benshell

After a month off for traveling and insanely hot weather making work in the garage/outside miserable, I'm back with some exciting news and a video to share! I pressed in the rolling bearing and reassembled, and the K4 is running smoother than ever! But just in case, this time around I didn't test run first... I took this video on the very first loop around the track with the new bearing! 20 cars was no problem until the entire train was on my 2.5% grade... then I had to lighten the load to 16 cars. I could probably add a little more lead weight to improve traction, but I probably won't bother for now. I just wanted another smooth running steam locomotive that could pull a decent train, and I feel I've reached that goal now. Also, it's running at top speed, which is about 20 volts DCC. A prototypical passenger train should run a little faster, but this is faster than I run most of my trains so I'm happy with the range. I guess the next step is getting PRR heavyweights. I'd been on the lookout for Aristo-Craft cars, but at this point I'll probably wait for the new USA Trains heavyweights. In the meantime, the K4 will be engaged in frequent freight service.

Yeah, it's unfortunate what has happened. I bought my last QSI Titan for my K4 back in February, just in time apparently. I'm still hopefully they will resurface at some point though. It's great hardware and software, it was just terrible customer support.

Wow!

This is looking and sounding great Joe! And the center set Kadee looks really perfect on it. It's such a nice improvement. (You really have me tempted to get one... and they are even on sale again at Charles Ro right now. Can't get everything though and I'm gonna have to continue to past on them for now!)

Great job! Improved lighting sure does make a huge difference! This makes me want to work on the firebox flicker on my steam locomotives. I only have one with firebox lights, but it's too dim to really see unless you're up close and looking for it.

Chuck, I will have to take a look at my USAT SD70s... I think one of them has also starting squeaking. Thanks for the suggestion! Joe, that's my thought as well. There are a lot of forces at work here, and if my motor alignment is even a thousandth off (likely!) it would have made this problem worse. I'm hoping roller bearings help. I wish I could say I finished the roller bearing assembly and tested this weekend... but just machining the bearing housing took all my limited shop time. I guess I also took the time to draw up the part in 3D and make a drawing (attached as a PDF), using Autodesk Fusion 360. I've been learning this program lately and I've been really impressed. Now I just need to press in the bearing and install. I guess I'm a bit hesitant on pressing in the bearing as there's no removing it (at least without destroying the part). I also realized that the bulky size of the new bearing housing is going to interfere with some supports for the electrical pickups. Oops. I guess I have another modification coming up. K4 Bearing Housing.pdf

Hi everyone, I've been behind on the updates here, but the story continues... I (barely--the night before) finished installing DCC for my June 25th open house and (mostly) put it back together to run. I ran the locomotive a few times around the track, pulling 10 cars. It was running well, except that the chuff sync was WAY off (I accidentally calibrated it for 4x the correct number of chuffs!), until it slowed to a stop! Looking into the engine from underneath, it appeared that a wire had gotten twisted around the motor shaft. So back to the workbench it went. Over the next couple weeks I opened it up again, did a better job with my cable management, fixed the chuff issue, thoroughly tested on rollers, and then fully reassembled. I also added more weight. Last weekend I tested again, pulling 18 cars quite well (no wheel slip). I was about to take a video, when, once again, it slowed to a stop and wouldn't budge! This week I've disassembled again... more than ever before... and found the motor shaft complete frozen in the gearbox bushing. I've never seen anything like this. First I tried using my lathe chuck + vise grips and I could barely turn the bushing, but couldn't remove it. I ended using my 1 ton arbor press and it was really tough to remove! I did some polishing of the shaft in my lathe afterwards and there are some pretty big scars left (picture shown), but the bushing is free now. However, I don't want this to happen again. I texted a machinist friend today and he said it was lack of lubrication. I did lubricate before I reassembled the first time, but I was running the engine pretty hard. Clearly I need to do something to solve this, so with my friend's advice I've ordered a Needle Roller Bearing, part # HK0509, with 5mm ID, 9mm OD, 9mm width. Next up, I will have to make an all new bushing to fit this bearing. I do enjoy the challenge of doing stuff like this, but I keep thinking the project is almost done! My USA Trains Big Boy runs so smoothly and pulls so well (that's my reference point) and I just really want another steam locomotive that's comparable (but smaller and easier to carry!).

