I operate with an NCE Power Pro 5 amp system. I have a pair of Athearn Genesis F3 units, an AB consist, with factory installed Tsunami sound decoders. These decoders are identified as TSU-GN1000.
I had not run this consist in months, but last night I fired up the consist and started backing out of the station. At first, all went well, but within a few seconds, the consist stopped. I realized that the B unit was operating properly, but the A unit was moving slowly, effectively bringing the consist to a halt.
So, I killed the consist and tried to run each loco independently of the other. The B unit performed fine, but the A unit had issues. I no longer had lights or horn or bell although the prime mover still made sounds.
The A unit ran very slow even when I applied full speed (28 speed steps), and it would not stop on command or go into reverse. However, left running, it eventually responded to the stop or reverse direction commands. If I lifted one side of the wheelsets off the rail and put the wheelset back on the rail, the loco would respond to the command immediately, but still run in slow motion. Still no lights, horn, or bell.
At that point, I came upstairs and found this thread on the forum.
http://cs.trains.com/mrr/f/744/t/209634.aspx
So, this morning, I removed all other locos from the layout and followed the blast mode procedure that Crandell offered up earlier in this thread. That got the lights, horn and bell to respond, but the loco still ran in slow motion and failed to immediately respond to speed commands, stop commands and reverse direction commands.
I have tried this procedure several times. On a few occasions, I have been able to get the loco to run properly for a few seconds but then it reverts to slow motion and I lose the lights, horn, and bell.
So, I replied to that other thread to ask for help. Meanwhile, I did some additional research and found out that the Tsunami decoder has a LOCK function. I never knew this, and I never messed with the CVs that control the LOCK function. So, I am guessing that the decoder went haywire and somehow locked itself up. Or, maybe the long layoff put it in some sort of sleep mode?
To solve my problem, I put the problem loco on the programming track (with a PTB-100 wired in). I checked CV30 and the value was 0 so the LOCK function was supposedly disabled. CV30=1 enables the LOCK function. I then checked both CV15 and CV16 which control the actual LOCK function.
According to the Soundtraxx manual, the values of CV15 and CV16 must match to LOCK the decoder. The values were both 0, but I took the extra step of resetting each value to 0 anyhow. I saw the loco jerk when I did that. Put the loco back on the mainline and everything worked, light, horn, bell, prime mover, and the proper speed and direction.
Crazy, huh?
Rich
Alton Junction
Yes, that is very strange. because the LOCK function only serves to lock CVs from being changed, it has nothing to do witht he actual operation of the loco. If it so happened to be locked, that may be anough to prevent CV19 from being set when consisting them, which could make for some odd issues.
One or the other may always read 0, I've not bothered with the lock function on any of the decoders that support it, so it could be that one actually was not 0 until you set it, thereby unlocking the decoder. But really, that shouldn;t prevent it from running, or for the bell and horn to work, or any of that other stuff - it would just prevent you from doing things like change the momentum, or change the speed table, or change the address, etc.
--Randy
Modeling the Reading Railroad in the 1950's
Visit my web site at www.readingeastpenn.com for construction updates, DCC Info, and more.
Good ol electronics, sometimes they make you wanna take up golf huh Rich?
I have experienced some ghosts in the devices before, I threaten to regulate the loco to the shelf before I can finally get the things to cooperate.
Your experince may vary.
"The true sign of intelligence is not knowledge but imagination."-Albert Einstein
http://gearedsteam.blogspot.com/
rrinker Yes, that is very strange. because the LOCK function only serves to lock CVs from being changed, it has nothing to do witht he actual operation of the loco. If it so happened to be locked, that may be anough to prevent CV19 from being set when consisting them, which could make for some odd issues. One or the other may always read 0, I've not bothered with the lock function on any of the decoders that support it, so it could be that one actually was not 0 until you set it, thereby unlocking the decoder. But really, that shouldn;t prevent it from running, or for the bell and horn to work, or any of that other stuff - it would just prevent you from doing things like change the momentum, or change the speed table, or change the address, etc. --Randy
This has happened to me before in prior winter seasons after a summer layoff. However, in the past, it was sufficient to reset the decoder to factory default and clear the consist (CV19) in the process. When that didn't work this time around, I tried the "blast mode" procedure which restored the light, horn, and bell, but not the proper movement. So, then, I took the step of clearing the LOCK function (CV30, CV15, and CV16).
