Author Topic: 'false' faulty decoders  (Read 641 times)

Offline renegade

  • Full Member
  • ***
  • Posts: 116
    • View Profile
'false' faulty decoders
« on: January 26, 2015, 11:07:46 AM »
ok, i do not know if this should be here, or under the DCC topic... Ed, please feel from to move stuff around, if that will make you feel better...


I have been driven to the verge of histerysis this weekend with decoders or rather, with my system not picking up the decoders...

I was tuning up some of my locos, tinkering a little with vStart values and momentum. etc etc etc.

I finally got to a loco (Walthers Trainline with a hard wired basic Roco decoder and 1 bulb for headlight)  placed the loco on the programming track, tried to read, and system says no decoder present...  I have the loco shell off, and am looking at the little green pc board that I have installed and tested a few months back.. so i KNOW there is a decoder...

I put another loco back on the programming track, scan, and it picks up the decoder.   put this Walters/Roco back, no decoder present...  WFT!!!!

OK, so perhaps the crappy Roco decoder decided to finally do me a favour and DIE! (it is an old, basic decoder, can only use 2 digit addresses, very limited speed tuning, etc...)

No problem, I have 2 Digitrax DH121 (brand new, but obviously a few yours old, in unopened packets, and a TCS T1, brand new unopened.  I pick one of the Digitrax decoders.  cut the Roco, and solder the Digitrax harness in its place.   I put the orange /grey wires on the motor.  I put the red/black on the 2 pickups from the wheels.  I do not wire a light in for now...

Put the loco on programming track...  scan, no decoder present!!!  WFT!!! REALLY!!!!   put on eof the other working locos on programming track, scan, detected 100%.   put the Walthers/Digitrax loco on the track, no decoder...

Now I am getting seriously fu#^%ed off at this loco...   unplug the Digitrax and plug in the TCS T1, same, no decoder present... unplug this, plug in the other digitrax, no decoder present.     
I am ready to take this loco and throw it agains the wall!!!!!


I cut the harness from the loco.  I have 2 athearn motors lying on my workbench, I pull one closer, get the Athearn DCC harness from digitrax, and clip it on, and plug in onbe of the DH121 decoders, and using crock clamps, connect it to programming track.  scan, no decoder present!!!!!   AGAIN put back a working loco.  scan, and detected fine!   I even tried a soundtraxx decoder connected with temp wires, scan failed.

So I am 100% happy that my DCC system IS working fine, I can scan and program fine.

My understanding is that in order for a decoder to be detected and working, it ONLY has to have a motor connected.  What are the odds that 3 motors AND 3 brand new decoders are all bust?  is there anything else I can try? check?  how else to test the decoders?  should I be looking for an expiry date or best before on the packaging of the decoders? 

The locos that ARE working and that I can program/scan have the following decoder and connections:
1) a Digitrax decoder with head and backup light, and motor (Kato)
2) a soundtraxx decoder with back and front light and motor (Life Like P2K)
3) a Roco basic decoder with only front light, and motor (Life Like P2K)

The items I tested, in various combinations are:
1) 2 different Athearn 5 pole motors
2) Walther Trainline 5 pole motor
3) 2 different Digitrax DH121 decoder with ONLY motor connected
4) TCS T1 decoder with ONLY motor connected
5) SoundTraxx DSD-100LC sound decoder with ONLY motor connected

please advise or guide me a little here, please....
Francois Kritzinger

Modeling HO American and S.A.R. And any high speed trains. And lego trains. And... Oh I just love trains!

Online capeklr

  • Hero Member
  • *****
  • Posts: 1556
    • View Profile
Re: 'false' faulty decoders
« Reply #1 on: January 26, 2015, 11:27:24 AM »

What are the odds that 3 motors AND 3 brand new decoders are all bust? 
Very unlikely!
is there anything else I can try? check?
Is the decoder getting power in the loco?
how else to test the decoders?
ESU makes a very nice decoder tester, that has the standard six pin, eight pin and 21 pin plugs as well as a spring clamp section where you can connect with wires.
should I be looking for an expiry date or best before on the packaging of the decoders? 
Never heard of an expiry date for decoders.

