Best Of The TP56/TP70 Kitbash thread N scale

[coosvalley]

Awesome. How slow will it go and keep running steady?..

[u18b]

Here is a link to the motors in question.

http://www.ebay.com/itm/262550362464?ssPageName=STRK:MEWAX:IT

In case the link dies one day, here is the title for words to search....
2PCS DC 1.5V~6V 18000RPM Micro silent K20 Motor Long Shaft Mini Solar DC Motor

$2.20 for 2 (like I said, a buck a piece).

If mine survives a weekend torture test of running a few hours with no heat build-up, I may buy more just to have.

[u18b]

Awesome. How slow will it go and keep running steady?..

I'm running analog right now.  Will install a decoder some time later.

So Analog, smooth DC, it smooths out around 23 sMPH +
Not great but not terrible.

Also, that is only with that one hunk of lead.   The weight could probably be distributed better.

And I won't know how much weight can be added until I move to some kind of test shell.


Might be better with decoder and DCC.

[narrowminded]

Very nice, Ron.  If you get me the resistance across the two motor terminals I might be able to hook you up with some 1/4 watt surface mount resistors (very tiny) that will fit with little to no additional space required, will handle the current without overheating, and will allow you to run a standard twelve volt controller.  That affords the convenience of having standard voltage supply but even more important, will also help when you try to get LEDs to work as they need 3 volts or so to light up and that may wind up with the lights not coming on until you're at warp speed.  The resistors will drop the voltage all of the way up the range for the motor but leave the lighting voltage at full track voltage.  Then try a Medvend low frequency PWM controller and you'll have the control you'd like PLUS... you can put a small capacitor or two in the lighting circuit and have near constant power lighting even when travelling slow.  And that controller will work nicely with any of your analog stuff.  We can talk about it if you'd like.   Peteski could probably help with this, too.

[u18b]

Very nice, Ron.  If you get me the resistance across the two motor terminals I might be able to hook you up with some 1/4 watt surface mount resistors (very tiny) that will fit with little to no additional space required, will handle the current without overheating, and will allow you to run a standard twelve volt controller.  That affords the convenience of having standard voltage supply but even more important, will also help when you try to get LEDs to work as they need 3 volts or so to light up and that may wind up with the lights not coming on until you're at warp speed.  The resistors will drop the voltage all of the way up the range for the motor but leave the lighting voltage at full track voltage.  Then try a Medvend low frequency PWM controller and you'll have the control you'd like PLUS... you can put a small capacitor or two in the lighting circuit and have near constant power lighting even when travelling slow.  And that controller will work nicely with any of your analog stuff.  We can talk about it if you'd like.   Peteski could probably help with this, too.

Good idea.   I'll check it out.

While I will almost certainly go with DCC, I want the design to be that someone who is analog could have one too.

[u18b]

As a VERY rough draft....

I laid the chassis on top of my TP56 drawing.

As you can see, with this motor there is lots of room for weight and a decoder.

[Dave V]

And don't forget about the GMD GMDH-3.

This thing really looks like it ought to have a red star with a yellow hammer and sickle on it.  Something about it screams "Soviet" to me even if it is GMD.

[u18b]

Resistance across the motor terminals is 14.3 ohms.

[u18b]

This thing really looks like it ought to have a red star with a yellow hammer and sickle on it.  Something about it screams "Soviet" to me even if it is GMD.

Dave, the blue version looks a lot better.
And it looks like you could pass a trimmed HT-C truck under it- meaning you could use this same chassis maybe.





I wish Aaron hadn't posted this thing.   If someone will print one, I might have to make one of these too!

[peteski]

This thing really looks like it ought to have a red star with a yellow hammer and sickle on it.  Something about it screams "Soviet" to me even if it is GMD.

That is because Soviets used to routinely copy/clone US designs.

Ron, I'm surprised that you didn't Beardenize the worm bearings. I would have just used the outboard worm bearing and the motor bearing on the other side would be more than adequate. The inside worm bearing might bind  (even if ever so sightly) unless the motor is perfectly aligned with both worm bearings.

As far as the aluminum motor case goes, I doubt very much. They are most likely steal coated wits some metal (maybe even aluminum). But they will be ferrous. Just check with a magnet (away from the internal motor magnets not to shew the results). The case is steel to concentrate and direct the magnetic flux of the 2 motor magnets inside.

[narrowminded]

Ron, I thought I would have specs for a motor like that confirmed through the resistance number.  I didn't and the sales sheet that you linked to didn't have them. 

What should be known to properly calculate the resistor(s), in addition to the armature resistance, is the amp draw.  If you can measure that running at 12 volts then a straight calculation can be done.  If it's not something you can do easily then just do it trial and error.  If you have a potentiometer of any sort, 150 ohm or greater and a wattage around 1/2 minimum, just wire it in to one of the track leads and dial it until you get your 6 volts at the rails with the truck running on the track.  When you get that measure the resistance at the pot and you've got your number.  It'll be around 100 ohm.  Make sure you're reading the correct two of the three leads, the pot wiper and the correct end of the sweep.