I have never messed with the LOCK function or those three CVs, so I cannot explain it except to speculate that the decoder went haywire when I accessed the consist after a long layoff.
The reset to factory default does work and presumably even when the decoder is locked, assuming that it was, indeed, locked. Which then raises the question, is the LOCK function ON as part of the factory default?
According to the definition, the decoder lock should be off by default, CV15 and CV16 both 0. Even a reset is not supposed to work when the decoder is locked. When locked, the only CV you are supposed to be able to change is CV15 which is the 'key'. The values only go from 0-7, if CV16 = 0 then the decoder is not locked, if CV16 = 1-7, then the decoder will only accept writes if CV15 = CV16. If you write 0 to CV15 and then attempt to read CV16, if you get 0 then the decoder is not locked, if you get no-ack then the decoder is locked and if yoou don't know the lock number you will have to try 1-7 one at a time until you can read CV16.
It's possible you got a 0 for CV16 initially because what you really got was a timeout and it never responded and may have been displaying the last known number.
Lock or not, it still doesn't make much sense because the lock does not interfere with operation. It was created specifically for dual decoder installs so you could configure overlapping CVs and have different values for each decoder - lock one, unlock the other and even though you are on the program track, or they have the same address and you are doing POM, only the unlocked one will be changed. Then lock that one and unlock the other, now only the second one will accept changes.
Yeah, Randy, I agree with what you are saying.
I normally consult the Tsunami Diesel Technical Reference manual which is a more comprehensive decoder manual, but in this case, I just read through the more basic Tsunami Diesel User manual.
The User manual points out that the LOCK feature is not enabled (CV30=0) as the factory default, and this is "to avoid inadvertently locking the decoder when the CV Lock feature is not needed".
The User manual also points out that "a locked decoder cannot be reset to its factory defaults until it is unlocked".
Also, a locked decoder "cannot be programmed in either Operations Mode or Service Mode".
Further, "Tsunami decoders are shipped from the factory with all CVs unlocked, that is, CV15 and 16 are both set to 0".
And, "if the decoder is unlocked, changing the value in CV16 will instantly lock the decoder".
The other relevant section of the User manual in this regard states, "Occasionally, something goes wrong and Tsunami will not respond as expected. Usually, this is caused by one or more CVs being programmed to the wrong value".
Lastly, "If you cannot get the decoder to reset, check to see that it has not been inadvertently locked".
So, I wonder two things, since I did not mess with CV30, 15, or 16 and never have messed with these CVs.
One, is the decoder in the A unit faulty?
Two, is there a decoder programming error in the consisting function of the lead loco?
I guess that I need to call Soundtraxx and discuss this issue.
I suppose it IS possible that the bits were randomly scrambled whiel the locos sat unused, but really that shouldn't happen, the duration of the non-volatile memory in the decoder is 10's of years at least. I guess I will find out as all my locos have been in storage for over 3 years now at the least, and some longer.
SO according to that, Tsunamis have doube protection against getting locked out, if you need to set CV30 to allow the lock feature to even work, and THEN set CV16 to the lock value you want. Which makes it double odd - because I can see accidently hitting 16 when you mean a different number when programming a CV, but to take two explicit steps, setting CV30 and CV16, there is practically no chance of that happening by accident.
If it stays funtional, you're probably OK, but if it goes bonkers agian, definitely call Soundtrax.