I have a couple of the MDC 2-bay hoppers, but never run them with my 1:29th scale. I don't have any Aristo-Craft 2-bay hoppers, but I could take a picture of them with something else. As I have an MTH GS-4 Warbaby on order, I'm hoping these hoppers will be useful as part of creating a buffer between 1:32 and 1:29 cars. Maybe these hoppers plus smaller 1:29th like flatcars or tank cars will work for these purposes.

Thanks guys. Chuck, yeah I never would have thought of turning the workpiece by hand with the mill running either, until I saw someone else do it! One of these days I'd like to upgrade my mill to CNC--it's just time and money!--although I'm also partly worried that CNC will make some operations more difficult. Today I removed the side rods, and without them the center (powered) drivers run perfectly smooth. But with the side rods off the other drivers roll smoothly by hand--at least somewhat so--but with the electrical pickups they don't exactly roll freely. And the side rods seem to fit well. But they must be binding slightly somehow, maybe as Chuck pointed out in combination with the pressure from the bottom "covers" being installed things have just tightened up a bit. I don't feel like working on this anymore right now, so I'm going to try and finish reassembling and installing the DCC decoder and see how it runs. Maybe it'll run well enough, and break-in with time as Sean suggests. The locomotive is so difficult to open that I've decided to put the decoder in the tender, with one 2" speaker in the boiler, and a 3" in the tender... only problem is that I forgot about a speakers until a week ago and they won't arrive until Monday. But today I did all the wiring to the tender, so I should have this wrapped up within the next few days. (I'm suddenly motivated to finish quickly because I'm hosting a garden railway open house a week from Sunday for a combination garden railway club and Ford Model "A" club tour. I was just thinking it would be nice to run something new that happens to be a Model A era locomotive.)

Thanks Ray, Chuck, and Joe! You guys all do so much work yourselves with upgrading, modifying, and kit-bashing equipment that you're an inspiration to tackle stuff like this. So this evening I started reassembly and I've got the chassis all back together--gearbox cover, electrical pickups, brake hangers/pads, trailing truck, axle box retaining plates, and lead weights. But then I tested it one more time and it didn't seem quite as smooth. I can't figure out why any of this would affect the motion, as none of this makes contact with the wheels except the electrical pickups and it seems like they should cause consistent drag. It's not that bad though... not compared to my Aristo-Craft Mallet and it still seems to work okay. But, I'm looking for all the performance I can get out of this one. Any ideas?

I knew it'd been awhile, but I can't believe it's been two months since I've updated this thread! I finally got back to making the new motor mount, and shortening the drive line, so now I'm ready to start reassembling... I did label most of the parts, but should have taken even more photos! Here are photos of making the new motor bracket, and the completed drive line running on analog DC. I started with a big piece of 1/8" steel. The original bracket seems to be brass, but that's more expensive and I want to add weight anyway. After some work on the bandsaw and a lot of milling on my LMS mini mill (light cuts in steel), I got the rough shape cut out: I did the round part as I often do on the mill, turning the part back and forth using vice grips holding the part over a 3/16" steel rod: I used a 4-jaw chuck on my lathe to hold the part for boring: I forgot to show the 1/8" slot I milled to make the bend easier to make, and in the right place, but I was so relieved when it worked. I did the same (but 1/4") for the second bend (as I had to drop the motor down to keep the same axial alignment). Here's the part test fit to the locomotive once I cut off the excess, drilled holes for the motor using the original part as a guide, and drilled and tapped 4-40 holes. The original is attached from the top, but I've drilled out the frame instead because it's far easier to attach this way. There wasn't room underneath for the screw heads to re-use the front two holes, so I've just used three screws instead. Finally, I had to shorten the driveline. Here's the original as I measured it. I found I only needed to cut off about 3/8" even though I moved the motor forward by more like 5/8", since there is a lot of play in the u-joint. and how far it slides onto the motor shaft. It was pretty easy to shorten on the lathe, and re-drill for the set screw in the mill. All assembled, it seems to run pretty smoothly:

Are you bending with the rail joiners connected? I've tried to do most of my bending in place, with the Split Jaw clamps connected to the previous piece. I'm on concrete roadbed, it's a great base for aggressive work with the rail-bender. But when I've pre-bent track I've done it on a smooth concrete patio, with a short extra piece of track attached so I can bend all the way to the end. It's extra effort to connect a piece of track just for bending, but the only way I've found to get the radius just right. (I've also bought hundreds of extra 6-32 stainless screws for the Split Jaw joiners--used to be cheap from Enco, not sure where to get them now.)

