Replacement Atlas N Scale Motor Universal Connectors

Atlas make lots of railroad locomotives and rolling stock in a variety of scales, and I have several of them because of their quality and they enable me to use the chassis for other builds.  In particular, I use their C-628 and C-630 N Scale models as the donor chassis for my N Scale DT6-6-2000 and RT624 kits.  The chassis has been revised over the years to make improvements, but one version has an issue with the driveshaft coupling to the motor failing.  In this post I’ll share with you my fix.

Below are a pair of Monon C-628s; the rear one is actually a dummy using my 3D printed chassis kit.

The powered chassis is a standard design, used on many of Atlas’s N scale locos, with a central motor and flywheels.

The chassis is held together by the two screws near each end, and the fuel tank, which clips over both chassis halves.

Inside at each end is a driveshaft linking the flywheel and the worm gear, which drives the truck towers. These simply pull out.

The motor is clipped in a cradle which in turn is clipped into the chassis.

With the motor removed you can see inside the flywheel; there’s a plastic universal joint, and it’s cracked.

The universal joint is press-fitted over the axel and uses the friction to spin it with the flywheel.  Even when cracked it’ll spin so the loco will probably run okay on its own.  But as the load is increased, such as adding a train, the amount of force on the split universal is stronger than the friction, and the axel just spins.  So if your loco seems to run okay, but won’t pull very much, this is most likely why.

The universal joint is a plastic tube with two pegs which fit into the driveshaft.  The hole in the tube will be smaller than the axel to create the required tight fit but the constant pressure on this particular material causes it to crack.

Replacement universal joints are available from Atlas, but these have been known to fail as well.  So I’ve 3D modeled the part and printed it in Shapeways Smooth Fine Detail plastic because it’s both accurate and also hard-wearing

The new part is a direct replacement for the original.

If the old universal is cracked it should simply pull off leaving a clear axel inside the flywheel.

I fitted the new universal by placing it with a pair of tweezers but not pushing it on fully, just enough to hold it in place.  If it’s pushed at an angle it too may crack.

I then used a flat screwdriver, as to give even pressure, to push it on fully so the universal is all the way to the back of the flywheel.

And that’s it.  The loco is ready to be reassembled.

These are now available in packs of two and four using the links below:

2x Replacement Atlas N Scale Motor Universal Connectors

4x Replacement Atlas N Scale Motor Universal Connectors

This universal is used in many of Atlas’ diesel locomotives and will fit all.

I’m juggling an HO project as well as testing recent 3D printed replacement parts, but my focus is on returning to work on customer’s layouts where possible, so who knows what I’ll be sharing next week!

A Baldwin DT6-6-2000 in HO – Trucks Part 7

Two weeks ago I shared with you the first 3D test print for my new HO scale DT6-6-2000, you can find the post here.  As you may recall I had a few issues with the print, but this week the second test print arrived and the results are very good.

I reprinted the two truck center halves and the four gears as all had issues to be resolved.  The final kit will need four truck center halves and eight gears.

The gears came out even crisper this time, I assume because the axel section of the gear is now a part of the gear rather than held in place by 3D print residue.

The gears up close are very accurate, but there’s still some residue on the axel which will need cleaning off. To do this I simply wipe it off with paper towel. The holes for all the gears will also need cleaning out and for this I use a drill to ream each hole.  If it has 3D print residue in the hole it will cause the gear to bind and add drag to the motor.  If all the holes bind it may even jamb up the gears.

This time everything fitted as planned.  The clip covering the worm gear will need to be taken off, and the drive shaft removed, in order to fit it into the chassis.

The gears, as shown below, are simply positioned in place to check they are a good fit.  Before I tested them under power from the motor I did all the cleaning mentioned above, and lubricated all the axel holes with light oil, from inside and out.  Then, with the wheelsets removed, I checked that the sets of gears turned freely by running my finger along them.  With the wheelsets refitted the whole assembly was a little tougher to turn, but this was expected given there are 3 more gears than in the original configuration.  But once the driveshaft and worm gear was fitted I was able to turn the whole assembly by hand.

The test fit into the chassis was also good; below you can see the original truck on the left and the new one on the right now facing the right way for a DT6-6-2000 or RT-624.

The clip covering the worm gear holds the truck in place, and a pin in the underside of the chassis fits into a hole at the center of the truck, creating the pivot point.

