide holes for wire runs.

(As an aside, after I acquired my O-18-a, another friend – Ryan Mendell – also picked up one, which he’ll use on his new Grand Trunk layout. And that led Stephen to find his own O-18-a – so we’ve started a club of sorts and have been sharing ideas for updating them.)

To make a long story short, I’ve done all that. I’ve added a LokSound Select, a TCS Keep-Alive (with a cut-out switch for programming, accessible from between the centre sills of the tender frame), LED lights, and a pair of ESU sugar cube speakers. It’s pretty crowded in the tender!

CNR 7456 Tender gubbins
(A view of the gubbins)

Up front, I’ve replaced the old open frame motor with a NSWL can motor, including a new bracket I fabricated from brass. This was a hurdle for me – but it turned out to be much easier than I thought it would be. The lesson learned is “Just go ahead and try, because it will probably work – and if it doesn’t, it’s just a bit of brass sheet”.

For this model, I decided to branch out from the typical model railway suppliers and experimented with Tamiya paints from my local plastic modelling hobby shop. I’m really impressed and will be using these a lot more on future projects.

But of course it wouldn’t be one of my projects without some sort of disaster. Yesterday, I reassembled the model and went to test it – and the decoder blew. I traced the fault to the bare contact on one of the sugar cube speakers, which came into contact with the bare brass of the tender interior. I thought I had secured the speaker enclosure to the underside of the top of the tender shell, but it worked its way loose. Lessons learned: Do a better job of securing the speaker enclosure and cover up those contacts.

Meantime, I’m in for another decoder – and a lot more fussy wiring. I’m kind of discouraged by that, so I’m not going to tackle it just yet. But I have plenty of time to get this model ready to run on Stephen’s new layout…

UPDATE: December 13, 2018

CNR 7456 - Fixed
(That’s more like it!)

On the weekend I was able to nip through an area hobby shop and pick up a replacement decoder – and yesterday, I installed it. This time, I made sure all speaker terminals were insulated (I applied Bondic to each one) and I also wrapped some of the interior of the brass tender shell with Kaptan tape.

The ESU approach to decoders once again proved its value: since any LokSound decoder may be loaded with the user’s choice of ESU sound file, and managed through LokProgrammer, I was able to buy the appropriate decoder – a LokSound Select Micro – with a diesel sound package preloaded on it. I then simply used the LokProgrammer to overwrite the package with my file for CNR 7456, which not only replaced all the sounds but also rewrote all the CVs to those I’d established before I blew the previous decoder.

The locomotive is now back together and running as it should. I still have a few details to address, such as a crew, window glazing and – perhaps – cab curtains. And I may want to adjust the brightness of those LED headlights. But the hard work is done!

As an aside, I picked the locomotive number – 7456 – back in the summer while visiting my friend Andy Malette. The choice was practical: Andy had a limited selection of etched brass CNR number plates and 7456 was one of the ones still available. Andy also supplied the lovely brass numerals for the cab sides. (Thanks for those, Andy!)

After deciding on 7456, I was pleased to discover a photo of the prototype when I visited the Andrew Merrilees Collection at Library and Archives Canada in September:

CNR 7456 - Merrilees

You’ll note there are a number of small differences between the prototype and my model of it. Notably, the coal bunker should be taller, the handrails are different on the tender and around the smokebox, and the headlight is lower on the smokebox front. The number board is also at the back of the headlight bracket, instead of at the front as it is on the model. However, I had already painted the locomotive when I found this photo, and a decided I could live with the discrepancies. Maybe on my next one…

CNR 3737 :: Reworking those big pipes

Much better:

Feed water pipes fixed

I hate going back and re-doing things – but sometimes, it has to be done. A case in point is the big exhaust pipes hanging off the back of the feed water heater on my brass CNR 2-8-2 project.

Given the size of the brass rod involved, these pipes were a royal pain to bend. And they soak up heat like nobody’s business, which made them a pain to solder in place, too. The problem was, they should drop almost straight down after coming out of the appliance – but I’d made the top bend too far back and then compounded the error to bring them back forward, around the front of the cylinder saddle. It really changed the look of the whole front end of the model and I wasn’t happy.

When I mentioned this to my friend Andy Malette at the start of yesterday’s work session, he advised that it’s up to me to decide if I can life with the inaccuracies – but that if I was having doubts, I should probably correct it now, before the locomotive is painted, etc. Andy also pointed out that brass is forgiving, and relatively cheap. If I don’t like something, I can unsolder it, either re-bend or fabricate another one, and reattach it.