I'm not familiar with this bender, but I have a lot of experience bending Aristo-Craft stainless using the Train-Li rail-bender. Yes, it's tough stuff, and requires over-bending quite a bit sometimes to get a curve to stay. And I bought more than half my rail used, which was even more difficult to straighten and then bend. It seemed like some of it was much more resistant to bending. But perhaps the biggest annoyance is that a lot of the new rail was twisted slightly. I could check a single piece of rail on a flat smooth table, and it would have a slight wobble. In order for the rail-clamps to align perfectly I found it helpful to straighten the rail, which required putting one end in a vise and using vise grips to twist the other end around. A 180 degrees seemed to be about where I had to start to get it to deform slightly. I also noticed in your photos that the curve isn't consistent all the way through, which I also found to be a problem. I found that sometimes you have to adjust the rail-bender along a curve to make the curve consistent. It has a mind of it's own. I used plywood guides for 20' diameter to measure (although many of my curves are more freeform, so just looking right was more important) and I didn't rely on the rail-bender settings at all. But after all the pain, it's well worth having stainless! It's really durable and it's great for its lack of oxidation when you have power through the rails.

Thanks for the insight, Ray. But as for getting the word out, I would have to say that my experience in my local club is very different. Most of the members are not in any large scale online community, and generally don't know about anything unless it was in Garden Railways or mentioned in our club newsletter. A few members get out to large scale shows or events like the NGRC, but MTH hasn't had a booth in recent years. And even then, it may take persistence and bold advertising to reach some. A couple friends went to a recent open house at Gold Coast Station where Charles Ro was present, and supposedly talking about their upcoming FEF-3 and passenger cars, and yet my friends came back unaware of these new items. I really think a significant percentage of people in the hobby haven't heard of the MTH 2017 catalog. That said, maybe this doesn't matter, because the same people in my local club who have no idea what's going on unless it comes to them in print also wouldn't likely buy anything!

That's great news, I've been waiting in anticipation this month to see if MTH changes the delivery date for the GS-4... maybe there's a chance that it really will come soon. Ray, do you have any idea what the production volume like will be from the 2017 catalog? It can't just be preorders, can it? It doesn't seem to me like they've advertised the 2017 catalog too much (e.g. not in Garden Railways Magazine at all). I might be interested in some MTH freight cars including a SP caboose 70-77036... but I'd like to get the GS-4's first and see how it looks to pull 1:29th scale freight cars with a Warbaby. I'm hoping with some smaller cars up first it'll look okay. (When I watch real train movies from the era the cars vary so much it looks some of out of scale!)

If only everything was the same scale, I'd have bought USA Trains Daylight cars. But, USA Trains never made a GS-4, and Accucraft and MTH did, and the difference is size between 1:29 and 1:32 is huge. I don't have any USA Trains cars to compare, but I do have almost exclusively 1:29th equipment and the 1:32 Accucraft cars look tiny by comparison. The Accucraft cars do roll really well (too well, actually--I can't leave them uncoupled on an almost level siding without them rolling away), and the paint colors are perfect, but they aren't very detailed inside or out, and with no lights or passengers. USA Trains is a far better value. But between MTH and Accucraft, I spent 2x the money for Accucraft because they have more car types (e.g. baggage car), they are the proper length, and the paint colors are better. Hopefully the cars are something Accucraft will re-run, because I think there is a need for 1:32nd passenger cars.

Wow, well this is an exciting announcement! It's great to finally see an official ad, and now finally be able to see the paint scheme variations and pricing. I was really hoping for a few hundred less, but I guess the bigger surprise is that the Big Boy hasn't gone up in price in awhile. I suppose I could pass on this, but of course I want one and I do like to vote with my wallet: more 1:29th scale steam locomotives is exactly what I want to see. I'm excited about the new heavyweight cars too. Not surprisingly, USA Trains is doing it right: fully detailed cars with LED lighting and figures.

Hi Doug, If you just want to run the train all by itself on a loop of track, you just need a basic DC power pack. A low powered one should be pretty inexpensive. For workbench testing of a new engine I use an ancient MRC Tech II from the 80's--a very popular model that I often see for sale at train shows for $5 or $10. It's not very powerful (meant for HO scale), and the DC output is just rectified AC so it's not clean power, but it'll work in a pinch. If you're serious about making an investment in the hobby, you'll want to either get a good DC power supply (look up Bridgewerks--they even have remote control units) or consider if you want to use a more advanced control system like DCC. Some Piko locomotives have built-in DCC decoders--you'll have to check yours. A locomotive with a DCC decoder will still run on regular DC, but may not run as smooth (especially on rectified AC from a cheap power pack), so we really would need to know details about your locomotive to make the best recommendation.

