N gauge decoders not responding

[MarshLane]

Evening all,

I’m hoping somebody may be able to give me some pointers because might Digitrax DCC System and these models are slowly drive me round the twist.

 

For the past 12 months I had one Kato engage model that just would not respond the lights were on but nobody was home. I was undecided as to what the decoder in the loco was, it was already fitted when I bought it, But I’ve since established that I am 99% certain it is a TCS. I’ve tried to do a reset doesn’t have any effect I’ve tried to program the decoder from both the handset and from the PC using the well-known piece of software that I forgotten the name of the moment., called JMRI Decoder Pro. Still no luck and I concluded the decoder must be faulty.

 

Tonight I decided to have a decoder fitting evening I have about seven locos have now got over the code to get with them. Started with another Kato trying to put a DN163K1C decoder in it. The American Models in theory should be so easy the existing light boards simply pull out and the replacement decoder board pushes in. In all cases these are pure motor not sound.

 

Successfully installed I put it on the way out the lights come on and I can turn the lights on and off using F0 but despite being set to address 03, no movement. Two iffy decoders seems highly unlikely. After 30 minutes of playing around I decided to put the model to one side and move onto the next one.

 

The next is an engage Fox Valley models Kansas City southern SD70ACe model, which uses a really small six-pin Dogitrax DZ126N decoder that plugs into the original light board. As simple as it gets. So having fitted the decoder, I placed the logo on the track, and nothing, zilch, zero. Quickly established that the pushing board was fitted the wrong way round, turn it round replace the locomotive on the track and hey presto the lights come on. Set the controller to address 03, and the same problem comes again no traction, no movement, but this time also no ability to turn the lights on and off. Now three iffy decoders can’t happen, So I am now thinking it is the DCC controller at fault. 
 

I have come back to the first Cato loco that I tried tonight and attempted to do a reset on the decoder the display gave me an old code including what appeared to be the word wrong at one point and although the power booster beeps as normal the loco does not shuffle. Same happens if you try to reprogram the address same happens if you try to reprogram a CV value.
 

In all of the above cases the DCC system is connected to a 2 m long isolated independent test try this isn’t on the layout it is just a test track and the model concerned is the only one on the track. You know the check that I had got track power I’ve placed another loco that was bought with a DCC decoder already and fitted, Select that address and everything is fine traction, lights, direction all fully controllable.

 

Can anyone offer any thoughts, ideas or suggestions of things that I could try, check or do? Something (probably very simple!) isn’t right here. My DCC system is a Digitrax DB150 booster with DT402 handset connected directly to it. Never had any issues or problems in the past.

 

All help gratefully received!

 

Rich

Edited June 18, 2021 by MarshLane

[Nigelcliffe]

What's your connection device from DB150 to PC ?    Reason for asking is the DB150 is a dumb device, can't read from decoders.   But, your connection device might allow reads to take place.   

 

A possible issue might be the "status edit" of the address/slot.   If set to something weird, you might find it doesn't work.  Status edit will be in your Digitrax manuals for the DT402.   

 

 

- Nigel

 

[MarshLane]

6 minutes ago, Nigelcliffe said:

What's your connection device from DB150 to PC ?    Reason for asking is the DB150 is a dumb device, can't read from decoders.   But, your connection device might allow reads to take place.   

 

A possible issue might be the "status edit" of the address/slot.   If set to something weird, you might find it doesn't work.  Status edit will be in your Digitrax manuals for the DT402.   

 

 

- Nigel

 

Hi Nigel,

Thanks for the quick reply. The PC connection is a RRCirkits loco buffer. The status edit of 03 is showing Stat 128.

 

Rich

[Nigelcliffe]

I'm stumped because you've no mechanism to read decoders.   So all that's possible is random prodding.   

 

Suggested random prodding is to check what programming mode is being invoked (page, direct, etc..).   But without decoder read-back I'm not able to offer much in the way of suggestions - being able to read a decoder is my starting place for problems. 

 

- Nigel

 

[MarshLane]

1 hour ago, Nigelcliffe said:

I'm stumped because you've no mechanism to read decoders.   So all that's possible is random prodding.   

