LokSound Conundrum

[Ray H]

My colleague has had a working perfectly Heljan O gauge class 40 fitted with LokSound v4 decoder for a fair while.

 

Within the last couple of days it has failed to respond to any attempt (using an NCE PowerCab/SB5 ) to control it. Instead its sound and lights come on immediately it is placed on his layout and it does beggar all else whilst it stays on the layout. It behaves the same if placed on a test track powered by the PowerCab.

 

It does exactly the same on my test track connected to the laptop via a LokProgrammer. However . . . . .

 

The LokProgrammer can (as far as I can make out) read all the decoder's CV values and if you reset the decoder using the LokProgrammer the loco will happily respond (as Loco 3) as though there is nothing amiss.

 

The loco reverts back to its errant ways described above the minute you try to restore the CV values read initially via the LokProgrammer.

 

There are 192 CV differences between the two sets of CV data which are a lot to try restoring singularly and testing after each value is restored.

 

Can anyone offer any suggestions regarding any critical CV that may have (some how) had its value changed?

 

I'm 99% certain that I have (if I can find them) details of the decoder's CV values taken a while back using DecoderPro (and possibly even via the LokProgrammer) so I may be able to compare that set of data against the data just read via the LokProgrammer. 

 

Thanks.

[Andymsa]

I would recommend getting the ESU decoder tester, you can test any decoder on it. Using the tester can eliminate the any issues with the loco.

[Ray H]

I've managed to locate a set of the loco's LokProgrammer data from last year and compared the two. There are 6 differences. % are expected as we have changed a few CVs more recently. The sixth is it currently has a value of 255 in CV19 which contains the Consist Address.

 

I'm about to head back to the LokProgrammer and change that CV's value back to 0 which is the value that it previously had and is also the default value for the CV.

 

I'll report back.

[Nigelcliffe]

CV19=255 is a legitimate consist address (127 in the reverse direction).   So, it could have been arrived at by operator error in pressing consist keys on a PowerCab (or other system).  

 

- Nigel

 

[Ray H]

That might have been how it happened.

 

Anyway, resetting the consist number to zero has fixed the problem. I just need to remember that's what needs to be done should the loco behave oddly again.