He’s right. So I did: I unsoldered the pipes from both sides, re-bent them, and put the new pipes in place.

Looking at an older photo from 2017 (below) and the lead photo, taken after yesterday’s work, I’m glad I spent the work session addressing the problem. It looks a lot better now.

Old piping
(This older photo shows my first, failed attempt at fabricating the exhaust pipe from the feed water heater)

Speaking of piping, there’s more to do – but probably not until sometime next year…

CNR 3737 :: Tender work “finished”

My friend Andy Malette and I basically shelved the CNR 2-8-2 project for the summer and fall, as real life took priority. But with the weather turning colder and the project languishing for many months, we decided to get back at it yesterday. With a few hours work, I can now declare the brass-bashing on the tender “finished”.

Tender - body finished - front

Tender - body finished - rear

Is any project ever finished? There’s more I could do, I’m sure. But at some point, one has to put down the tools and decide it’s time to move onto the next phase of the project. As I look at these photos, I see a few things I could still do – but other than straightening some bent handrails and drilling a hole to run wires for the rear light, I’m happy with things as they are.

Next week, it’s back to the front end…

ProtoThrottle Progress

Setting up ProtoThrottle

Over the weekend, I set up the decoders in my GE 44 Tonner and my Gas Electric to work with the ProtoThrottle, and I’m very pleased with the results.

Introducing the ProtoThrottle to a layout is a multi-step process.

– The ProtoThrottle must be connected to the layout’s DCC system (which I wrote about earlier this month).

– Each locomotive that will be used with the ProtoThrottle must have its decoder setting tweaked. This isn’t necessary to run with the ProtoThrottle – but doing so allows one to leverage all the capabilities of this realistic control stand.

– For each locomotive, a configuration must be built and saved within the ProtoThrottle itself. This includes address, plus settings such as the braking behaviour, notch points for the throttle, and rules governing the lighting switches.

For this work, it definitely helps to have a programmer at the workbench. Depending on the decoder being used, that’s going to be either something like DecoderPro (JMRI) or – in my case – the ESU LokProgrammer. (These are good ideas for anybody with even a single sound-equipped DCC locomotive, regardless of whether one’s using a ProtoThrottle, because they greatly simplify setting CVs.)

To provide an idea of what’s involved, I’ll share the adjustments I made for my 44 Tonner. I’ll also share some of the adjustments I made for the gas electric, to demonstrate some of the changes one might consider for a locomotive with different performance characteristics.

ProtoThrottle set up for CNR #1

Proto Throttle - Port Rowan

The ProtoThrottle can store up to 20 configurations. These include the locomotive address, function mappings, throttle notch settings, and other options. These are some of the values that went into the configuration for my 44 Tonner, which is equipped with an ESU decoder and a sound file with the Full Throttle features:

Under the Configuration Function menu (CONFIG FUNC), I set the Horn to F02, Bell to F01, Brake to F10, Brake Off to F — (not set), Aux to F09 (to enable Drive Hold) Front (F) Light to F00, F DIM #1 to F00, F DIM #2 to F12, Rear (R) Light to F05, R DIM #1 to F05, R DIM to F12.

The next task was to match the engine sounds from the decoder to the notches on the ProtoThrottle. When I move from Notch 3 to Notch 4 on the throttle, I want to hear the model notch up accordingly. To determine the notches for my 44 Tonner, I first ran the locomotive with a regular DCC throttle equipped with a speed step indicator. Working with 128 speed steps, I increased the throttle one speed step at a time, and made a note of the speed step at which the engine sound changed – in other words, the point at which the decoder generated a “notch up” sound. I then picked values that lay between the notching steps.

For example, if the decoder notched up from 2 to 3 at speed step 20, and notched up from 3 to 4 at speed step 35, I decided that notch 3 would be set to speed step 29.

Having noted the values, I then returned to the ProtoThrottle. Under the Notch Configuration menu (NOTCH CFG), one sets the speed step that each notch on the ProtoThrottle will send to the decoder. As noted earlier, this can be set for each of the 20 configurations saved in the throttle. Based on my tests, I set the notches for CNR #1 as follows:

1 = 8, 2 = 17, 3 = 29, 4 = 40, 5 = 49, 6 = 60, 7 = 70, 8 = 90

Finally, I configured the brake handle. I tried both approaches, and decided that I did not gain anything by using the Variable Brake capability. So in the OPTIONS menu, so I set this to OFF. I also set the emergency stop to OFF, since I’ve never needed it using other throttles on my layout.

