CNR 3737 :: Piping

A muscular face

For many sessions now, the work on CNR 3737 – my S-3-a Mikado – has involved removing piping from the boiler, to the point where it was starting to look like a tube. On Friday, Andy Malette and I started adding piping – and already it’s a definite change for the better.

CNR 3737 Piping

CNR 3737 Piping

We started by removing the rest of the handrails (but keeping the stanchions in place), so they’d be out of the way. Then I bent up and added the exhaust pipes from the Elesco Feedwater Heater. This required a fair bit of trial and fit to get the pipes to hug the smokebox. We then installed the cold water supply pipe from the feed water pump. Next, we added the Hancock check valve on the top of the boiler, then fitted the hot water pipe from it.

I still have to add the condensate pipe, which runs from the side opposite the water supplies, down the smokebox, under the boiler, and back along the length of the locomotive towards the tender.

Before wrapping up the session, we managed to add the four sand lines, too.

While there’s a lot to do – and still some stuff to remove/reshape – it feels like we’ve turned a corner in this project. Thanks in part to its piping, this model is going to have a lot of character – and a very different look than it did when I bought it.

(Thanks for another great work session, Andy!)

CNR 3737: Test of Wisdom

On Friday, Andy Malette and I returned to working on our CNR S-3-a Mikados. With other commitments we had taken the summer off, and much of the autumn – our last day in the workshop was in mid-May – and it was time to get back at it.

Andy had cut and filed some sheet brass for us to fold into the covered steam turret located immediately in front of the cab. (Thanks, Andy!) I removed the exposed turret and the various lines that radiated from it, then bent the shroud and soldered it into place. It took some doing, and some cleaning up afterwards, but it’s in place.

CNR 3737 turret shroud

The next step was to start plumbing the turret (and the air pump, and the feed water heater, and…) … but I looked at the photos and looked at the model and nothing was making sense. We still had an hour set aside to work on things, but I realized that due to a combination of things (including lack of sleep the night before), I just didn’t have the focus to tackle the plumbing on Friday. So, we called it a day.

It was hard to do that – it has been months since we worked on these locomotives and I’m enjoying the process as much as watching new models come together. But I realized that I could do more damage than good if I kept at it. Upon reflection, it was the wisest decision I could’ve made.

I reminded myself of this today, while revisiting the model in the comfort of my own workshop. I again took a look at plumbing and, after installing one pipe between air pump and turret, pushed back from the bench and called it a day. Again, a hard decision to make – but the right one.

CNR 3737 turret shroud

I will look at the project later this week. Meantime, Andy and I are planning another day in his shop, later this month. I’ll do my best to get more sleep beforehand!

CNR 3737 :: Pump for Feedwater Heater

CNR 3737 feedwater heater pump

On Friday, Andy Malette and I held another session at his place to work on our CNR S-3-a class 2-8-2s. Having installed the Elesco Feedwater Heater during our previous session, we moved onto the pump that supplies water to this appliance.

The water pump sits ahead of the compressor, under the running board. But before I could install the water pump, I had to make room for it. This involved cutting away part of the running board on the fireman’s side then fabricating a new section of running board.

CNR 3737 - Al Paterson prototype photo - left side.

Using my fireman’s side photo, I planned the running board modification. I measured and marked the running board, then used a pair of tiny metal shears to cut the running board in two places, from edge to boiler. I then used the resistance soldering iron to remove the isolated midsection. More heat and a wire wheel in a Dremel Tool took care of the lumps of solder left behind on the boiler. I also removed the sand lines, which need to be re-routed.

Before installing the new running board and the water pump, it seemed like a good time to fill the holes left by the model’s check valves.

CNR 3737 - feedwater heater - check valve hole
(That hole in the boiler, just behind the foremost sand line, is the location of the original check valve. There’s one in the same location on the other side of the boiler, too)

I found a piece of brass rod just larger than the hole, and used an appropriate drill bit to open up the hole to accept the rod. I then tinned the hole and the rod and used plenty of flux and the resistance soldering rig to secure the rod. One it was in place, I used a cut-off disc to carefully cut the rod as close to the boiler as I dared. I then ground down the rod flush with the boiler, and finished up with progressively fine grades of sanding paper.