 

Suggested random prodding is to check what programming mode is being invoked (page, direct, etc..).   But without decoder read-back I'm not able to offer much in the way of suggestions - being able to read a decoder is my starting place for problems. 

 

- Nigel

 

Thanks Nigel,

Ok, I have tried both paged and direct mode, with no response from the decoder at all, although the booster has 'beeped' as it should do.  What item of equipment - either in addition to, or change, from my current setup - would you suggest? The DB150 is probably 10-12 years old now if not a bit older.

 

An interesting twist to this tale. It looks like a decoder issue with the Digitrax decoders, possibly all three of them.  I took one of the DN163K1C boards out of the Kato model, and replaced it with a TCS K1D1 board, straight swap, nothing else, and bingo, everything worked fine on Address 3 - lights, traction all controllable, reprogrammed the board to the loco address, booster 'beeped', loco shuffled.  Selected 4682 on the controller and hurray, everything still works!  Put the Digitrax board back in and zilch, back to where we were.

 

So I can only conclude that it is something wrong (either a fault or CV value) with the Digitrax decoders, but why it should affect three or four boards, which have been bought at different times from different places, I am really not sure.

 

Rich

Edited June 17, 2021 by MarshLane
Spelling

[Nigelcliffe]

To be able to read decoders one of the following: 

 

a - Digitrax PR3/PR4.   This can act as either a programming track device, or as a Loconet-PC link (replacing/duplicating the LocoBuffer).      Advantage:  it can be a loconet-pc link (but you already own one),  and can load sounds into Digitrax sound decoders (if you have any).   Disadvantage:  not quite as good at reading decoders as others, support is slightly remote in US, though UK dealers are usually helpful.  

 

b - Sprog.   Stand-alone computer-interface device for programming, test running locos, (and optionally running a layout).   Do NOT connect to the same track as your Digitrax setup (or any other DCC system).   Advantage:  probably reads more troublesome decoders than most other things, been around for ages, solid UK support from its maker.   

 

c - DCC++.   Arduino DIY (and semi-DIY) option, can build a combined programmer and command station for very little money.   For this situation, achieves much the same as Sprog, for less money.  But its yours to maintain, update, etc.. rather than a manufacturer supported box. 

 

d - upgrade the DB150 to one of the DCS-series of command stations which support a programming track. 

 

 

 

 

Long shot thought on Digitrax decoders.   Have you accidentally engaged the "test" mode in the decoders ?  This exists for Digitrax sound decoders, and it disables the motor output, no idea if its present in other decoders, but it may be (many makers use common firmware, and it would be needed for the sound-bug piggy-back decoders).   I think its CV128, and (I've not checked this!!) guess that zero is "normal running".   

 

 

- Nigel

  

 

[MarshLane]

Nigel,

Thanks for the info, much appreciated, and thanks for your time. I’ve acquired a SprogIIv4 from CoastalDCC this morning, so once that arrives, I’ll let you know the outcome. I’d never realised that the Digitrains DB range doesn’t ‘read’ decoders.
 

Being that the American locos run double and triple headed, then speed-profiling the fleet of 14 to match the ‘master’ loco is another task that I suspect the Sprog will be better for. Something else to do some reading up on!

 

Thanks

Rich

[Nigelcliffe]

Speed profiles are easier to do "on the main", for which you don't get "read back".  But you do have the reference loco under control, and can tweak the other loco(s) until they match the reference loco.   To do it, initially read the decoders to give a starting value, then if all changes are made with JMRI, the changes are known and recorded, so you know the values in all of the locos.     
Its about equal between doing it with the Sprog alone, or doing the reads with Sprog, then moving to layout with DB150/LocoBuffer to do the writes to adjust the speeds.  

 

If you set up the "profiles" in JMRI carefully, you use the same file location for the "roster" for both the Digitrax interface (LocoBuffer) and the Sprog.   Thus, the same data is used and is visible when using either interface, so no need to manually copy anything or keep it "in step" when moving between the two hardware setups. 