Offline Freightcars

  • Full Member
  • ***
  • Posts: 123
    • View Profile
Re: 'false' faulty decoders
« Reply #2 on: January 26, 2015, 10:01:19 PM »
Renegade,

Hmmm, that's a dilly of a pickle.  Nothing makes your weekend like some decoders not doing what they should.  :lamer:

A couple of quick questions:
1. What DCC system are you using to program these decoders?  You've mentioned in the past you use Roco and Digitrax (and maybe some others) but what are you using in this case?
2. What programming mode are you using?  Paged, register, physical, direct?
3. Whats the setup between the DCC system and the programming track?
4. Is the loco in question wiring 100% correct and reliable?  Trace the wiring using a multi-meter and see that you get continuity from the track to the decoder for both power pick and motor wires. Also check that wheels, wipers and track are clean.

You are right, for any decoder to read it needs some kind of load.  This is normally on the motor wires (orange and grey).  The load is needed so that the decoder can acknowledge the commands its getting.  You obviously also need to have to have the power wires (red and black) connected to the track.

Some things to consider:
  • Some decoders (especially really old ones) don't have the capabilities to be read.  I stand under correction but the early DH121's may be one of those types.
  • Others will only respond if read using the right mode on the programming track. If you're using paged mode, try one of the others
  • Some sound decoders such as Soundtraxx Tsunamis need higher current to be read.  This has to do the design of the decoders and them been a more power hungry and the fact that command stations only put out very little current when programming.  To overcome this you need a programming track booster such as a SoundTraxx PTB-100.  Personally I use a Digitrax PR3 with a 16V power supply and that reads pretty much anything.

To give an answer to your questions:
The odds of all 3 decoders and motors not working is highly unlikely.  You could try subsisting the decoders into a known working loco and vice versa to your test motor setup to prove that everything does or doesn't work.  If a known good decoder doesn't work on your test setup then you know where to look for problems.

As capeklr has said, ESU makes a real good decoder tester.  These have almost every type of connection you may need as well as a motor, LEDs and a speaker to test decoders.

There isn't a best before / expiry date for decoders.  Old decoders are just that... Old.  Old decoders will generally work with new DCC systems but there might be some catches such as needing to use a specific speed step mode to get them to respond.

HTH,

Niel

Offline renegade

  • Full Member
  • ***
  • Posts: 116
    • View Profile
Re: 'false' faulty decoders
« Reply #3 on: January 27, 2015, 09:15:12 AM »
Ok, some further experimentation last night, just to muddle the water even further...

I have another loco that works 100% with Digitrax decoder, and the decoder can ALSO not be read...  however, if I punch in the address and RUN the loco, it works fine... Also, one system WILL program the decoder, and the other will NOT program the decoder... so I am 100% happy that my decoders are not faulty... but something is not kosher with the programming...


Niel, I will answer as complete as possible... 
1.   I use 2 systems...  my Console is an ESU ECoS II (50200)  my other system is the much loved/hated Roco Digital system.
2.   For both the ESU ECoS and the Roco system, I use direct programming on a programming track. ie, select a CV and set a value...
      The ECoS does a 'scan' of the decoder, to load all the data and display it for you to adjust.  The Roco can not read decoders without the Computer interface (which I do not have plugged in at this stage, so can not say if it will actually read the digitrax decoder...
3.   a)  Using the ECoS, I have a length of track connected to the PROG OUT connections of the console. The system is configured to boost PROG OUT, to allow for sound decoders to function on the programming track.
      b) The Roco system simply has the same length of track to the TRACK OUT connections (PROG OUT only on computer interface)

I agree RE the connections..   to achieve this, I have the chassis on the prog track, with motor connected to orange/grey and wheels to red/black...  so I am very much certain I have this bit 100%.  I even 'bypassed' possible bad wheel contact, by placing aligator clips on the rails, and on the red/black wires just before the decoder. so I am 100% the decoder WILL get signal from the track.

The DH121 'reads' 100% fine on the Digitrax system used at GFG and PMT before that, as well as using the Roco computer interface. so that is not a problem.   The ECoS will also not write to the decoder at all.

