For twin motor DCS locos, how are motors wired?

The two motors are indeed wired in parallel with no intervening electronics so they always see the same voltage.  The yellow wires and the white wires are connected together on the circuit board right at the 5-pin connector.  Adding jumper wires will have no discernible effect for the current levels at hand.

Your notes say it all.  The tach'd wheel starts slipping but the tach and PS electronics don't "know" that is happening.  The power applied to the motors is decreased as it's easier to spin a slipping wheel at the commanded speed than to pull a load at the same speed.  As the applied common voltage decreases, the other motor eventually stops (stalls).

If you have a Z-4000 or some other indicator of track current, watch the current as the slipping starts.  It should drop as less power is needed.

As Stan stated the motors are in parallel.  The only issue I see with Stan's explanation is both motors are still getting the same voltage pulse so why has one slipped and the other stalled.

It maybe that one motor is not performing to standards and this is causing an imbalance.

Unless you are overloading this and the applied power can no longer drive the motors, but that doesn't explain the slip unless the engine has shifted weight under the load and incline.  How are you moving tach between motors, normally the wires aren't that long, and the tach bracket isn't in place?  G

Stan and G, thanks for your quick replies.  You both told me what I needed to know, same voltage applied to both motors.  Once I was told that, I believe Stan's explanation is the most likely.. the motor with the weight is stalled, while the other one spins with very low current draw. I just surprised it does it so quickly.  There is a brief but noticeable hiccup even at 25-30 SMPH

G- I am sure the engine has shifted weight under the incline.  Remember with a GG1 and a P5a the motor and drive wheels are close to one another.  Not like a diesel where they are separated by the length of the locomotive.  Adding weight over the slipping wheels solves the problem on a GG1, because those trucks are separate. That does not work on a P5a with a single rigid frame. (A VERY dumb design, by the way)

G-In answer to your question, I simply swapped motors, tach reader and all. Its easy to do because the trucks are so close

One solution might be to wire the motors in parallel. The resistance of a stalled motor is higher. Most of the voltage would be dropped across it, which may convince it to start turning.  Of course my SMPH would be off, but I can live with that.

I've had several locomotives where one motor slips and the other stalls.  All it takes is a slight imbalance of the weight on the drivers and that's what can happen.  I have the MTH P5a, and it has six driving wheels and two motors.  Any track irregularity and the driving wheels with the encoder will spin and the others will just sit there.

I have the Legacy PRR T1 Duplex, it has the same kind of issue if you get the right track conditions.

The fact that in incline is involved suggests to me that you have the same kind of issue.

John,

This happens from time to time on engines with traction tires, specifically the ones on the motor that has the tach reader.  If there is smoke fluid or any other oil on them, it will allow that motor to spin easier and starve the opposite motor creating your described condition.  I think traction, or lack of, is your problem.

Dave

I'm sure traction is the issue Dave, the cause I believe in my case is slight irregularities in the track.  It happens in specific places, and I know one of them does have a "hump" that needs to go.  Some locomotives are just way more sensitive to the issue than others.  As far as the T1 is concerned, there are two other identical ones in the club, and we've all seen the issue at times, so I think it's something about the mechanical design of that particular locomotive.

But the motors ARE wired in parallel!  Did you mean to say series?

I'm not sure what you mean by a stalled motor having higher resistance.  Given the same applied voltage, a stalled motor draws more current than a spinning motor.  That suggests the stalled motor has a lower effective electrical resistance.

If you do wire the motors in series, the current flowing through the motors must be the same whether or not one motor is stalled and the other is spinning.  So for DC can motors as we are discussing, motor torque will be the same.  I don't know if this ties in to your idea but a consequence of series wiring.

Also, I'm not clear about your statement that sMPH will be off.  Irrespective of parallel or series motor hookup, the motor controller will attempt to spin the tach'd motor at the commanded speed.  If a series hookup, then only half the voltage is available.  If you are in command mode with, say, 18V on the track, then you would have half the starting/stall torque to get moving and half the available running torque per motor.  But to extent this torque is enough to start and move the engine to speed, then the sMPH should be correct.

I think we all agree that traction is the issue.  So without the benefit of the rubber tires, Lionel's magnetraction system, etc., then I think you are stuck fussing with adding weights or tinkering with the track/wheel smoothness in the areas where they are slipping.  In the category of fantasy scenes-we'd-like-to-see, it would be too cool to have your engine eject sand under the wheels upon DCS command.  I can also imagine some electronic anti-slip motor control algorithm but don't hold your breath.

The real solution, and I'm not holding my breath for this one, is to have the tach on both motors.  You could adopt a simple solution of just getting the "slow" one up to the command speed with the assumption that the other one is likely spinning.  Obviously, this isn't going to happen any time soon, maybe this is a good place for a custom retrofit.

A little uP board that just took both tach inputs in and forwarded the slowest one would allow the PS2/3 electronics to remain unchanged and insure that one motor wouldn't stall as we see here.

One other solution, with which I solved this problem on my P5a is to buy another copy of the same loco with a different cab number and double head them.  With my P5a, one is a box cab and the other is the "modified" version.

Any excuse to buy another loco.

Well, that's an expensive, although cool solution.

Pat,

I didn't have any excuses when I added the P5a Modified to my Box Cab and lashed them together. The synchronized auto pantographs and steam release are just great to watch. 

Just out of curiosity:  Have you tried running the P5 [ especially ] or the G without the pilot trucks but otherwise under the same conditions ?

SZ

I have not tried that, though from looking at it on the one problem track section, I suspect that might be enough to tip the balance.  I did replace the springs with lighter springs for the pilot trucks.  That helped but not enough for the hump in the track.

