Rivarossi Revival - Repowering Old Steam with New Motors

[wm3798]

We all know I have become enamored of relics from the Dark Ages of N scale.  I suppose it can be traced back to the rebuild I did some years ago to turn a Rivarossi Pacific into a Western Maryland K-2 http://www.wmrywesternlines.net/loc_steam.php


In that episode, I kept the original motor, but upgraded it to DCC, super detailed the boiler and improved the lighting.

Now, having retreated to my N scale fetal position with DC throttles and simple machines, I'm keeping the model more or less intact, but with new power flowing to the drive train.

Through the past year or so, I've accumulated quite a few of these old dinosaurs, some that run well, and some that didn't.  A few had fatal Zamac disease, but have been good for parts donors for some of the others.  So now I have on my workbench three examples that I plan to repower.  A light Pacific, a Mikado, and an 0-8-0 switcher.

Each has been disassembled to the point where I could verify that the only thing wrong with them was the motor.  The frames are sound, the gearing is fine, and the mechanisms turn effortlessly.  So the next step will be to install a new power plant.



If you're looking for a description of how to elevate a 50 year old model to the state of the art, you'll probably want to look elsewhere.  But if you have one of these laying in a drawer for want of a good motor, and you just want to run it, then you might have something here to follow along.

Progress will likely be sporadic, but I'll try to document the process as I go along. 

[wm3798]

On the recommendation of Mad Max @mmagliaro , I'm starting with these:
https://www.ebay.com/itm/Minebea-10mm-Square-DC-Motor-DC6V-9V-7-2V-11700RPM-6-Pole-Rotor-Large-Torque-Toy/323989254916?_trkparms=ispr%3D1&hash=item4b6f43b304:g:U0cAAOSwTfhZjbPW&amdata=enc%3AAQAFAAACcBaobrjLl8XobRIiIML1V4Imu%252Fn%252BzU5L90Z278x5ickk5v8gVt3hEWLVg%252F253w6XCZMagnAFuvDI3elQ3ik5mIuKyHiSmgxMDw12Ahy8nNzmNoBGGKoGKpDRfOeYC47Bl7AUPd5N0nbp9v9OOo9miuCkuDia71xQo0kGmaJH4q3aNKh5ZDEluOMO4k%252BLGej%252B6p%252BuIa8%252B9opS1hj3BNnhanjm%252BPf24iK5pGPnPZ8EdeAX5yEKxSvXSl3hjQIwD3%252Fow6%252Bwq6ybTv3iyeyObTBIl%252FAJHmDhY%252Fwm4t%252FVRFfGHbdausr2wJzxs388%252BMv6Y9WfY41qKyYHoKtBlvSR86qtJ0QmrbdEb8TwwGBn8fRU%252BXm9Csc10eny7a1ez5BFPHKk%252BWLfE6%252FScQzo9UGo6NRM04zX2VH5c2x7sAABSGpjW1GId5Edc8ZvAnxKHUVsFnu0%252B6ZFIJJCIQDfRJQBoR2O6JSm9TVUS5gDWvTZhWElUOhvJf28UO9aHRQ%252BEt7getdRAQ5%252BEK4z0vO7lJXJenjWZtucCzB9pzQrTZa9s9XA3t30Meren20Nubbs9gCB%252F0FZkXk%252FhYMKaggvoXJKk2r66ZlRf%252BWG0xvmu%252B8is4fC6sQFCyfIKv%252Fvx9%252Bzah2EpW%252Bd1AFsX3zKyUeHOIe8q8wLAqlum35lqywrN6S1gprVl7B69fhRnE7egH%252BFTBy2bx9nY7pI4hCW%252F4qeXMPKVfk9cSRBgFicAGJOPqQO10thnH4cOP1wVHpfuvQTMSzuuYv8pwZodN51P85P%252BGbguj0gYK8AfsJRVtNPjfnJPA4DRYK6kY0SEOSA1RAXSfckAs2FNw%253D%253D%7Ccksum%3A323989254916a7e6883d149a41c2b0db0ebe426f6c86%7Campid%3APL_CLK%7Cclp%3A2334524

The Minebea 10mm motor is rated for 3-9v, but is high torque, which will be ideal for moving all those siderods, brass gears, and high friction Minitrix and Roco wheelsets.  As you can see, they are also dirt cheap, which as you know, is my favorite color. 
There are three hurdles I'll need to clear as I undertake this experiment.