Edit: But you know, I have a couple of locos I got back after more than 5 years of storage - long story but that as a few years back when I still had my ayout up and even though a couple were some of the irst locos I installed DCC in some 13 or so yeas ago, I just set them on the track and dialed up the address and they worked fine, mix of Digitrax and NCE decoders. Even my old PCM T-1 with Loksound 3.5, it ran fine without having to reset or reprogram it. I have no idea of when the ywere last used but the brass RS-3 I bought on ebay turned out to have an ANCIENT Digitrax decoder in it and it too ran fine. More recently I picked up a Mantua 0-6-0 that the seller said didn't run when tested. I saw the decoder peeking out in one of the photos - sure enough, it has a decoder installed and configured to disable DC. That one I did reset because it wasn't the road number and it wasn't 3 for an address. Plus a horrible install, wires floating around all over and the decoder WAS scotch taped in. Was. Now it floats around in the cab.
rrinker I suppose it IS possible that the bits were randomly scrambled whiel the locos sat unused, but really that shouldn't happen, the duration of the non-volatile memory in the decoder is 10's of years at least. I guess I will find out as all my locos have been in storage for over 3 years now at the least, and some longer. SO according to that, Tsunamis have doube protection against getting locked out, if you need to set CV30 to allow the lock feature to even work, and THEN set CV16 to the lock value you want. Which makes it double odd - because I can see accidently hitting 16 when you mean a different number when programming a CV, but to take two explicit steps, setting CV30 and CV16, there is practically no chance of that happening by accident. If it stays funtional, you're probably OK, but if it goes bonkers agian, definitely call Soundtrax. --Randy
I stop the locos and physically separate the A and the B units. Sure enough, the B unit is performing as intended, but the A unit is running slowly, very slowly, and not responding to commands, no lights, no sound.
I do everything that I can think of, kill the consist, make sure that CV19=0, reset to factory default, confirm that the decoder is not locked, re-program on the programming track, .............nothing works. More screwing around and all of sudden it works.
I guess that I need to call Soundtraxx on Monday. This is sure strange.
richhotrain I guess that I need to call Soundtraxx on Monday. This is sure strange. Rich
richhotrainSo, I emailed a detailed explanation of the problem yesterday to Athearn Tech Support.
Good luck with that. If it is now working it will be hard to fix.
"What to do when a tsunami sound decoder stops working properly?"
"I wil take Rip it out, and install Lok Sound for $800 Alex".
Ricky W.
HO scale Proto-freelancer.
My Railroad rules:
1: It's my railroad, my rules.
2: It's for having fun and enjoyment.
3: Any objections, consult above rules.
ricktrains4824 "What to do when a tsunami sound decoder stops working properly?" "I wil take Rip it out, and install Lok Sound for $800 Alex".
I just may.
richhotrain I was going to contact Soundtraxx this morning, but their web site advises Athearn Genesis owners to contact Athearn directly, saying that Athearn Tech Support employees are trained in the repair and replacement of Tsunami decoders. So, I emailed a detailed explanation of the problem yesterday to Athearn Tech Support. I hope to hear back from Athearn today. Rich
I was going to contact Soundtraxx this morning, but their web site advises Athearn Genesis owners to contact Athearn directly, saying that Athearn Tech Support employees are trained in the repair and replacement of Tsunami decoders. So, I emailed a detailed explanation of the problem yesterday to Athearn Tech Support. I hope to hear back from Athearn today.
Thank you for your inquiry. Have you ever purged all of the old addresses from your DCC system? We suspect that you may have an older "ghost" address in your system that is causing this problem.
richhotrain ricktrains4824 "What to do when a tsunami sound decoder stops working properly?" "I wil take Rip it out, and install Lok Sound for $800 Alex". LOL. I just may. Rich
LOL.
Wow , ESU has really gotten expensive...
If you don't believe ESU can have issues like any other electronic componet, I have a bridge for sale super cheap, and I accept credit cards.