I have NOT tried to use different modes on the ECoS to do the rea, and will do so tonight, and see if that makes any difference.  The funny thing is, using the Roco system, I AM able to write data to the decoder that the ECoS will not scan or program.   After changing values using the Roco, I can test it on the ECoS and it works.   Lets say I change the address using the Roco system.  switching over to the ECoS, I can address the loco using the new address, so I know the Roco DID write to the decoder successfully...  The ECos is in fact 100% happy to CONTROL the loco with the Digitrax decoder, it simply refuses to program it!!!

The ECos has configuration parameters that allows the PROG  track to be boosted to compensate for the sound decoders (up to 4.2 Amps if required)  so I know this is not the issue.   also, the DH121 is a very basic decoder.

After brining the Roco system online last night, to test side by side, I can confirm that all my decoders, and the motors are 100% working.  there was nothing wrong with my test or programming methodology.

It apears that ehe ECoS firmware does not like the Digitrax and TCS decoders.  Ironically, when I was testing the Roco Multimaus for Roco back before it launched, they had the exact same problem with the Multimaus!  The Multimaus would not program Bachmann, Digitrax or TCS decoders I had, but the Roco, ESU and Soundtraxx I had all worked 100%... they did a firmware fix (Multimaus V1.02) which I also tested for them to confirm the problem was gone.


What I am wondering about... I have read on a few places that some systems have problems reading/programming some decoders.  In many cases the solution seems to be to add a resistor to the programming track outputs.  In some cases it seems to be placed in parallel, and in other serial with the programming track.  a +/-1K resistor seems to be a fairly commonly suggested value.  I have NOT seen anything specifically relating to ECOS, but I am still looking...


Oh, and the comment about decoder expiry date was a little tongue in cheeck humor, I do not REALLY think they have an expiry date...
Francois Kritzinger

Modeling HO American and S.A.R. And any high speed trains. And lego trains. And... Oh I just love trains!

Offline renegade

  • Full Member
  • ***
  • Posts: 116
    • View Profile
Re: 'false' faulty decoders
« Reply #4 on: January 28, 2015, 01:39:34 PM »
So, last night I finally got things to work, thanx to the suggestions made by Neil (Freightcars)...

up until now, I have ONLY tried direct mode programming, since 1) it is the default, and 2) it has not caused any problem with either the ECoS or the Roco systems...  until now that is...  it would appear that in one of the recent firmware updates, direct mode stopped working for some decoders...  it seems that in the OLDER version of the firmware, the ECOS wouold automatically handle programming mode, by trying Direct, then Paged and then register modes.  they changed it along the way that the user now has to SPECIFY what mode to use, hence my problem started...  sounds like a step BACK if you ask me, not an improvement...

When I switched to Paged Mode, I was able to read and write the Digitrax decoders (and the TCS).  which is a big relief.   

But it is still inconvenient, since the ECoS has a very nice interactive, plain english programming capability, that allows you to to program without having to know which CV/combination of CVs to set to achive anything.  since this can not be changed from Direct to Paged mode, I can not use this, I have to use the advanced programming dialogs on ECoS, and swith to paged mode.   I can then read/write any CV with any value I choose.   I have to understand though, things like what bits to set in CV29 to allow long address, or how to construct the 2 values to be used in CV17/18 to actually save a long address...   not of these are a problem to me, just inconvenient that I now have to do the work the ECoS can do for me.


So in short, some decoders WILL happily program on the ECoS in direct mode, other need paged mode.


I also experimented with the resistor many on the internet suggested must be added to the programming track, using 1/4W resisters in various values from 100R to 1K... none made any difference.  the smaller values just heated up very fast.  even toggling between current 'off' and current 'on' on the programming track made no difference. 

either way, thanx Neil for reminding me there are other programming modes that I can try!   :thumb:

regards
« Last Edit: January 28, 2015, 01:51:34 PM by renegade »
Francois Kritzinger

Modeling HO American and S.A.R. And any high speed trains. And lego trains. And... Oh I just love trains!

Offline Freightcars

  • Full Member
  • ***
  • Posts: 123
    • View Profile
Re: 'false' faulty decoders
« Reply #5 on: January 28, 2015, 06:23:58 PM »
Hey Francois,

Glad you got it working   :thumb:

Niel