Er, OK -- but I really aimed that question at John S. [ since I mentioned both the P5 and G ], especially since both were, I believe, converted from 3 rail.

SZ

The first MTH P5a had the tach tape on the motor that drove the

wheels without traction tires.The wheels without the tires would

spin without the P5a moving. By switching the motors so the

motor with the tach tape powered the wheels with the tires, the

P5a ran much better. Adding a little weight to the end with the

traction tires also helped the P5a run better.

The later P5a had the tach tape on the motor that powered

the wheels with the tires so MTH corrected that problem.

Several points:

Stan: I don't know quite what I did with my brain yesterday.  I meant to say series.  As for the SMPH being off, that depends on the algorithim used.  I was thinking of something different, but you are probably correct and it might be OK.

Gunnerjohn, On the P5a I pull the springs out from under the four wheel trucks, and then weight the trucks with lead.  I have no problems with traction/wheels spinning. I am not sure you have to add the weight if you keep it three rail, as the wheels are heavier, and the deeper flanges make easier tracking. But I would definitely try pulling the springs out.

pa: You are correct about MTH changing the tach drive position over the years. I moved the tach drive to the motor that powers the two axles.

SZ. On the GG1, I did try pulling the pilot truck, and it did not help. The pilot truck on the GG1 is fairly free, and can easily move up and down without taking any weight off the drivers.  My problem is entirely due to lack of traction on the tach motor truck in an uphill, superelevated curve.   Working on the track has solved the problem somewhat, in that speeds in excess of 25 mph shows no hiccups

Barry: I agree putting several together and watching the pants go up and down in synch is a hoot.

Pat, I did not settle for just one additional P5a:

John:

Great solution, and in true PRR practice, the modified is in the lead, making the crew just a little safer.

That's a great idea, I considered that when I was looking at the issue.  One question.  Did you just swap the motor positions?  Could it be as easy as that?

Origially Posted by gunnerjohn

One question.  Did you just swap the motor positions?  Could it be as easy as that?

 

yes

I guess the other question would be, did that improve things?  The other suggestion of adding some weight in the back makes some sense, though with the diecast shell it would probably take quite a bit of weight to make a difference.

I guess I could swap the motors around and see how that works, sounds easy enough.

On the P5a units, yes.  That plus removing the springs and adding weight on the four wheel trucks turned them into strong pullers.  Basically you are doing two things:  1) Changing the tach drive puts the decoder on two driven axles, which are less likely to slip. 2) Removing the springs from the four wheel trucks ensures all the loco's weight is on the drivers.  Adding weight to the four wheel trucks just allows them to track better.

Any one of those three P5as you see in the photo above can pull that entire train.  The three are there for the pantograph show, and to follow, as Pat said, prototype practice

Swapping the decoder motor does not help on the GG1, its just changes which wheels slip.  This is obvious, because each motor powers the same number of axles.  So I can either slip going up a grade, or down, depending which truck has the decoder

Stay tuned to this thread, I plan to try the series wiring within the next four hours     

Swaping the motors was easy, just unscrew them and move. The wires are long enough. There was much better pulling power with the motor change and adding

a little weight to the end with the tires changed the balance point just

enough to add a little extra.

I put in weaker springs, I'll give a try to taking them out totally.  Adding weight may be interesting, I'll see what I can fit into them to add weight.

Sounds like I can "improve" this enough to overcome it's deficiencies.

Your issues may be all weight related, but I just finished working on an ABA set with powered slave.  The trail A was dragging one motor.  Occasionally the motor would partially rotate.

I checked resistance of the motor and no load current which seemed high, but no drastically bad.  The motor wouldn't start rotating No Load out of the truck until about 2.2V, and it would stall around 1.2V

This unit had 1 hr total on it.

I used CRC electrical cleaner spray and cleaned out the motor.

No load amps dropped by about .1A, Rotation start about 1.2V and stall at .5V.  Did both motors with similar results.

Unit runs nice and smooth now.  I don't know if it is varnish, or corrosion, but I have seen significant difference in motor performance.

If you are going to max and engine load, or run a MU, I highly recommend cleaning the motors and doing some basic bench testing to ensure they are matched closely.

I see a fair share of Slave unit that have shorted motor FETs or engine with one bad motor.  I don't believe mechanical coupling can limit these imbalances.  If a motor doesn't generate the torque with the given voltage it just locks up.  G

G-  Thanks for the advice. How are you cleaning the motors, pulling them and disassembly, or squirting the CRC through a hole?

 I've never experienced this as a big problem. I have seen it. On a new loco, it goes away with some break in running. It seems like a bigger problem on the bench than I notice on the rails.

 I don't have any tough grades to test. I always run long trains with a lot of power both in front and as helpers. I've been watching this whole thread with great interest.

Joe

It is unlikely you would see this on diesels, where the trucks are placed at the ends.  The weight in that case is distributed fairly equally between the trucks, so there is less of a possibility of wheelslip

I take the motor off the truck and squirt through the holes.  Then dry out.  G

Certainly a reasonable thing to check, I'll just unplug the motors and run them individually from a bench supply at low voltage and see if one has lower speed/torque than the other.

G-Thanks. I agree that getting the motors to have the same low starting voltage would go a long way to solving this problem.  Unfortunately, in my case getting the shells off is a real pain, particularly the Box cabs.

The shell on the P5a comes off very easily, what's different about he box cabs?

Normally nothing, but remember I converted them to 2 rail.  I fixed the pilots, attached couplers to the frame, extended the handrails down the steps, and filled in the side rails:  Easily my most challenging conversion:

Before (photo taken by SandPatch Tom):

After

Well, that probably does change things John, I wondered why it would be an issue, now I know.