TCS ALD4 troubleshooting

[ridinshotgun]

I have a TCS ALD4 decoder that has suddenly started acting wonky on me.

It ran great for a few months (3-4) then around Christmas it started acting funny.  When I would put it on the DCC layout at our club it was like it was already selected on the throttle, which it wasn't, and would power up and take off full blast.  Then another time it would do nothing and I couldn't dispatch the loco.  Nothing nada zilch like it wasn't on the system.  It would run fine though our our DC cleaning track.

I put it aside today until I could put my programming track together to troubleshoot it.  When I first put it on the programming track it read fine then I started reading sheets and changing some CV's and it started acting up.  Sometimes I was getting an error code that there was a short.  It would change the locos address to random changing numbers every time I tried reading the basic sheet.  then it would start pulsing on the track and try to take off at full speed while reading sheets or writing changes.  I was also getting that ozone smell from the engine.

The decoder is installed correctly with klapton around the frame by the motor tabs and it is situated in the frame tabs snugly.  For some reason when I try to do a factory reset of the decoder the CVI block that needs to be changed is grayed out.

I am leaning towards a decoder issue.  What are your all thoughts on it?

[peteski]

Since you mention dispatching the loco I assume that you are using Digitrax system. That functionality is all within the DCC command station and has no effect on the decoder inside the locomotive. That is, no decoder programming takes place on the main line while that address is being dispatched.  I'm talking about a single single loco which is not part of any consist.

As far as the programming track problems go, are you sure that the wheels and all the other electrical contacts between the track and the decoder are clean and solid?  While it is possible for decoder to develop intermittent problems, most often the problem is intermittent electrical contact.  I know that you mentioned that it runs fine under DC control (with the decoder installed) but still, make sure that all the contacts are solid and wheels are clean.

It also sounds like you are using JMRI. Try to program the loco directly from the DCC system (bypassing JMRI).  Doing a decoder reset would be a good thing to try.

[ridinshotgun]

Since you mention dispatching the loco I assume that you are using Digitrax system. That functionality is all within the DCC command station and has no effect on the decoder inside the locomotive. That is, no decoder programming takes place on the main line while that address is being dispatched.  I'm talking about a single single loco which is not part of any consist.

As far as the programming track problems go, are you sure that the wheels and all the other electrical contacts between the track and the decoder are clean and solid?  While it is possible for decoder to develop intermittent problems, most often the problem is intermittent electrical contact.  I know that you mentioned that it runs fine under DC control (with the decoder installed) but still, make sure that all the contacts are solid and wheels are clean.

It also sounds like you are using JMRI. Try to program the loco directly from the DCC system (bypassing JMRI).  Doing a decoder reset would be a good thing to try.

Yes Digitrax and I know that no programming occurs on the main.  The problem is as I stated and no programming was taking place on the main.  Only plugging in the UT4R and trying to acquire the loco through the Digitrax command station  We were not running JMRI at the time.

The loco and contacts are very clean.  I took this loco apart due to over oiling by the previous owner and completely cleaned it before installing the decoder.  It ran perfectly fine after that and ran maybe 2-3 hours after install before developing these problems.  How would dirty contacts cause the auto powering up and taking off at full speed?

I have no ability to program outside JMRI.  I have to use a sprog to do my programming and don't have any ability to access a system where it can be done without JMRI.  Our club layout is disassembled indefinitely and no one locally has a system I can do it on.

How can I do a reset?  The CVI that needs to be changed is gray out as noted and can't be written.

[peteski]

Yes Digitrax and I know that no programming occurs on the main.  The problem is as I stated and no programming was taking place on the main.  Only plugging in the UT4R and trying to acquire the loco through the Digitrax command station  We were not running JMRI at the time.

The loco and contacts are very clean.  I took this loco apart due to over oiling by the previous owner and completely cleaned it before installing the decoder.  It ran perfectly fine after that and ran maybe 2-3 hours after install before developing these problems.  How would dirty contacts cause the auto powering up and taking off at full speed?

I have no ability to program outside JMRI.  I have to use a sprog to do my programming and don't have any ability to access a system where it can be done without JMRI.  Our club layout is disassembled indefinitely and no one locally has a system I can do it on.

How can I do a reset?  The CVI that needs to be changed is gray out as noted and can't be written.

If you have no means to bypass the JMRI (which doesn't allow you to perform the reset procedure) I don't know how else you can try to do that.  Maybe someone else who is familiar with SPROG will chime in.
Can you read (and/or write and verify) any of the CVs back from it on programming track?

How can intermittent contacts (like dirty wheels) can cause the decoder to misbehave?  Some decoders will act "funny" when the DCC packets are interrupted certain way.

[ridinshotgun]

I can read and write some CV's on the programming track but nothing sticks and I do get random red blocks in the CV table.  Whenever you try to write it goes all crazy and writes CV's but not what entered and some CV's that I didn't want to change get changed.

I would think that if a loco isn't acquired on the main it wouldn't be getting any packets?  I was never able to acquire the loco on the main when it did it's best NASCAR impression!

[Mark W]

In JMRI, open your Program window on the Programming Track.  CV8 is read only via pages, but go top left next to the menu dropdown and choose Reset > Factory Reset..

Sounds like a short may have scrambled something.  If a reset does not resolve, send it in. 

[ridinshotgun]

In JMRI, open your Program window on the Programming Track.  CV8 is read only via pages, but go top left next to the menu dropdown and choose Reset > Factory Reset..

Sounds like a short may have scrambled something.  If a reset does not resolve, send it in.

Thanks.  Someone showed that to me a while ago but forgot about that feature.  When I set up the track tomorrow I'll try it that way.

If that doesn't work i guess a call will have to go into TCS.

[peteski]

I would think that if a loco isn't acquired on the main it wouldn't be getting any packets?  I was never able to acquire the loco on the main when it did it's best NASCAR impression!

Acquiring a loco on a DCC system does not mean that the DCC system knows that the loco is on the track and responding to the DCC packets.   Current version of DCC we are using is unidirectional. The DCC packets from the command station, through the booster, are broadcasted (over the track) to all the decoders, but the decoders themselves never provide any feedback telling the command station that they received the packets.

You should be able to acquire a loco on your DCC system if that loco is not even sitting on the track.  At least that is how it is on the all the DCC systems I have ever dealt with.  Yes, some DCC systems and decoders use a transponding feature, but that is not a part of "acquiring a locomotive".

If you could not acquire that loco on your DCC system then I would try to find out why (it is not the decoder problem). But I agree that if the DCC system was not sending packets addressed to that locomotive then the chance of the decoder being fooled by intermittent electrical contact is fairly low.  Is is possible that when the loco took of there was some DC voltage present on the track?  But mind you, I don't totally discount the possibility that the decoder is bad after all, but it would be good to try few things before putting the blame on the decoder.  If you have no means to do so then I guess the only thing left to do is the send the decoder out for repair/exchange.

[ryan_wilkerson]

I had an AMD4 act like this so I sent it in and received a replacement in 10 days (coast to coast shipping). TCS support is the best and that's why I keep buying from them.

[ridinshotgun]

Well some resolution tonight.  i tried a reset and kept getting a short on programming track message.

So while searching for the paperwork for the decoder I found an older ald4 decoder in my stash and switched it out and the loco ran smooth as can be and programmed with no problem.  Put the old decoder in another loco and had the same problems.  So to call TCS tomorrow and get a replacement.

[peteski]

So it did end up bring a blown decoder after all. Nice troubleshooting!