Salient question

I tried a completely

I tried a completely different engine with another DASR to make sure the MRC system was ok. No, no smoke, smell or heat, I did try a factory reset but no luck. I am going to open it up in the next day or two and check to be sure I did not loose a connection. 

NCE Recovery Mode

Alco,

If you can get access to an NCE DCC system you can perform Recovery Mode Programming. 

Recovery Mode is one of the great tools NCE offers. 

Over the years I have on occasion oopsed with an NCE decoder and ended up with what seemed like a dead decoder. 

This procedure has fixed a few of them.

In simple terms this "force Programs" the decoder back to the way it came from NCE cv by cv 

NOTE this is beyond a simple MFR reset.  It takes a min or so and if it is quiet in the room you can hear the loco responding to the system actually programming CVs.  The motor chatters/ticks a bit for each cv.

Info here

https://ncedcc.zendesk.com/hc/en-us/articles/207904033-Recovery-mode-for-dead-NCE-decoders

Hope this helps

Ernie

2 CVs to try on the programming track.

First, try CV 30 = 2.

Then try CV 8 = 8.

on the PTB-100 look for the yellow led to flash when programming.  The Green LED should be solid and not flashing.  

Also try a reboot of the computer it'self as the MRC PC interface tends to lose it's connection with JMRI from time to time.

Thanks for the tips. I dont

Thanks for the tips. I dont know any one with a NCE system, good to know though. 

I tried to set CV30 to 2 but it did nothing, The motor did not make a sound when I tried it. I am going to try it again in a day or two before replacing the decoder. I will monitor the PB100 leds when I try it to verify what they do. 

As I had just updated JMRI, I verified that I have the correct version of JAVA as I thought maybe that was it. 

. .

I browsed thru JMRI for the NCE DASR definition.. There is no reset to  these decoders (you can thank NCE for this) so CV30 nor CV8 will get you ahead. Only process was the one mentioned by Ernie but that was for ALL NCE environment: system and decoder. 

That leaves you with JMRI.

To Restore Defaults; JMRI -> Open the NCE Silent Drive decoders definition: Da-SR  ->   Go to the CV Tab and write full sheet. These are the 'default values'.  Try one CV to see if it gets thru. If yes fire off the full sheet.

If "one CV" does not go thru, then you have a more serious issue to deal with.

What version of JMRI were you running and what version did you update too ?

Marc

I have not got back to

I have not got back to checking it yet, The NCE decoder manual did say that CV30 set to 2 should reset it to factory defaults.

I had just updated JMRI to version 4.14, I will give that a try, Thanks Marc.

Try using your throttle cab

Bypass JMRI all together and use your Prodigy cab on the programming track.  Enter programming mode and attempt to read the address.  If successful, the address should be displayed on your cab.  If it will not read, it should display "err".  Also, check to see if the PILOT light on the command station is flashing rapidly as this is an indication that the system is attempting to send packets to the decoder and is waiting for an answer.  

. .

Did some digging.

NCE started including a reset via CV in decoder firmware V3.5.  Before that it might be  "present" but undocumented or not at all.

The JMRI definition I browsed was for 3.0 thru 3.4 and indicates no decoder reset available.  Not to confuse with the ' reset to defaults ' on the Roster Entry page/tab.  That does not reset a decoder.

So depending on the CV7 value if higher than 33, it should do a reset when CV30=2.

If you can't even read CV7, you are up a creek. if you tried CV30=2 and got nothing...  :-(

Be cause of the lack of a reset in it's early decoders, NCE came up with the process Ernie suggested.

Marc

Its alive

I took the shell off and inspected everything, and all looked good. I disconnected the motor leads from the decoder and tested the motor with my old power pack and it ran fine. Put the leads back on and put it on the program track, without the PB100 in line. Using MRC not JMRI I did a address read and the motor now made the normal sounds during programing. The address came back with a number I never used, 64? and the engine ran using 64. I reset CV30 to 2 and had an address of 3 and it ran. I reprogramed the address back to 2102 as it should be and it runs. I checked CV8 and its 11 for NCE. Dont know what I did, JMRI must have scrambled something? I can run the engine with the JMRI throttle.

Thanks for the help. Now back to getting JMRI to work.

It's alive

  I love it when that happens :> ) ......DaveB

Update?

Did you update JMRI recently?  I always try to run the most recent stable version which is 4.14 as of now. 

Also after re-reading your OP, I think a glitch on the programming track occurred.  A temporary loss of contact with the engine will not stop the programming but it will cause it to "skip" a CV, so when programming resumes the CV and data are out of sync IE. CV 29 is being fed data for CV 31 as an example.  Reading the address and getting a result of 64 could indicate that happened.

Also the green and red LEDs  on your MRC computer interface should always be flashing.  Green indicates DCC power and Red indicates communication with JMRI. If the red LED is not flashing then there is no communication to JMRI.   When this happens, I usually just reboot the computer to get it working again.

Nelson; Yes I had update JMRI

Nelson; Yes I had update JMRI to 4.14 and JAVA is up to date. I think the LEDS were flashing on the interface. 

Your right power down and power up fixes a lot of things. 

. .

Surprising. I also own a Soundtraxx PTB-100 and it has never caused issues with non-sound decoders for me.

So how were all your previous attempts to reset done ?  From what I read, you tried the reset several times without results. One thing for sure, if the decoder ended up on address 64, none of your attempted resets went thru with the exception of the last one you tried. I would not even bother to guess what might have transpired.

One thing I can guaranty, it was not a hickup and one CV's data going to another CV. Ain't the way it works. But then you are running MRC, so anything could happen  : -)

Another of those fascinating little mysteries of life. As long as it works again, you learn at no additional cost.

Say CV1=64  and CV17=200, CV18=54.. Something accidently changes CV29 away from 38 to 6.. bingo you are now on address 64 and away from 2102   But any of your many resets should have resolved this.

I've been dealing with JMRI 4.14 issues this weekend and they revolved around CV29. A Production version is simply another Test version, kid you not.  There is always a possible glitch that can sneak into any version of JMRI.  There are many cooks and not as many tasters.  They do some rigorous testing, but things can slip thru

As Nelson suggested, it is wise to try it with the native throttle of your system if/when JMRI seems dazed and confused.

Marc