That completed the configuration of the ProtoThrottle for CNR #1. I saved the configuration, then turned to the decoder itself.

Using my LokProgrammer, setting the characteristics for the decoder in CNR #1 was intuitive and adjustments were easy. It required a fair bit of time, however, as I would make a change or two, then switch to driving mode and test my updates.

I wanted to use a Full Throttle file from ESU, but while ESU offers a sound package for a 44 Tonner, it has not yet been upgraded to include Full Throttle features. The great thing about Loksound decoders, though, is that I can load anything into the decoder for now – and upgrade it to the proper sound file if/when it’s made available. 44 Tonners were powered by a pair of Caterpillar D17000 V8 prime movers. I scrolled through the Full Throttle options and decided that the file for CP Rail’s oddball CAT 3608-powered M636 would do for the time being. (Again – I know that’s not right. But I can update the sound if/when the correct file is available with Full Throttle features.)

I won’t list every value here – that would take a book – but I will share the thinking behind some of the key decisions I made. (I’ll include the LokProgrammer language for those who use it, but also try to explain it so it doesn’t sound like gibberish to those who do not.)

Under Motor Settings, I enabled Back EMF and the heavy load/coast load settings that enable Drive Hold on a Full Throttle-equipped Loksound decoder.

Still under Motor Settings, I then used the Three Values option (Voltage Start, Voltage Mid, Voltage High) to adjust the motor speed. In the LokProgrammer, there’s a graph for this, with a slider. I dropped the top speed (V High) from 255 to 50. That may seem slow, but I get frustrated when I’m running on a layout with a throttle that offers me 128 speed steps, and I’m stuck using about 25 percent of that because anything higher is too fast. What’s the point of having 128 steps if you’re never running above speed step 30? So on my own layout, I knock down the top speed of every locomotive so that I can take advantage of the full range of speed steps on the throttle. According to this neat article about the prototype, GE 44 Tonners were limited to a top speed of 45 mph, “although it’s doubtful many actually achieved it”. What’s more, the top speed on my layout is a blistering 20 mph. Scale speed is subjective – what works for me may not work for others – but to my mind, setting the maximum voltage to 50 seemed to provide the right top speed for this little locomotive.

Under Driving Characteristics, I set Acceleration Time to 170 (42.5 seconds from full stop to top speed) and the Deceleration Time to 255 (63.75 seconds from full speed to stop). High values for these settings serve two functions. First, they allow the prime mover sound on the decoder to ramp up before the locomotive moves… or drop off to idle while the locomotive continues to roll (representing the momentum of a heavy object rolling on rails). Secondly, on the ProtoThrottle they smooth the transition from one speed step to the next.

Obviously, one can get into real trouble with the deceleration set at 255. On my layout, the 44 Tonner running at full speed (which is not very fast) will roll about 11 feet before coming to a stop if I simply drop the throttle to “idle”! That’s where the brake handle comes in. Under Brake Settings, I set the Dynamic Brake to 64. This will bring the locomotive to a stop from its maximum speed in 16 seconds. I arrived at this value by testing the locomotive to find a brake that was responsive enough to allow me to stop the locomotive where I wanted to fairly reliably, without being too aggressive. With the Dynamic Brake set to 64, CNR #1 will go from full speed to full stop in about 15 inches when the throttle is shut off and the brake is applied.

The following Function Mapping are relevant to the configuration settings in the ProtoThrottle. To set up the front and rear lights so they work with the throttle’s rotary switches, I mapped the physical outputs for the front light to FO(Forward) and FO(Reverse), and the rear light to F5. To enable dimming, I mapped the logical function on F12 to “Dimmer”. (For each light, I also entered the Function Outputs menu and set them up as dimmable lights with fade in/out, knocked down the brightness a bit, and enabled the Dimmer and LED mode special functions.)

Again, these are all personal preferences, based on setting values, then running the locomotive and making notes of what worked and what didn’t. If you have a ProtoThrottle, don’t simply do what I did: do your own tests and pick settings that are right for you.