The white of the truck center is barely visible through the truck side frames, but it can be painted, taking caution not to paint the axel holes.  With the chassis reassembled, albeit with one new and one old truck, I lubricated the gears using oil designed for plastic gears from LaBelle. You can find a post about these products here.

The final thing to do was test the new truck, so I connected a basic DC controller to the chassis (there is no DCC chip fitted yet) to see how well it runs.

In the video below you can see I ran the chassis at full throttle in one direction then threw it into the other.  Because of the two large flywheels, the direction change is not instant, but it’s still a heavy load on all the gears and they seem unaffected.  The chassis is in a foam holder because the first time I did this the sudden reverse toppled the chassis over!

The new truck ran very smoothly and the slow speed looks okay, but the real test will be when it’s on some tracks.  As I’m an N scale modeler I only have a short section of HO track at home, but later this week I’ll be able to test it on a layout.  The chassis has an 8 pin DCC socket so I can plug in a chip and see how slow I can make it crawl, but I feel confident that I’ve resolved the issue with rotating the trucks, so now it’s just a matter of finishing the shell.  I’m currently modeling the cab interior and I will share that with you when it’s finished and before we go to test print.

Replica Railways/Bachmann OO Locomotive Split Axle Repair

In January of 2019, I shared with you my design for a 3D printed axle set to fix Mainline steam locomotives with the large drive gear on the center axle.  You can find the post here.

The manufacturer Mainline was incorporated into the Bachmann company and several of the designs stayed the same.  These are what I call the large center gear models, and have the square shafts on the wheels, such as this modified Hall locomotive below.

But another popular brand which Bachmann also incorporated was Replica Railways.  These locomotives, such as the LNER B1 below, have a similar design to the Mainline locomotives in that they are a split chassis design without wires, but the main difference is the motor and the power transmission to the wheels.

These have a totally different motor that drives the rear axle via a smaller gear than the Mainline locos.  The side rods then drive the forward wheelsets which have ungeared axles.

Along with the rear gear being smaller, so are the axles, and sadly the wheels do not have square shafts.  This causes the wheels to get out of quarter when the axles split and everything jams up.  These are also made from the same material as the Mainline ones and it’s very common for them to crack and split.

This particular model had at some point been repaired in the past and the standard method for this was to glue the damaged axle in place. This means when the axle finally breaks down completely there is glue in the wheel which prevents a new axle being fitted.  In the photo below you can see the pocket around the center stub is full of glue.

To get this out I drill several holes in the glue until it starts to break up and can be pulled out with tweezers.  This also depends on the type of glue that was used.

Below you can see the holes I started drilling in the glue.  Make sure you don’t go too deep and start drilling into the wheel. The way to check is to look at the swarf coming out.  This particular glue was fairly transparent so as soon as a fleck of silver came out I knew I was as the bottom of the pocket.

Once all the glue is removed you should have something which looks like this.  The rear axle, which has the geared axle, usually has traction tires.  The center wheels have a larger connection for the side rods as this is also where the piston rod and eccentric rod attach.

The three axles have been drawn up and printed in my usual material, Shapeways Smooth Fine Detail Plastic.  I’ve put them on a sprew, which doesn’t actually touch the parts, for ease of printing.

Because these axles have round holes it means unless the wheel is a good fit it can easily rotate inside the axle, but if it’s too tight the wheel will split the new axle, just like the original.  Because the Mainline axles have square holes this is less of an issue.  I’ve purposely drawn the hole in the axles ever so slightly smaller than the wheel center stub.  I use a drill to carefully ream out the hole because, as I have shown in other posts, there will always be 3D print residue in the hole and it needs to be a perfect fit.

The axle can then be pressed onto the wheel; it needs to go all the way into the pocket.

The corresponding wheel can also be fitted but at this stage, it’s vital to get the quartering right.  This means one wheel needs to be rotated 45° to the other.  Each wheelset must be set in the same configuration.

All three wheelsets can then be replaced in the chassis.

Often at this point is I get asked which way round all the rods connect, so I’ll run through the sequence.  Below you can see the simple side rod connects the center wheel to the front wheel.  The joining side rod, shaped to represent a side rod joint, connects the rear wheel to the center.

The joining side rod is connected first, hooking over the connecting peg.

Second is the simple side rod sitting on top of the joining side rod.

Third is the spacer, with the grooved side facing up.

Fourth is the piston rod, which fits over the spacer.

And lastly, the eccentric rod fixes into the wheel with the special screw which holds it all together.