If you don't have a pot but do have a 100 ohm resistor, just wire that in as it will be close.  That motor will probably draw about 50mA normal running which would want a circuit resistance of about 120 ohms so a 100 ohm resistor plus the armature should put you about right for that.  The wattage would be about 3/8 at those numbers so either needs a 1/2 watt, or two 50 ohm, 1/4W (or more), one on each lead.

Hook the resistor(s) up in either (or both) of the leads going to the track.  Start running the truck and reading the voltage at the track, increasing speed until you get the six volt reading.  Then read the voltage before the resistor.  If it's twelve volts or reasonably close, you're done.  You've got your numbers.  If it's reasonably close you could also just set the 12V before the resistor and then read the track.  If you're not exceeding 6.6 volts at the track, or if it's less than 6V but has as much top speed as you want,  you could call it done.

If you've got one 1/2W or two 1/4W resistors you should be good but it would be wise to run it a while and check the resistor(s) for heat.  They can run hotter than you might like without hurting themselves but if they're really hot, like uncomfortable to touch, they might hurt surrounding parts.  If too hot then you can increase the wattage or split the load over more resistors in series.  That should not be necessary but always good to check.  Expect a resistor temp of about 100 degrees F maximum.

[peteski]

I have not yet tried to scientifically figure out a voltage dropping resistor for a motor but I know few things for sure:

1. it is pretty much impossible to accurately measure the static resistance of the motor.  That is because the armature wingdings are all interconnected, the position of the commutator in relation of the brushes makes a difference, and the resistance of the contact point between the commutator and the brush also seems to be different when the armature is stationary vs. when it is spinning and passing current.

2. measuring current of a free running DC motor is pointless because the BEMF generated by the spinning armature results in a current draw being much lower than when the motor spins under load.  We are at this point dealing with impedance, not resistance.

While adding a series resistor to a motor is sometimes used to lower the voltage being supplied to the motor, it also creates problems with motor control. The motors impedance creates a non-linear load, so the motor will not linearly reduce the voltage.

But as I said, this type of circuit is being used when using 12V to power a motors rated for lower voltages.  Kato Portram LRV uses a 3V pager motors and some interesting looking electronic circuit to drop the 12V to 3V linearly.  I suspect that they use MOSFET transistors. One of these days I'll try to reverse-engineer that circuit.

[u18b]

Ron, I'm surprised that you didn't Beardenize the worm bearings. I would have just used the outboard worm bearing and the motor bearing on the other side would be more than adequate. The inside worm bearing might bind  (even if ever so sightly) unless the motor is perfectly aligned with both worm bearings.

That's a good idea.

I guess I didn't at first because the mesh of the worm/gear is really fine.  Less than a mm, so only a few hundred nanometers.

But since the outboard bearing is locked in place by the gear cover, then as long as the motor is pretty straight, it should be no problem.

I'll make that change.

And just for fun, I'll see if I can measure the current draw before and after to compare.

It may be late tonight before I can get to it.

[coosvalley]

I would love to see someone shoehorn a gearhead into u18b's design...To me a switching loco should run at, well, switching speeds!

I'm loving seeing all of these critter design options take shape..

[narrowminded]

Using just a resistor is very crude but also very small.   With a resistor in the supply the voltage does change in relationship to the load so unless your service is a perfectly fixed load, therefore a perfectly steady amp draw, you won't be able to calculate the resistor to hold voltage to the last decimal BUT you can get it awfully close and totally acceptable in this service.  And it will never be perfect because it DOES change as the load changes but it's also in a tolerable range because the load changes are in a tolerable range.  Using a dropper resistor, when the amp draw goes up, the voltage goes down, and vise verse.   We're using these with a manual or eyeball control and aren't worried about a few milliamps of wasted energy as you might be in another device, especially battery operated, it's all tolerable, and is why it's not uncommon in small power chassis applications.  It's easy enough to try and I think worth the convenience of just running on standard 12V and will be very helpful when it comes to lighting with leds.

The reason I asked what the resistance was is I have some very complete data on various size motors and have found them usable for other manufacturers' motors when I didn't have such data and have used the easy check of the terminal resistance as an indicator of the similarity.  I had found that I could accurately calculate a resistor size with their data when the motor lined up dimensionally and had similar resistance across the terminals because the amp draw followed those two pieces of information.  If not perfect it will at least be a good starting point and after testing adjustments can be made if needed.  For the little effort it takes I think the dropper resistor would be worth a try but if the gearing isn't acceptably matched to the motor RPM don't expect great things regardless of how you get six volts to the motor.  That kind of mismatch can't be fixed with electronics.

I still think that the one item that's going to cause fits with the performance on these things is insufficient gear reduction for these small motors, therefore low torque and high RPM.  The voltage supplied to the track, whether correct as supplied or corrected in some scheme, isn't going to be the problem.  And a proper gearing will work nicely however the voltage is supplied, correct or crudely corrected.  That's been my experience and I don't think it's effected by longitude and latitude.