Sounds great, but I'm getting an error on the video: "We could not locate the item you are trying to view.". Could you post to YouTube and link that here? That's what I do as it's easy to post and easy to view.

Very nice. Looks like it's in pretty good shape! With all those trees (pine?), I can imagine the effort getting to this point!

Great thread idea. Chuck, I'll have to second your first item: Hakko soldering station An old Window desktop computer in the garage dedicated to trains/electronics projects (JMRI, QSI software, Arduino IDE, quick Google lookups, etc) Little Machine Shop Mini Mill (mostly for 7-1/2" gauge projects, but comes in handy for G-scale coupler adapters, etc) Logan 10" lathe (also mostly for 7-1/2" stuff, but I do use occasionally for large scale) Automatic wire stripper (it's easier to think of the bigger tools, but here's a small tool I sure appreciate)

After a nearly two week wait for the new motor (FedEx has seemed especially slow lately), followed by some other priorities, I'm back to working on the K4 for the moment. Today I did a test run with the new Pittman GM9236S020 motor, hooked up to the QSI Titan. Before going through a lot of effort to modify the locomotive and/or motor, I figured I should make sure I'm happy with the speed and power draw of this motor. I used a lot of electrical tape and rigged up quite a contraption, with my multimeter measuring the current. The test was mostly successful. I'm quite happy with the top speed (maybe not prototypical for a passenger train, but faster than I normally run trains). And the current draw was only about 1 amp, pulling 8 cars on level track. Although I'm not sure I'm measuring the amps accurately, considering PWM/BEMF, however I am using an RMS meter if that makes any difference. If I hold the locomotive back, the drivers just spin, pulling about 1.5 amps. If I also push down on the engine, I was able to see power draw up to 2.5 amps, while the wheels continued slipping. I'm sure I could force a stall if I really tried, but I put more pressure on the locomotive than any amount of weight I could possibly add! The problem today was that at low to moderate speeds the engine wasn't running smooth at all--worse than I've ever seen. I will have to put the engine back on the workbench on rollers and see if I can improve this. I have to admit that I really don't understand motor PID settings at all, so there is certainly room for improvement. It doesn't have to be perfect, but it has to be better than the start you see in my video! (My kids picked the cars--but they are more prototypical for the era than a boiler-less, cab-less steam locomotive!)

Hi Bill, I think you may be confusing DCC and DCS. They are very different and incompatible systems, although Protosound 3 decoders (or whatever they are called--I use DCC) are also compatible with DCC systems. Protosound 2 is not DCC compatible, so you will need to stick with MTH DCS to use them.

That's great Bill! I appreciate hearing stories like this. What's especially impressive to me is that they took the time to help you troubleshoot it so thoroughly over the phone and enable you to fix it yourself. A lot of companies today just want to you set it back without a second thought. Best of all, now you have some experience with electronics troubleshooting.

Congrats! Sorry they are out of the most important piece, but I'm sure it'll come back in stock very soon (within a month, probably less, based on my experience). Also, re-reading my post above, I have a few updates: 1) They've redesigned the USB interface so it's much better (and cheaper!) than the old one. However in the meantime I've been using their RS-232 serial interface which has also been great. 2) I was completely wrong about the issue I had back in early 2014... the CAN bus needs to have terminating resistors! That was the issue. From the docs I read at the time I was thinking the resisters were only necessary when using the CAN-CAB throttle, and that just between the USB interface and command station I could just wire the CANL and CANH terminals together. That is not the case and it's amazing that it worked for me anyway--up until open house day, anyway. 3) A third issue I also had in April 2014 was that trains were mysteriously coming to a stop even with plenty voltage on the track. The issue was my DCC signal was getting corrupted so far away from the booster. There's a problem with inductance when you have two parallel rails carrying a digital signal. I had just barely finished my mainline in April 2014 and more than tripling the length of my track brought this issue to the surface. I did three things to solve it: I installed three MERG boosters in weatherproof enclosures in the yard, close to the tracks, I added more jumpers to the track, and I twisted the wires to minimize inductance. I haven't had any other issues. The system has been working perfectly for almost three years now. But I have an extra command station on hand just in case, and I use it for my programming track. I also bought a CAN-CAB recently and have used in on my workbench and also to run HO trains.