Testing the loco is always, hopefully, a happy moment and I like to test motion before the loco is put on the track.  Below is a video of the first test.  The slow speed is a little lumpy partly due to the motor but this went away once it had done a few laps of a test track.

Because the hole in the axles needs to be such a good fit and I’ve come across some variations in the size of the wheel pegs, I’ve provided a spare axle with the set so you can use this to check and see if it fits or whether it needs enlarging.

The axle and gear set can be found here.

Despite being harder to install than the Mainline kit, mainly due to the quartering issue, these do work well as a repair and I’ve fixed many damaged Replica Railways locomotives which, without the part, would be redundant.

I have some more 3D printed gears in my latest test print and once I’ve installed and tested those I’ll share them with you as well.

A Baldwin DT6-6-2000 in HO – Trucks Part 6

Last Friday I had a delivery from Shapeways containing the new truck centers and gears for the HO Baldwin DT6-6-2000 project.  They look good, and although there were a few issues, I thought I would share with you how they came out and what was wrong, rather than order some more and just show you the finished item.

I usually order several parts for different projects at the same time, with some of them combined on a sprew.  Below you can see the truck center frames and lots of parts including the four new gears; they are located to the left of the sprew.

The four new gears are small and I didn’t want any parts of the sprew to touch them as it would need to be cut off, leaving a rough surface.  So I surrounded them in a cage, which worked well.

The gears were free to move about, but couldn’t fall out.

Close up, the new gear is a good match for the original; the new gear is still covered in 3D print residue, which is why it looks a bit fuzzy.

The truck centers came out very well and appeared to be a direct replica of the original, with the desired changes.

The first problem came when I started fitting the gears into the holes.   Each gear has a shoulder and an axel on each side.  The axel fits into the hole and the shoulder acts as a spacer to position the gear in the center of the frame.

But in the 3D model, I’d forgotten to make the shoulder part of the gear, so it was 3D printed as a separate part which just happened to be close enough that the support material held it in place.

The axel and shoulder simply come off and you can see all the support material in the middle of the gear.  The same thing happens on both sides so you end up with a flat gear with a hole in it.

Thankfully, this is why we print test pieces, and I was able to quickly fix the 3D model, so next time I print these gears they’ll be one whole piece.  Sadly, for now, it means I can’t test all the new gears in the truck center.  Interestingly though, once I had cleaned up all of the 3D print residue, the shoulder fitted into the hole in the gear so precisely I did wonder if I could make it work, but that made the axels too short and they wouldn’t stay in the right place. It does go to show how precise the 3D printer is.

The next test was the driveshaft and worm gear.  These have brass bearings either side of the worm gear which clip or slot into the top of the truck center, and it fitted well allowing for good free movement, but without clamping the worm gear.

As you can see below I did try and fit it all together with the gears, and they did turn before they fell out.  The big problem here is with the tubes which stick out from either side of the truck centers.  The ones on the left are longer than the ones on the right.  These tubes are the fixing locations for the truck side frames which hold the wheels and power pickups in place.  One is longer than the other, because the shorter of the two also clamps the power pickups in place.

But as you can see below, I got these round the wrong way.  The power pickup is clamped by the longer of the two tubes causing the truck side frames to flare out.  And this made it impossible to properly test the trucks.

This error has also been fixed in the 3D model and a new set has now been ordered.  To be fair, apart from the issue with the gears and the tubes, the truck centers came out very well, and with the corrections made, I feel we shall have a decent working truck with the asymmetric axels positioned the right way round.  In the photo below you can see on the left truck, even though the side frames are flaring, the wheels are all in the right place.

These truck centers will work for both the Commonwealth and Tri-Mount Trucks, so they can be used for either the DT6-6-2000 or the RT-624 models.  While the new truck centers and gears are being reprinted, I’ll finish off the 3D model of the shell, and I plan to share that with you next week.

A Baldwin DT6-6-2000 in HO – Body Shell Part 5

This week I’ve been looking at ways to improve upon the progress I’ve made with the Baldwin DT6-6-2000 HO Scale body shell.  With the main bodywork done, it’s attention to all the little details which will greatly improve the model, making them as best as they can be for HO scale.

This particular model is based on Santa Fe’s 2602 as pictured below; this is the N Scale version built by Dirk Jan Blikkendaal.

The 2602 had two single horns which I 3D printed as part of the shell, however several people used the kit to represent other railroads’ DT6-6-2000s with different horns.  So I’ve made the horns separate.  I intend to provide a choice of horns that will fit into the same mounting holes.  This will also make them less susceptible to damage before the model is finished and on the railroad.