• The drive shaft in the motor is only 1mm, so I'll have to make a collar big enough to hold the OEM worm gears and universals.
  The 10mm case is square, the old Rivarossi mount is round... but fortunately much bigger.
  and Current from the tender is supplied directly to the old motor via the bottom pole of the motor itself, which also serves to receive the drawbar, making it a critical mechanical connection as well as an electrical one.

Since the concept is still in its larval stages, I have some ideas of how I might tackle these three issues, but I have no clue whether they'll actually work until the work commences.  Regarding the first issue, Max kindly supplied me with a short length of stainless tubing with a 1mm (+/-) inside diameter, so that will just be a matter of cutting and fitting.  The other two issues will have to be addressed once I'm confident that I haven't broken the motor by putting on the collar.

A fourth hurdle, of course, is the motor voltage maxing out at 9v, when typical power pack output is going to land between 12 and 14v.  Max provided me with a diagram for wiring in some diodes to help step that down, and I'm hoping he replies here and shares that.
So with that, I'll get back to work, and ask you to look forward to the progress reports as they take place.

Lee

[wm3798]

So I pulled the crappy rubber magnet out of the Rivarossi can to see if there's some hope of making it work as a sleeve for the new motor.  Afterall, if I can figure out how to get the motor secured in the old can, securing THAT to the frame should be easy.

Eureka!
 






We're close... still no cigar, though.

I'm hoping I can figure a way to modify the plastic cap at the back, because the motor has screw mounts at the front of it, and they almost align with the vent holes on the front of the can.  My thought is that with a little judicious drilling and/or filing, I can get the motor firmly anchored to the can, which would in turn firmly anchor the combined unit to the locomotive.  Then it's just the riddle of mating the worm gear to the shaft, and the power pick ups to the terminals on the motor... 

I wish I was a little more fluent in 3-D Printing...  There's probably a way to print the exact parts I would need for a secure motor mount...  Any thoughts?

Lee

[wm3798]

Hmm.  Only a partial solution.  I've got to get the business end of the motor further into the frame... too much motor hanging out the back.  Maybe I preserve the backhead from the old motor, and cut the front of the can off.  That will have to wait till next time.  I really need to get some work done!!

Lee

[wm3798]

Really close.  It appears that with a little bit of fine motor skills with a fine motor tool, I can square up the circle just a bit to get the back of the motor to snug into the plastic cap.  The surgery should be minimally invasive, and still enable me to use the original mounting screws, as well as the original pick up terminals, since I can align the motor terminals to be in direct line with them.  Using the OEM sprung brush caps, I can probably make the connection with a simple press fit contact, wireless and solderless!  But I'll have to nibble away that circle very precisely to make the fit snug enough.

I may end up attaching the back head with some E6000, which will bond it securely, but being a rubber cement, I should be able to pry it off if there's a problem...  Now seriously.  Back to work.  That guy wants to file for permits tomorrow.  

Lee

[mmagliaro]

Here's the zener picture.


The zener voltages can be whatever you want to drop.  So in the case of the Minebea motor, with a limit of 9v, if you want to drop 3 volts, you might choose a 3.3v zener.  Zeners come in a very wide, versatile range of voltages.  You can get 2.9, 3.3, 3.6, 3.9, 4.2, etc.

Next, you have to worry about power handling.
Using the P = V x I   (P = power in watts, V = voltage, I = current in amps).
The Minebea motors draw a maximum, I think, of abou 80 mA under heavy load.
If you use 3.3v zeners, we have:
P = 3.3 x .080 = 0.264 watts.  So a 1/2 watt (0.5 watt) zener would be plenty big enough.  1 watt devices are also pretty small and are easy to find, and I'm sure you could find the room for one somewhere in that Rivarossi frame.