CNR 3737 - check valve hole - filled
(… and the hole is filled. I still need to polish the scratches out of the boiler)

I then fabricated a new running board segment from .020″ brass sheet. I used CA to spot-glue the segment I removed to my piece of brass sheet, with the outside edge flush to a sheet edge. I then used a pin so scribe the inside edge, which needs to follow the shape of the boiler and fit around the boiler bands. I made sure I had extra sheet stock to the left and right of this pattern, then sliced the original running board segment off the brass sheet, sanded off the CA, and cut out the new segment with shears. I then filed the new running board segment to final shape.

The new segment sits above the original running board, with tightly curved ends that meet the running board segments ahead and behind it – hence the need for extra material to each side of the new running board. I located the point at which I needed to make the bends and folded them down using pliers. I then measured the height of each folded down section and used shears to cut off the excess material. There are also various notches along the boiler edge to allow for the passage of piping, so I marked those out and filed away the unneeded material.

With the new section fabricated, I soldered it to the boiler then used photographs to locate the water pump. It’s centred just ahead of the boiler band over the third driver.

I cut and folded an L-shaped bracket for the pump, then soldered the pump to it using the provided mounting post. I trimmed the bracket so it would support the pump in the proper location, then soldered this to the boiler.

I finished up by placing the boiler on the chassis to ensure that the new water pump doesn’t interfere with the valve gear. It barely clears – but it’s sufficient.

Every work session brings this bog-standard USRA Mikado closer to my chosen CNR prototype. Compare the photo above with this photo of the stock model, taken just before we started this project:

OMI Light Mikado - Left Side

The main reason I’m doing this project with Andy is I did not have the know-how required to tackle it myself. I’m learning more at every session, What’s more, I’m throughly enjoying this project and look forward to what we tackle next!

CNR 3737 :: Feedwater Heater

CNR 3737 - feedwater heater
(Well, that’s looking very different…)

It’s been a while since Andy Malette and I worked on our CNR 2-8-2 brass-bashing projects. It’s just that time of year when other commitments get in the way. But last Friday, we got back at it by installing the Elesco Feedwater Heater on each engine.

The prototype S-3-a class did not come equipped with these. According to Canadian National Steam! – an essential resource for anyone modelling The People’s Railway in the steam era – shop forces in Battle Creek, Michigan added the appliance to CNR 3737 in September of 1940. I’m assuming that this was done as the Grand Trunk Western prepared to sell this locomotive to parent CNR – which occurred on February 18, 1941. CNR 3737 would retain its Grand Trunk Western livery at least until December 1948, when the Canadian government settled the duty, sales tax, and War Exchange Tax.

(As an aside, at the same time as they fitted the feedwater heater the Battle Creek shop forces also swapped in Boxpok main drivers – third driving axle from the front. Thanks for that, guys! I’ll have to retrofit that in a future work session. Andy and I are still discussing how to approach this work.)

A close look at our collection of prototype photos shows that on the engines that Andy and I are modelling, the feedwater heater was installed by cutting into the top of the smokebox and adding a platform. Side-on photos show that daylight can be seen in the lower corners, between the feedwater heater and the smokebox. With the aid of the flashlight on my phone, I’ve tried to capture what that looks like on the model:

CNR 3737 - feedwater heater - close up
(BTW, the boiler is set loosely on the chassis for these photos so there are gaps that will disappear when it’s properly screwed together)

Following Andy’s lead, I measured, marked and then cut the smokebox top to accept the feedwater heater. (I can admit that taking a cut-off disc in a Dremel tool to the top of a brass steam engine was a bit of a brown trouser moment, but I did not slip. There are no nasty gouges to fill.) I finished the hole with some careful filing. I then cut a platform out of a piece of brass sheet and soldered it into the hole. Andy and I considered building arched side walls for this platform but in the end we decided that they would be completely obscured by the feedwater heater, so there was no point.

The feedwater heater itself is a brass casting provided to us by our friend Simon Parent, who created them for his 2-10-2 models. He also provided us with the spoked pilot wheels. (Thanks, Simon!)