For the headlight, I looked at several different ways but I’ve decided to use a simple 3mm warm white LED.  I’ve recessed the headlight surround to allow for a clear lens with a diameter of 4.5mm to be fitted if required.

Inside the headlight, I’ve designed the right shape to receive the LED from inside the shell.

With the LED fitted in as far as it will go, and with the flat spot indicating the negative connection at the top, the front of the LED will be in the right place to allow a lens to be fitted.  My only concern is the light will also travel through the body as the material is translucent.  So it will be important to paint the inside of the model and inside the headlight with a matt black prior to fitting the LED.  I do this on my other N scale locos as well.

The last detail, for this week, is the windscreen wipers in the main windows.  On the real locomotive these are very small but so are the windows.  In order to model these in HO Scale, they will need to be etched in brass.

The actual wiper will be a simple shape with a triangular stop to prevent it from sliding too far into the locating hole.

To ensure the wiper will be in the right place and securely mounted, the locating hole is rotated to the right angle.

I still have a few details that I want to improve or add, but then I think the model will be ready for a test print.  But for now, here’s how it looks.

The Baldwin RT-624, which is the successor to the DT6-6-2000, will also be getting these updates and improvements once I know they work.  The 3D printed parts for the trucks and gears from my earlier post are nearly ready and hopefully will be sent out later this week.  I’m looking forward to testing them and getting the right trucks on the chassis which I will share with you as soon as I can.

A Baldwin DT6-6-2000 in HO – Body Shell Part 4

This week I’ve some more progress to share with you on my HO scale DT6-6-2000 project.  I’ve been making lots of small changes, many internally which are hard to see, but also several to the exterior.

The body shell model, as shown below, now has all of its brass handrails positioned, along with their corresponding locating holes.  Unlike the N Scale version, I will only be producing brass handrails for this locomotive because a 3D printed set, at the correct scale thickness, will simply be too weak and they’ll break very easily.  The brass will be strong and when fixed into the mounting holes should stand up to the little knocks and bumps all of our trains accidentally get.

Last week I left you with my idea of adding the Preci Models DCC auto uncouplers directly to the locomotive.  To make that work, I needed to figure out the connection between the chassis and the body shell, then how to mount the Preci motor.

Because DT6-6-2000 is longer than the donor C-630 chassis the coupling is moved forward, allowing the original coupling mount hole to be used only as a body shell fixing.  As you can see below, with half the body shell hidden, I’ve created a hole in the body shell above the chassis hole along with a cutout for an M3 nut.  The nut will drop into the cutout becoming captured and can be glued in place.

Then an M3 bolt can be used to secure the body shell to the chassis.  Because the chassis tab with the hole fits into a recess in the body shell, both will be in the right place.

With the body shell secured, and without the need to modify the chassis, I can now look at the couplers and the Preci motor.  The Preci motor can be mounted in many ways but normally it’s glued to the back of the Kedee coupling as shown below. (Pictures from http://www.precimodels.com).  This allows the actuating string to run parallel to the coupling to give a straight pull.

With anything like this, my first step is always to model in the parts, so here is the Preci motor.

Using the model I can then accurately position the motor, and, as you can see below, there’s not enough room between the coupling and the chassis to mount it.  I also find mounting the motor like this can be rather fiddly as you need to glue it in just the right place.

My solution is to mount the motor vertically in a 3D printed hole.  This has the advantage of correctly locating the motor horizontally and vertically.  The large hole doesn’t go all the way through, creating a pocket for the motor to sit in.

The motor would be fitted before the body shell is bolted on top of the chassis.

The space between the coupling and chassis is still tight, but as the rotating part of the motor is so small and will only have a thin string attached to it, there should be room.  Connecting the string should also be done before the chassis bolt is fitted for ease of access.

With a bit of luck, and mostly modeling time, I’m hoping to have the body shell modifications done by next week.  I also intend to make the horns separate parts to allow different horns to be fitted, which was a recommendation from a fellow modeler.  Currently, as shown below, the horns are 3D printed directly onto the shell.

I’ll also be adding brass windscreen wipers, a cab interior and routing for wires to give cab lights.  The N scale version had provision to add working headlights, but due to the size constraints, the marking/number board lights on the noses were not illuminated.  But I’m considering this for the HO version.  They will still be small but it’s not impossible.  I’ill share my progress with you next week.