The idea is simply that zeners conduct only in one direction, like a traditional diode, BUT, they break-over and conduct in the opposite direction once you get over the "zener voltage".  And after that, they drop that zener voltage as a fixed amount.  So you put two of them back-to-back, as I showed, and then in forward or reverse, you will get a fixed drop of 3.3 (or 6.2, or whatever) from the "reverse" zener, and another 0.7 or so from the "forward" zener (because they still drop about 0.7 in the forward direction like a conventional silicon diode.   So in this case, if you want to get as close to a 3v drop as possible, you might want to go with 2.4v zeners (the lowest commonly available value) which would get you 2.4 + 0.7 = about 3.1 volts total drop.

This does mean that as you turn up your throttle, the engine gets nothing until you get over the diode voltage, so you do lose some of your control range.  If your throttle puts out 12V at full-dial, then a 3.1v diode drop, will cause you to only have about 3/4 of usable range on the dial.  But that's not too oppressive, considering that in most cases, we can never use the upper 1/4 of a throttle's speed range anyway because the engines are running like drag racers at that point.  This is NOT DCC where you can fix the upper limit and program speed steps.  Having said that, if you wanted to use zeners in a DCC engine, it ought to work fine, but it would be easier to just program the decoder to not output a voltage higher than 9 (or whatever) to the motor.

[peteski]

. . .
This is NOT DCC where you can fix the upper limit and program speed steps.  Having said that, if you wanted to use zeners in a DCC engine, it ought to work fine, but it would be easier to just program the decoder to not output a voltage higher than 9 (or whatever) to the motor.

Since DCC uses PWM circuitry for running the motor, even if you reduce the maximum voltage (by setting the specific CV value), the motor driver still subjects the motor to full DCC voltage pulses (12V). I would be leery of using a lower voltage motor in that situation.  Actually, the Zener diode circuit would be ideal to use for running low-voltage motors under DCC decoder control.

Actually, since here we are specifically discussing DC powered models, none of the above is relevant, but since you brought up DCC, I thought I should mention how DCC decoders drive motors.

[mmagliaro]

Since DCC uses PWM circuitry for running the motor, even if you reduce the maximum voltage (by setting the specific CV value), the motor driver still subjects the motor to full DCC voltage pulses (12V). I would be leery of using a lower voltage motor in that situation.  Actually, the Zener diode circuit would be ideal to use for running low-voltage motors under DCC decoder control.

Actually, since here we are specifically discussing DC powered models, none of the above is relevant, but since you brought up DCC, I thought I should mention how DCC decoders drive motors.

True, I overlooked the fact that the DCC decoder is just a "PWM throttle".   Yep, you are right.  You would want to put the zeners in there. 

[nickelplate759]

Since DCC uses PWM circuitry for running the motor, even if you reduce the maximum voltage (by setting the specific CV value), the motor driver still subjects the motor to full DCC voltage pulses (12V). I would be leery of using a lower voltage motor in that situation.  Actually, the Zener diode circuit would be ideal to use for running low-voltage motors under DCC decoder control.

Actually, since here we are specifically discussing DC powered models, none of the above is relevant, but since you brought up DCC, I thought I should mention how DCC decoders drive motors.

I'll bet that the motor doesn't care all that much about peak voltage being more than the rated voltage but does care about average (however that is calculated for PWM) - unless the peak is much, much more than the rated 9V. 

[wm3798]

Okay so one of my Retro Bros on the face toobs recommends this motor as an almost direct swap.
https://www.ebay.com/itm/DC-6V-12V-27000RPM-High-Speed-Mini-DC-Motor-DIY-RC-Slot-Car-Drone-Toy-Model-/192873670885?hash=item2ce82aa8e5&fbclid=IwAR1jXmt4DKtxc1WfI_EQ2-XWZ2ECnVmfBN2V52ZXH01kd6guWWQnyJBVQb0

12v rated, 2mm shaft, so definitely less fiddling to do with it, but open frame, skew wound... looks a lot like the later motors that Rivarossi used during the Con Cor era.