For our models, Andy and I had to slightly modify the connections. As provided, they have two small pipe connections on the back, at the edge, that run parallel to the handrail, one above the other. I had to remove the upper connection and relocate it to the front, pointing down. This involved clipping away the unneeded connection, filing the end to shape, then drilling a hole for the new connection, soldering in a piece of brass rod, and then soldering a flange to it.

CNR 3737 - feedwater heater

Finally, I tinned the mounting plate and the bottom of the feedwater heater, and Andy hit it with a propane torch – in and out, quickly! – while I held everything in place with pliers. That casting needed a lot of heat…

I need to clean up a bit of solder on the end cap – easy enough to do – and decide whether I want to fill the vertical notch. The prototype had a solid cap, but the mass of this casting will make it difficult to fill without unsoldering the new pipe connection and flange. I might just leave it. It adds character, and the caps could certainly be swapped from engine to engine – or even left off: one of the photos of Andy’s prototype (3702) shows it left bare.

With its new pilot and now a feed water heater, the model is really starting to take on some CNR character. There’s still a lot to do, including reworking the face of the smokebox and adding the water pump ahead of the air pump – a modification that will require us to chop the running board and elevate a portion of it. And of course, there’s that nasty Boxpok driver!

I’m not sure what we’ll tackle next, but we’re returning to Andy’s shop in just over a week. I can hardly wait!

CN 3737 – Cab back and railings

CNR 3737 - Al Paterson prototype photo - left side.
(Click on the image to filter this blog for all posts about this project)

Last Friday, Andy Malette and I held another joint work session on our CNR 2-8-2 projects. This time, work continued on the cab.

The first order of business was to finish the cab back. In a previous session, we’d squared off the rear of the roof – something the CNR did to make it easier to hang curtains to protect the crew in cold weather. This time, we added a back wall to the cab roof:

CNR 3737 - cab back

The wall is simply a piece of brass sheet, cut to match the curve of the roof and with two windows added according to prototype photos. Some of these cabs had the back wall flush, while others – like CNR 3737 – had a lip. Two small lengths brass were added under the side roof extensions, next to the back wall, and then trimmed and filed to length to complete the major modifications. This work required one to get in and out quickly with the resistance soldering probe, so as to not unsolder the roof extensions. I was really pleased that I was able to do this with no rework required.

As the above photo shows, we also added stanchions and railings to the cab roof. This was a relatively simple operation: mark and drill the holes, tin the stanchions, string them on a wire to keep them all properly aligned, then add lots of flux and hit them with the heat.

CNR 3737 - cab roof rails

We left the wires long to the rear of the stanchions, then trimmed them after soldering. At the front, the handrail loops 180 degrees then bends parallel to the front cab wall, so we did that too:

CNR 3737 - cab roof rails

The cab still needs an armrest under each window, but we’ll add that after painting. I think it’s pretty much done, and can be set aside while we start on the next phase. I’m not sure what that is, but I’ll find out at our next work session. I’m looking forward to it!

If you ever get a chance to learn from someone who knows their way around a brass engine… do it! (Thanks for teaching me, Andy…)

LokSound Love for 1532

CNR 1532 - LokSound installation
(Replacing the decoder in the 10-wheelers looks challenging, but it’s really just a case of mapping the wires and doing things one wire at a time)

Over the past week, I’ve done a fair bit to advance my hobby goals.

I’ve resumed working on trees for Port Rowan, and I’m pleased with the progress: I applied my bark mixture to nine more armatures this morning.

I had another work session at Andy Malette‘s place, as he and I convert USRA Light Mikados into CNR S-3-a 2-8-2s. (More on that in this post.)

And I finished converting the core fleet of steam locomotives to LokSound Full Throttle Steam, with the installation of a LokSound Select into the boiler of CNR 1532 this morning. With that, I’ve finished the two moguls and two 10-wheelers that I use in regular operating sessions. I’m loving the new sounds and the motor control. This is what I was looking for.

I have a couple other steam locomotives to convert, but I can do them as time allows.

All in all, a fine week!

Through the Lynn Valley


(You may also watch this directly on YouTube, where you may be able to enjoy it in larger formats)

Another day, another video of one of my CNR moguls equipped with Full Throttle Steam – the new sound packages soon to be released by ESU for their Loksound Select and Loksound V4.0 decoders.