Proto Throttle - first run with Gas Electric

I also set up my gas electric. Many of the settings are the same as in the 44 Tonner – in both the model’s Loksound decoder and the ProtoThrottle configuration. For example, the front headlight settings are the same. Since the model does not have a rear headlight, I disabled those settings in both the decoder and on the ProtoThrottle.

As a passenger unit, I wanted the gas electric to have a higher top speed than the 44 Tonner. Therefore, using the slider under Motor Settings, I gave it a top speed of 100 (versus 50 for the 44 Tonner). Note that this does not mean the gas electric goes twice as fast as the 44 Tonner: each model has a different drive train set-up, including unique gear ratios. So I set the top speed based on each model, by setting a value, testing the unit on the layout, and adjusting as necessary.

I also wanted it to have snappier throttle response so under Driving Characteristics, I set the Acceleration Time to 125 (versus 170 for the 44 Tonner) while keeping the deceleration value at 255.

The introduction of the ProtoThrottle has definitely been worth the investment for these two models. Switching with the 44 Tonner is a completely different experience than it was with a standard DCC throttle. And driving the gas electric with the ProtoThrottle makes a straightforward passenger run into a much more engaging experience. I’m glad I did this, and I look forward to setting up more locomotives to take advantage of this throttle. As mentioned in a previous post, I need to upgrade the decoder in my CNR RS18 and the ProtoThrottle is the incentive to move that project up the to-do list.

Now, when will I see a “Proto Johnson Bar” controller for my steam engines?

CNR 3737 :: more tender work

Someone recently asked me how work was progressing on the CNR 3737 project. The short answer was, “It wasn’t”. The long answer was that Andy Malette and I were both busy with other things and just couldn’t find a free Friday that worked for both of us. That’s fine – it’s a hobby: It fits between the other things in life.

But after a long hiatus, we managed to get together last week and make some progress. This time around, I added railings to the tender:

CNR 3737 Tender Railings

I marked locations for stanchions and soldered a bunch of brass strip in place, leaving each piece longer than needed. I then soldered the railing to the stanchions, using a scrap of strip wood as a non-conductive spacer to make sure they railing was a consistent distance off the tender walls.

My prototype has separate railings along the coal bin, whereas the locomotive Andy is building has one continuous rail that follows the lines from the coal bin down to the rear deck, around the back and back up the other side. I definitely had the easier project.

The rear railing is actually two pieces, soldered in place then trimmed to meet on the stanchion next to the ladder at the back of the tender. the railings simply end behind the coal bin walls.

CNR 3737 Tender Rails

On one of the coal bin railings, a couple of brass fittings are soldered in place at each end. These are the electrical plug-ins for the rear light, which will go on the water tank deck.

After shooting these photos, I did some clean-up. I filed the stanchions flush with the top of the railing. In the process, I managed to break a couple of the solder joints, but the repairs were quick – thanks, in part, to my new low-profile vise, which opens enough to hold an S scale tender body, and which can be used as the grounding point for my resistance soldering unit.

It’s nice to be back at this project. It’s taken a long time and I’m now at the stage where I want to get it done and move onto the next thing. Plus, of course, I want to see the locomotives in action on the S Scale Workshop exhibition layout!

CNR 3737 :: Tender

I’ve been tardy in updating my blog because it’s been very busy lately, so this is actually a report on two work sessions with my friend Andy Malette. Both focussed on the tender for CNR 2-8-2 number 3737

Let’s start with a reference photo – the stock tender that came with the URSA light Mikado from Overland:

CNR 3737 - stock tender

In the first session (held at the end of January), I reshaped the side walls forward of the coal bunker. On the stock model, these slope back to the deck. But CNR 3737 has a semi-enclosed cab, which meant these needed to be modified. The trick is the fine strip of beading along the top of the side walls: We wanted to preserve that.

A careful application of heat and a single-edged razor blade lifted this off, about one third of the way back along the bunker. I was then able to cut and file away the angles on each side. Finally, I cut and shaped new wall sections to build up the front of the side wall. Once these were soldered in place, I carefully re-bent the bead and soldered it down. Here’s the result:

CNR 3737 - tender mods

When I got home, I realized that the tall walls to either side of the coal bunker doors would also interfere with the back of the semi-vestibule cab…

CNR 3737 - tender mods

… so, off they came:

CNR 3737 - tender mods

The deck to either side of the coal doors is pretty messy now – but the good news is, my prototype photos show spilled coal all over these small decks, so I’m not going to worry about it. I will have to do some clean-up and filling around the side wall extensions that I added, though.