Any thoughts?  They're certainly cheap enough.  I might order a couple for the next batch of projects.  (I might have all of the current batch done by the time the slow boat gets here from China...

Lee

[Chris333]

This is probably your square motor:
http://www.narrow-garage.com/SE15HORTYP.html

[peteski]

This is probably your square motor:
http://www.narrow-garage.com/SE15HORTYP.html

Lee's motor is a Mineba, part# SS10G0RTME.  Similar, but different. 

[randgust]

Mildly skeptical but interested to follow anyway.

Torque is relative and you may only get that at peak RPM, basically, until you try it, you just won't know.

One of the dumber things to do, but it's effective, is to just take the raw motor, hook it up to power with an ammeter on it, and let it run in various speed ranges and brake the shaft with your fingers.    See if the motor actually heats up.

I just ordered some 6mm motors with 5.14 gearhead reduction off of Amazon, what a dissapointment.   Man, did they run hot even at 3.5V as soon as subjected to load.   Glad I didn't pay much.   

Motor efficiency is hard to measure except by field testing.   The old Rivarossi motors usually predictably failed by the rubber magnets gumming up, increasing load, creating heat, and the rear bearing would actually melt out against that plastic cap, and it's done like a dinner and seizes.   I've got a couple of scrap Rivarossi motors out of my D16sb project if you need them, I think both might still even work.

The other thing is - if I remember right - the Rivarossi motor goes in that plastic universal "u", which could be drilled bigger rather easily, so a motor with a larger replacement shaft wouldn't be a problem.   I do like your idea of using the old shell as a mount though, that's pretty cool.   I think in the same situation I'd try to shim up the inside of the old metal case with glued styrene so that the entire thing was just press-fit into the old round part and similarly shim/tape the case to press-fit into the frame rather than relying on the screws in the back of the black plastic cap.  As long as you don't have to mill the darn frame to fit something, it should work.

Years ago I saw an epic rebuild of an Atlas 0-8-0 with a Faulhaber gearhead, the top speed was about 15mph with the motor wide open, and the back of it was almost hanging into the tender so the owner had put on big honking cab curtains.   The nicest tiny and inexpensive five-pole open frames I've found so far were out of the original Kato 11-103 chassis (the fixed frame four-wheel one), which may be the same motor as in the NW2, haven't checked.  Great motor though, robust and good performer.  Buy chassis for motor and throw out chassis.

The tiniest 12V motor I have any confidence in is the one that Kato puts in the 11-105/6/7 chassis (metal closed case).   It's really small, native 12V, and high RPM, but surprisingly efficient under load; i.e. doesn't tend to overheat at low RPM and higher load.  Same as the 11-103, buy chassis for the motor if you have to.  That case dimension has become fairly common, Bachmann's 44-tonner had a double-shafted version available for a while as well.

[sirenwerks]

I've been meaning to ask, Lee, I have an old Minitrix K4, loco only/no tender./box  It's been forever since I ran it, so I have no idea what it's doing these days, but if you want it, it's yours (as soon as I find it).


Bryan

[peteski]

Mildly skeptical but interested to follow anyway.

Torque is relative and you may only get that at peak RPM, basically, until you try it, you just won't know.

Randy, you have that backwards.  For permanent-magnet electric brushed DC motors the maximum torque is at stall (not at peak RPMs).  When the motor is stalled, there is no BEMF being generated, so the maximum current flows through the winding, creasing maximum magnetic field (thus maximum torque).

I have those 6-pole motors Max recommend to Lee. Actually I found those on eBay and pointed Max towards them. He got some and was very impressed by their torque.  While I don't have any measured results, I can tell you that for what they are (standard motors with iron-core rotor), then have very impressive torque.  My test is my calibrated fingers pinching the spinning shaft.     They really are impressive.