I’ve spent a little more time running the locomotive and I’m getting much more comfortable using the Heavy Load and Coast features to bring the sound to life.

As I’ve mentioned in previous posts, these are beta sound files. The production versions should be released soon. Watch the ESU/Loksound USA website for details.

Meantime, I’m getting ready to replace decoders in more locomotives. It’s a great time to be modelling steam!

CNR 86 – Full Throttle – 2nd Run

I’ve made some more adjustments to the Loksound decoder in CNR mogul 86 and CNR 10-wheeler 1560, which are loaded with Full Throttle Steam packages from ESU. And therefore, I’ve made a follow-up to yesterday’s video… this time focussing on 86 in action on my layout…

In this video, I’ve highlighted a number of sounds generated by the decoder. Some are automatic, some are user-controlled, some are both. The video features braking noises, the air compressor, bell, whistle, injectors and dynamo.

In the first scene, the locomotive drifts into St. Williams. In the next, it works hard to start the train out of St. Williams (with Full Throttle’s “Heavy Load” function engaged). Finally, the engineer drifts over a bridge in the Lynn Valley (with Full Throttle’s “Coast” function engaged), before opening the throttle to build speed for the run into Port Rowan.


(You may also watch this directly on YouTube, where you may be able to enjoy it in larger formats)

I still have some minor tweaking to do – notably, to adjust volume levels – but I’d say I’m 98% of the way there. Colour me impressed!

Working hard, and drifting

I now have LokSound decoders installed in two of my steam locomotives – CNR mogul 86 and 10-wheeler 1560 – and I’ve loaded them with pre-release versions of the soon to be released Full Throttle Steam sounds and features, thanks to Matt Herman at ESU North America.

I’m still tweaking the sound and motor controls, but I’m 90% of the way there and wanted to share a quick video to illustrate one of the features I really like about this new line.

The Full Throttle Steam series will include a function similar to “Drive Hold”, which is a key feature in ESU’s Full Throttle Diesel sounds. Drive Hold is mapped to a function button and is turned on and off just like activating a bell sound. When it’s engaged, the feature locks the locomotive’s motor at its current speed. Turning the throttle knob will not adjust the speed of the train. But it still adjusts the sound of the locomotive.

Here are two ways it can be used:

If one is pulling away from a station, one can open the throttle to start the locomotive, then lock the motor once a desired (still slow) speed is reached… then continue to increase the throttle to make the locomotive sound as if it’s working harder to get the train underway. The exhaust will be sharp and strong, as if the hogger has put the Johnson Bar right into the corner.

Once one is at track speed, the motor can be locked and then the throttle can be turned down to represent pulling the Johnson Bar back closer to neutral. The exhaust note will be softer and quieter. At its extreme – turning the throttle knob all the way to speed step zero – the exhaust sound will disappear entirely, as if the hogger had shut the throttle. The locomotive will now drift indefinitely, simulating a prototype that’s being carried along by the train’s mass and momentum.

I’ve shot a very brief video that illustrates both of these features. First, I show CNR 86 starting from a station stop. At St. Williams. Next, I show CNR 1560 switching from throttle to drift as it passes the station.


(You may also watch this video directly on YouTube, where you may be able to enjoy it in larger formats)

It does take a little bit of practice to do this smoothly – but 20 minutes of playing with this feature should fix that. It should be noted that one does not have to use this feature: One can control the locomotive in the conventional way and still get a sense of working hard and drifting by writing high momentum values into CV3 and CV4. But using the motor speed-lock feature is a much more powerful way to accurately replicate the sound of steam.

In fact, the best solution is a combination of these two approaches. I notice the first locomotive (CNR 86) speeds up abruptly as it’s leaving the scene. Increasing the value in CV3 (acceleration momentum) should take care of that, because it will smooth the transition between the locked motor speed and the throttle setting once I release the motor. I may also increase the value in CV4 (deceleration) to help smooth the transition when slowing down. For me, that’s part of the fun of experimenting with DCC.

A special thank-you to Matt at ESU, who prepared these pre-release files for me as part of our Full Throttle Steam recording session at TrainMasters TV last Friday. Matt tells me he will release of the first Full Throttle Steam decoder files very soon, and I’ll be sure to update the blog when he does.

It’s a great time to be modelling the steam era!