Atlas N scale MP15DC with NMP15 decoder

[Rich_S]

I've tried searching this forum and could not find an answer to my question.

My MP15DC with the factory installed NCE NMP15 decoder will not start moving until speed step 15. I've read through the Atlas / NCE manual and have tried adjusting CV 2, CV116 & CV117 as they suggest and still no joy.

I'm using a Digitrax Empire builder system with a DB150 command station and DT500 throttle. I have CV29 set to 34, using the long address.
I've set CV5 to 255 and CV6 to 128. I've also set CV3 & CV4 to zero and CV95 to 128.
I did use JMRI and my PR4 to set the speed table to a standard curve, but I do not want to tweak the speed table until I can get the model to begin moving at speed step 1.

I've tried numerous settings in CV116 & CV117 and nothing seems to work, is there something I'm missing?

[peteski]

Have you considered that the problem might be mechanical binding in the mechanism, not electrical (decoder) problem?  Can you remove the decoder and check how  smoothly it run in DC and how much voltage it takes to start??

As for setting CV2, if the loco does not move at speed step 1, you need to increase value of CV2 until the loco does start moving at speed step 1.  CV2 is a start voltage. It is the voltage required to get the motor running at its minimum speed.  So try setting it to a higher value (5, 10, 15 . . .) and then narrow the value down to where the loco will start creeping at step 1.

CV2 is the most important CV for what you are trying to achieve, yet you didn't mention what its value is.

I don't know what the factory defaults are for CV166 and 117, but I would leave those alone for now.  Set them back to factory defaults.
As a test, set CV2=50, CV5=255, CV6=150 and see if the loco starts running (faster than you would expect) on speed step 1.

I"m not veru familiar with NCE decoders. The seem fairly basic. This is the complete list of the CVs.

[Rich_S]

Have you considered that the problem might be mechanical binding in the mechanism, not electrical (decoder) problem?  Can you remove the decoder and check how  smoothly it run in DC and how much voltage it takes to start??

As for setting CV2, if the loco does not move at speed step 1, you need to increase value of CV2 until the loco does start moving at speed step 1.  CV2 is a start voltage. It is the voltage required to get the motor running at its minimum speed.  So try setting it to a higher value (5, 10, 15 . . .) and then narrow the value down to where the loco will start creeping at step 1.

CV2 is the most important CV for what you are trying to achieve, yet you didn't mention what its value is.

I don't know what the factory defaults are for CV166 and 117, but I would leave those alone for now.  Set them back to factory defaults.
As a test, set CV2=50, CV5=255, CV6=150 and see if the loco starts running (faster than you would expect) on speed step 1.

I"m not veru familiar with NCE decoders. The seem fairly basic. This is the complete list of the CVs.

Hi Pete, I can assure you the problem is not mechanical, the locomotive moves at speed step one with an old spare Lenz decoder I have. I have CV2 set at 50. Also CV116 and CV117 are for kick starting the motor and according to the documentation the lower the value in CV116 the more kicks or pulses per millisecond and the higher the value in CV117 the more voltage is applied per kick.

I'm beginning to think there is a problem with the decoder as I've tried to change a few of the other values, including the factor reset and it no longer seems to be responding to any programming. I've also tried entering a value of zero in both CV15 and CV16 thinking somehow the decoder became locked, but that did not help as well.

The most disappointing aspect, my JMRI is no longer able to read this decoder and I cannot read CV values on my current DCC system. After the day I've had fighting with this decoder, I'm about half tempted to go back to DC   

[peteski]

Ok, so we can eliminate mechanical  issues.  You didn't mention that originally, so I had to ask.

Rich, I don't  think I ever had to mess around with any of those kick values on any of my decoder installs.

The reason I asked to set CV2=50 is that this should result in the the average motor voltage at speed step 1 being so high that no kicking would be needed to get it running.  At speed step 1, the loco should just take right off at a decent clip.

So yes, the decoder might be faulty. Not sure what warranty NCE offers, but I would look into that.

[Mike C]

  @Rich_S   Does the decoder support BEMF ?  The only time I've seen this was on a Tsunami 1 when I turned off the BEMF  ( it was a tuning alternative tip for Blackstone locos )  With the BEMF off I had to put a value of 15-25 into CV 2 to get the loco to move at speed step 1 ......Mike

[Rich_S]

  @Rich_S   Does the decoder support BEMF ?  The only time I've seen this was on a Tsunami 1 when I turned off the BEMF  ( it was a tuning alternative tip for Blackstone locos )  With the BEMF off I had to put a value of 15-25 into CV 2 to get the loco to move at speed step 1 ......Mike

Hi Mike, I think the decoder has gone out to lunch. JMRI can no longer identify the decoder and when I try to program it using my DT500, it does not seem to accept the new input. I've put a value of 99 into CV2 and it still does not move at speed step one. According to the NCE documentation, CV2 will accept values between 0 and 100. Also per their documentation I've tried putting a value of 2 in CV30, but that does not seem to reset the decoder.  I think this decoder is going to go back for replacement or repair.

Thanks everyone for your thoughts and input, but I'm now thinking I got a lemon right out of the box.