While I was doing that, Andy was prepping for our next session (held yesterday). He cut some channel and angle to length and drilled it for me so I could build new steps at the front of the tender. Thanks to his prep work, the assembly went quickly. Compare this image to the stock photo:

CNR 3737 Tender - front steps

Each ladder assembly consist of 14 pieces. Andy tells me his took a lot of time to assemble, and he was surprised mine went together relatively quickly. Of course, what goes around comes around: The other project during yesterday’s session was building a three-piece assembly for the rear number board. It consists of two C-shaped brackets and the number board itself… and for the life of me I could not get everything to solder properly. Andy eventually stepped in and got it mounted – and I will have a lot of clean-up to do on the rear wall of the tank:

CNR 3737 - tender number plate

The tender still needs a ladder on the fireman’s side, plus railings, power conduit, rear light, and other details. But it’s already looking a lot more like it belongs on the CNR.

CNR 3737 :: Piping near the cab

CNR 3737 - Piping.

I’ve been exchanging notes with a reader and he mentioned he’s hoping for more progress reports on my CNR 2-8-2 – so this one’s for you!

Progress has been slow, so there’s not much to report. Our schedules have conflicted more often than not, so my friend Andy Malette and I have only been able to hold a couple of work sessions over the past several months. That’s fine – it’s a hobby, and the work will wait until we’re able to do tackle it.

At our last session – late last year – I installed some piping ahead of, and underneath, the cab. Unfortunately, I’d forgotten to take along some brass castings that were essential to this work, so I had to revisit the piping, cutting into it in some places and re-bending it in others. But the result can be seen in the photo above.

I still have to do the other side.

I have to confess that tracing pipes on photographs makes my head swim. The pipes duck in and out, behind appliances, under running boards, and so on. And the old photos are often a bit grainy or taken at a typical, 3/4 front view from track level. So sometimes, it’s a guess at best. Now, it’s an educated guess: I’m following piping diagrams from various sources, including the Model Railroader Cyclopedia – Volume 1. But those sources are guidelines, at best. As any student of steam locomotives knows, appliances and piping typically varied from unit to unit, based on where and when the locomotive was last shopped and what was on hand at the time.

I find that printing out the photos – or sections of photos – in a larger size helps. I can then use a selection of markers to trace each pipe in a different colour. Assuming, of course, I can see them in the prints…

Locating piping is further complicated by the fact that the pipes cannot always go where the prototype put them. In the photo above, I’ve had to run the lowest pipe parallel to the running board until it clears the space required for the trailing truck, then curl it downwards. On the prototype, this pipe cuts across that open space at more of a 45 degree angle. But of course, I want my model to be able to negotiate the curves on the S Scale Workshop modules (for which I’m building this model) – and my own layout.

Back to staring at photos and taking notes. It’s all part of the learning process.

Steam Locomotives (the Cyclopedia)

More accurately, Model Railroader Cyclopedia – Volume 1: Steam Locomotives:

Steam Locomotives - Cyclopedia

This arrived for me this week, after a discussion with my friend Andy Malette about research materials for our CNR Mikado project. Andy noted that this book taught him a lot about the various appliances on steam locomotives, as well as the myriad of pipes that connect them. So, I grabbed a copy via ABEbooks. And Andy is right – there’s a ton of information in this tome.

The caveat is, the information is of course “ex-Works”, “best practices” and so on. If you’re detailing a locomotive, as we are, it’s important to check prototype photos of the exact locomotive you’re trying to model. This is particularly important with steam engines, and even moreso if they’ve been around for a while: just like a subdivision can start out looking like it’s built from Monopoly houses, yet acquire character through the passage of time, individual steam engines often developed a unique character as shop forces worked to keep them in service, and to modernize them.

In fact, that’s one of the joys I’m experiencing in doing this project with Andy. We’ve each picked different numbers – I’m doing 3737, while he’s chosen 3702 – and the two locomotives are very different. The plumbing is different. The location of appliances is different (for example, on Andy’s locomotive, the location of the feed water pump and the air pump is reversed). The smokebox fronts are different. The sand domes are in different spots. And so on. When we’re finished, we will have two locomotives of the same class that each exhibit their own character, and have their own back stories.

This is what makes prototype modelling so rewarding. As a friend is fond of saying, “Details Matter”.