Atlas C-420 Issue

[BCR751]

According to the Atlas web site, the Lenz decoder (LE063XF) was factory installed in the C420 up to 2013 (not really "ancient" times). They began using NCE decoders (AN12A0) in 2013. If you're not really sure if yours are Lenz or NCE, read CV7 (Version Number). It will be 45 for the Lenz and 36 for the NCE. CV8, Manufacturer's ID will show 127 for both (= Atlas).

If they are in fact Lenz, I suggest you read and post CVs 2, 29, 67, and 94 to help us figure out what to do next.

If they are NCE decoders, the problem may be the torque compensation settings, CV116 and CV117. The factory default settings for both are 0, which results in a high start voltage. To get a lower start voltage and overall better slow speed control these need to be increased to 3-6 and 35-50, respectively.

But  -  if you're already on the way the replacing them with TCS decoders, I'll be happy to take the Lenz decoders off your hands!

I got into the decoder today and the results are as follows.  These are values in the decoder when purchased new.

CV 7 = 45   So the decoders are indeed Lenz (I don't know if they are 'scale speed' motors or not)

CV 2 = 8
CV29 = 34
CV67 = 4
CV 94 = 255
CV 116 = 255
CV 117 = 255

Compared to the numbers in the above "quote", these values will likely require adjustment, I'm just not sure where to start

Doug

[BCR 570]

My C-420 runs with a TCS AMD4 decoder which is a medium length drop-in decoder for many of the Atlas/China diesels.  They were also in my C-630M and M-630s until recently.

Tim

[BCR751]

My C-420 runs with a TCS AMD4 decoder which is a medium length drop-in decoder for many of the Atlas/China diesels.  They were also in my C-630M and M-630s until recently.

Tim

I'll likely go the same route. 

Doug

[jeffstri]

I don't see anything unusual in the CVs. Note that CV116 and 117 are not implemented in the Lenz decoder - it operates at a lower frequency than the NCE so it doesn't need torque compensation to run smoothly at low speeds.

All Atlas C420's came with scale speed motors, so with default decoder settings they will run slower at any given throttle position on your NCE PowerPro than a loco model with a higher speed motor and/or gearing.

Have you attempted to change the speed table CVs?  If you're using the default decoder settings, comparing the loco speeds at a particular NCE PowerPro throttle position doesn't tell you much, other than that the locos have different voltage/speed characteristics. That's a locomotive, not decoder characteristic. You can't expect two different locos, even if they have the same decoders, to have the same throttle/speed profiles. To match loco throttle/speed profiles of two different locos, you need to change the speed table CVs of one or both of them. 

For the Lenz you'll have to use the 28-step speed table (CVs 67-94), since the LE063XF has no CVs for the mid and top speeds when the minimum speed is controlled by CV2. For a 28-step speed table, CV94 determines the percent of the track voltage applied to the motor at the maximum throttle setting. For example, if CV94 = 100, the NCE PowerPro (or Digtrax, Lenz, etc control station) applies 100/255 = 39% track voltage at the highest throttle position (maximum CV value is 255). The CV67 value determines the percent of track voltage at the lowest throttle setting, e.g., with CV67=10, 3.9% of track voltage is applied at the lowest throttle position. I set the CV67 value at the lowest needed to get the loco to barely creep, and CV94 to get the maximum realistic prototypical speed, and then as a first approximation linearly interpolate the other speed table CVs between them. This can get done in a few minutes with JMRI DecoderPro. You'll also have to change the value of CV29 to 50 to activate the 28-step speed table.

Just replacing the LE063XF with TCS decoders will not automatically result in speed matching. You'll still have to go through the same process.