At $800 I wouldn't be installing sound in anything
What's nice is the actual price hasn't changed in 5 years, I just checked, Modeltrainstuff sells Loksound Select for the same price I paid for one from them in 2012. But they've added features (more than just the Full Throttle) and have new recordings for many locos.
richhotrain richhotrain I was going to contact Soundtraxx this morning, but their web site advises Athearn Genesis owners to contact Athearn directly, saying that Athearn Tech Support employees are trained in the repair and replacement of Tsunami decoders. So, I emailed a detailed explanation of the problem yesterday to Athearn Tech Support. I hope to hear back from Athearn today. Rich I heard back from Athearn today, and here is their response: Thank you for your inquiry. Have you ever purged all of the old addresses from your DCC system? We suspect that you may have an older "ghost" address in your system that is causing this problem. So, tomorrow, I will Clear All Cab Memory and Clear All Advanced Consists which is what I have been advised to do on the NCE-DCC forum. I won't know if it works unless it happens again because the problem is sporadic. Rich
I heard back from Athearn today, and here is their response:
I had similar problems last night with a Stewart engine with a Tsunami. I had already sent the decoder back to Soundtraxx once for repair, it came back working OK except for lights so I piggybacked a light-only decoder on it. Anyway, engine ran fine for a while, but I noticed when running at slow speeds the break 'squeal' was constant. Tried resetting the brake sound sensitivity CV, and then the engine began running very slowly. I changed CV 66 and 95 from 70 to 150, and then it took off like a rocket. I changed it back and forth until finally it seemed to work OK. I ran it for a while, then when it was sitting on the track the lights started flashing. I tried to get it to move, it moved an inch or two and then stopped - no sound, no movement. I couldn't get it to do anything. My guess is it died (again). I've pretty much given up on it, and on Tsunamis. Doesn't matter how many great features you have if it doesn't work!
To make the night complete, several freight cars' couplers failed (in one case, the head of the coupler broke off; in another, the extended 'cushion' coupler box broke off), and when I tried to replace the Stewart engine with my 0-8-0 steam engine to finish the switching I had to do, I found it's couplers had disappeared. Front coupler had sheared off, rear coupler and cover for the coupler box were completely gone.
As Al Kalmbach said, "model railroading is fun!"
richhotrainI did some further testing yesterday including Clear All Cab Memory and Clear All Advanced Consists. Nothing I tried worked. So, I powered off the layout and went on to do other things. I returned later and powered up the layout once again. Called up the problem consist and it worked like a charm.
If you cleared all advanced consists, then how could you call up the problem consist and have it work at all? Or did you reconsist the problem consist?
maxman richhotrain I did some further testing yesterday including Clear All Cab Memory and Clear All Advanced Consists. Nothing I tried worked. So, I powered off the layout and went on to do other things. I returned later and powered up the layout once again. Called up the problem consist and it worked like a charm. If you cleared all advanced consists, then how could you call up the problem consist and have it work at all? Or did you reconsist the problem consist?
richhotrain I did some further testing yesterday including Clear All Cab Memory and Clear All Advanced Consists. Nothing I tried worked. So, I powered off the layout and went on to do other things. I returned later and powered up the layout once again. Called up the problem consist and it worked like a charm.
After I cleared all cab memory and cleared all consists, I did not recreate any consists.
What I meant to say was that the problem loco could not be corrected as a single loco at the time that I powered down the layout. When I returned and powered up the layout once again. the problem loco no longer exhibited the problem.
By the way, I passed this issue on to Athearn. Athearn suspects a bad decoder and has offered to evaluate it, but it is no longer under warranty so I would have to pay for repair or purchase a new decoder from Athearn. Not sure what to do at this point.
richhotrain By the way, I passed this issue on to Athearn. Athearn suspects a bad decoder and has offered to evaluate it, but it is no longer under warranty so I would have to pay for repair or purchase a new decoder from Athearn. Not sure what to do at this point. Rich
This weekend, I decided to take on the Athearn Genesis Erie F3A once again. Same old problems. No lights or sound, then lights and sound, then no lights and sound again. The loco was unresponsive to directional changes unless I lifted the loco off the rails and on again. Worst of all, the loco would only run at slow speeds.
So, yesterday and today, I tried everything including soldering the wires to the decoder in place of the plastic clips. Nothing changed. So, I bit the bullet and removed the Tsunami sound decoder and temporarily replaced it with a spare non-sound decoder that I had laying around. Everything worked perfectly. So, it was the decoder that was faulty. Now, I need to buy a new sound decoder so that I can permanently fix the problem.