Motor oscillation on TCS K3D3 for Kato NW2?

[Wutter]

I've installed a few of these TCS K3D3s before but this is the first time I've come across this, so just reaching out to see if anyone has seen this before....

After installation, the decoder took programming its address without issue.
When I try to start movement the motor makes a buzzing sound and putting a finger on the flywheel to try and "kickstart" it one way feels like it is oscillating without actually moving in one direction or the other.
I have probed and checked that the frame halfs/pickups are properly isolated from the motor, and that the halves are isolated from each other. 
Lights can be controlled on/off/direction swapped without issue during this motor buzzing.
Removed decoder and tested motor independently in the chassis on DC without any issues.

Any ideas?

[woodone]

Have you tried doing a reset on the decoder?

[Wutter]

Update on this:

Did the CV8 to 2 reset.
Reprogrammed an address again without any issues.
Same problem when using on main.
Probed the motor contacts on board with a multimeter to see what is being output when commanded to speed step 7, fluctuates somewhere between 2 and 8 VDC (this seems not normal)

[peteski]

That behavior is very strange. While I have no solution to your problem, Decoders uses PWM to produce variable voltage to run the motor.  It is basically just pulses.  Not sure a DC voltmeter will indicate that type of voltage correctly.   The pulse polarity does not flip like the DCC track voltage, but it is still bunch of pulses.

I'm saying that because if the voltage to the motor fluctuated between 2 and 8 V the loco would be going from almost stopped (or crawling) to about half of its full speed.  Or does its speed actually fluctuate *THAT* much?

You also mentioned that te armature buzzes without actually spinning?  Does the voltage measured by the voltmeter actually flip polarity?  If the polarity does change, the (digital?) )voltmeter might be too "laggy" to show the polarity changes.  Probably need a scope for that.

The motor in that loco is a typical 3-pole permanent-magnet DC motor.  It is smaller than the ubiquitous 5-pole Kato motors used in most of their models, but it has the same open frame design.  The difference is that it has 3 skewed poles and uses the super strong "rare earth"  magnets.

If I had to guess, I would say the decoder is defective.  Can you try the decoder in another loco (or another decoder in this loco)?  Or just connect the decoder to some loose motor if you have one hanging around.

[Jim Starbuck]

I doubt if it will make a difference but a couple things I have seen cause odd behavior in TCS decoders is having analog conversion turned on and BEMF turned on.
Perhaps try turning those off and see if anything changes.

Jim

[Wutter]

Closing update to this:

Turned turning off BEMF and DC mode but didn't seem to change anything. Took the decoder out and ran it around a little bit longer in DC just to confirm everything was okay.

I bought the equivalent Digitrax decoder for the same NW2 to make sure that everything was working fine on that specific frame (and it seemed to go in without any problems). But then I took one of my older NW2s and swapped the problematic decoder into it and it worked the first time without issues. So it seems like something was just preventing the motor from commutating and it was getting stuck in an odd position and causing the buzzing when I was originally testing it.

Thanks everyone for the input!

[peteski]

Interesting.  So are you saying that the original motor had some sort of mechanical bind (either internally or externally) preventing the armature from rotating? And now that "bad" loco is working ok?

[Wutter]

Interesting.  So are you saying that the original motor had some sort of mechanical bind (either internally or externally) preventing the armature from rotating? And now that "bad" loco is working ok?

Yeah that is what it seems like, I could manually turn the flywheel by hand when the loco was off but when it was being commanded by the TCS decoder originally there seemed to be something holding it from turning while I was trying to help "kick-start" it to move.

[peteski]

Very strange, but I'm glad the problem is resolved.