 

 

- Nigel  

[MarshLane]

Hi Nigel,

Thanks for that advice, I'll let you know how I get on.  For interest, I opened a support ticket with Digitrax in American yesterday.  Very impressed with their turn around times, added it at 1820ish our time, and had a response within 45 minutes.  In case it helps anyone, their response was:

 

Richard,

It is highly unusual that 3 different decoders would act the same way.

Reset the DB150 to factory defaults.  Place the MODE toggle switch on the DB150 in the Center OP position and then press the SWCH key, press 39 and then the c key (which is also the Cloc key). You then go back to the command station and push the MODE toggle Down to Sleep and then Up to Run.

 

Reset the decoders to factory defaults CV8=8.

 

Once you reset both, test on address 03 without doing any other programming.

 

If the decoders are still non responsive, You can send the decoder in for replacement. 

Digitrax Help Desk.

 

Cannot ask more than that really - but it suggests it really is an unusual situation.  I'll try that when I get time (probably mid-week) and by then the SPROG should have arrived, and I can read the decoder settings. I'd like to see if I can find the fault, so I (and anyone subsequently reading this) knows what the problem is and (hopefully) how to rectify it.

 

Rich

[MarshLane]

On 18/06/2021 at 08:43, Nigelcliffe said:

b - Sprog.   Stand-alone computer-interface device for programming, test running locos, (and optionally running a layout).   Do NOT connect to the same track as your Digitrax setup (or any other DCC system).   Advantage:  probably reads more troublesome decoders than most other things, been around for ages, solid UK support from its maker.   

 

Long shot thought on Digitrax decoders.   Have you accidentally engaged the "test" mode in the decoders ?  This exists for Digitrax sound decoders, and it disables the motor output, no idea if its present in other decoders, but it may be (many makers use common firmware, and it would be needed for the sound-bug piggy-back decoders).   I think its CV128, and (I've not checked this!!) guess that zero is "normal running".   

 

Hi Nigel,

From the reading I have done on the motor only (ie non-sound) Digitrax decoders there isn't a test mode, that's only on the sound services.

 

The Sprog unit has arrived this morning, so I'll try and read the CVs this evening.  Is there anything specific to look at, that might give a clue as to what is wrong please?

 

Rich

[Nigelcliffe]

2 hours ago, MarshLane said:

The Sprog unit has arrived this morning, so I'll try and read the CVs this evening.  Is there anything specific to look at, that might give a clue as to what is wrong please?

 

 

In this order: 

 

Can it identify the decoder maker (Digitrax) and possibly type.   That tests reading CV8 and CV7.   CV8 should always give a sensible value, CV7 might or might not, depends on decoder model (and I've limited experience of Digitrax decoders).   Proves the decoder is responding correctly to "read" requests.  

 

Is CV19 set for consists ? (ie. a non-zero value).  If yes, that's likely to be your problem, change CV19 to zero. 

 

Is CV29 sensible ?  Sensible includes:  28/128 speed steps (not 14) and whether its on long or short addressing (based on what sort of address you're expecting).  

 

In all cases, the loco motor should twitch when reading or writing indicating decoder is functioning.  

 

Does the decoder now run on the Sprog (turn on track power in software, open software throttle, try running it) ?   

 

If nothing seems sensible, try the decoder reset mechanism (drop menu on the decoder pane in JMRI) and repeat the tests.     And if still stuck, I'd be replacing the decoders, but three going at once seems very unlikely.  

 

 

 

- Nigel

 

[MarshLane]

Nigel,

Thanks ever so much for your help.  Some degree of success tonight.  Two of the three problematic decoders, were talkative to the Sprog, and doing a reset on them, then resulted in everything working.  The third just won't identify itself to DecoderPro at all, there is no lights flashing or twitching, and the power light on the Sprog starts flashing constantly for 30 seconds, with DP reporting 'no locomotive found', so I will send that back to Digitrax for replacement, as I suspect its dud somehow.  

 

However, thanks to your help and the Sprog unit, along with Decoder Pro, I now have nine chipped locos available all of which are fully working, so many thanks.  I'll now identify which is the slowest, and takes the highest speed step to start as the 'master' loco and try and profile the rest to match it.