Cooped Quick update for everyone. I received the decoder back from Soundtraxx yesterday, they had replaced a faulty diode. I put back in the loco and it runs a treat now, no sign of shutting down due to overheating! Dan
Quick update for everyone. I received the decoder back from Soundtraxx yesterday, they had replaced a faulty diode. I put back in the loco and it runs a treat now, no sign of shutting down due to overheating!
Dan
That is good to know. I have stored this link in case I ever see this issue again.
Rich
If you ever fall over in public, pick yourself up and say “sorry it’s been a while since I inhabited a body.” And just walk away.
That is great news Dan.
Pete
I pray every day I break even, Cause I can really use the money!
I started with nothing and still have most of it left!
Got it to overheat tonight in the loco that appeared to run ok........... added aluminium foil heat sink, still overheated. Factory reset as suggested as a last resort by Soundtraxx, overheated. I'm done, sending it back to Soundtraxx, now all I have to do is get the dealer I bought it from to give me a receipt! Perils of buying at shows. Reputable dealer with a website etc so I don't think it will be a problem.
The stall test I did was 0.7A not 0.07A. I pressed down until the motor stopped turning. The power supply is a basic Bachmann DC power supply. I checked the track voltage first then switched to amps and measured the current draw.
There may have been some binding originally which contributed as the loco was new, which may be why the problem seems to be easing, but it is now clear that the sound is the major contributor.
Randy,
Sorry, but I missed the OP stall test results. He is doing a stall test with 12 volts.and showing .07 amps. This is by pressing down on the loco. Is the motor still turning? For a true stall test should not the motor should be stoped? Also what type of power supply is being used? If the power supply drops amps when testing will this affect the reading? When I do a stall test I use a 3 amp power supply and test at 16 volts.
About the glue. I use tub and tile caulk ( water based). This has always worked for the decoders I have worked with. TSU 1000's and the TSU-750's .
OP, Is there any binding in the gear drive that could be causing over load?
It's possible that the decoder is defective, in that the heat sink plate inside the shrink wrap was not properly attached, or someone skipped the thermal paste. It just seems odd that a decoder that claims ot be able to handle 1 amp is overheatign at well under that (You shouldn;t be getting anywhere near .8 amp unless you actually stall the motor - running, especially with no train, it should be well under .5 amp). I haven;t run my Stewart for more than maybe an hour straight at a time, but it doesn't overheat. Then again I think it draws maybe .5 amp stalled. The decoder just sits over the rear truck, it's not pressed against the metal frame of the loco.
--Randy
Modeling the Reading Railroad in the 1950's
Visit my web site at www.readingeastpenn.com for construction updates, DCC Info, and more.
Cooped Yes, the decoder does 9 flashes indicating the 'Over temperature fault'. I had a response from Soundtraxx, they have recommended an aluminium heat sink so I'll try that (I wonder if some thinck gauage foil will do the trick.). I'll see if I can get hold of some of the paste Rich has suggested too. Dan
Yes, the decoder does 9 flashes indicating the 'Over temperature fault'.
I had a response from Soundtraxx, they have recommended an aluminium heat sink so I'll try that (I wonder if some thinck gauage foil will do the trick.). I'll see if I can get hold of some of the paste Rich has suggested too.
I read about the error code in the book, but have never seen it or heard of anyone experiencing it.
Soundtraxx is pretty good about getting back to customers. I don't see why aluminum foil wouldn't work, it's worth a try. Maybe fold or roll it into a coil or something similar. I would recommend against gluing the decoder in place. I use double sided tape to hold my decoders in place. As you can see, you never know when you may need to remove or relocate them.
Like Randy indicated, maybe another load test is in order.
Inspired by Addiction
See more on my YouTube Channel
Since this same decoder runs fine in another loco with no overheating problems, I still suspect somethign wrong with this particualr loco. Try the stall current test again. Several times, so that it gets tested at different points of the motor's rotation. Even better, put the jumper back in and run it on DC with the meter connected to read amps - although momentary spikes won;t show on a digital meter, like I mentioned. Clearly there is somethign different about this loco compared to the other one you tested the decoder in which ran fine. A TSU 1000 should not need an external heat sink for a motor that draws .8 amps stalled. A TSU 750 - yes, it needs a heat sink. .8 amps is actually overloading them.
After running the engine for a while tonight it seems that although it seems to overheat regardless of whether the lights or sound is on it seems to occur more often and have to cool down longer when the sound is on.
PennCentral99 As I read this thread, I had a few ???? You said the decoder overheated, did it flash an error code? Seems strange that you removed the tender shell (max air flow) and still overheated. Also, you placed the decoder in another loco and it ran fine. I am interested in the troubleshooting and hearing the outcome.
As I read this thread, I had a few ???? You said the decoder overheated, did it flash an error code?
Seems strange that you removed the tender shell (max air flow) and still overheated. Also, you placed the decoder in another loco and it ran fine.
I am interested in the troubleshooting and hearing the outcome.
The other loco I tried had the decoder in contact with a substantial metal chassis, so it's possible this was acting as a sufficient heat sink.
woodone Sounds like it is time to do a motor stall test to see just what amps the motor is drawing. Also you might try gluing the decoder the to metal weight for better heat transfer. The decoder taped the the weight might transfer the heat as well has gluing.
Sounds like it is time to do a motor stall test to see just what amps the motor is drawing.
Also you might try gluing the decoder the to metal weight for better heat transfer. The decoder taped the the weight might transfer the heat as well has gluing.
He did the motor current test and it is acceptable.
Read all the messages in this thread.
Also, no glue but use Radio Shack heat sink paste and strap the bottom of the decoder to a metal plate using small nylon ties. Suggested method in the Yahoo SoundTraxx DCC Group but mainly for the 750 decoder.
Ok, I took out the PCB and soldered in the decoder. Firstly I was very pleased to find I managed to solder all the connections properly (I am a very novice solderer). It ran ok for a while, but then the decoder overheated again
I'm going to contact soundtraxx today. I was wondering why then did it work ok in my other loco, the only thing I can think of it the position I had it in that loco was in direct contact with the chunky loco metal chassis so I may still need a heat sink in good contact............
Exactly. If there's a backup light, you need 2 resistors. Doesn;t matter which wire the reistor goes in, but it does matter which wire hooks to the blue decoder wire and which hooks to the white (for headlight) or yellow (for backup light). However, if you get it backwards, no harm done, the LED just won't come on.
Excellent, thanks. So I would simply wire a 1K 1/4 watt resistor between the decoder and the LED?
Dan.
The latest K4 Spectrum I have came with an LED for the headlight and that one is about 8 years old. I imagine yours is the same. You can wiggle the boiler front off very carefully and see it just behind the firebox door. I use a 1K 1/4 watt resistor with all the LEDs I use. I find that this value works good for the various track voltages my locomotives encounter between friends layouts and the clubs layouts.
Well, sounds like it could be time to say sayonara PCB then! Just one question, if I keep the existing headlight, do I need a resistor? If so what rating? I'm not sure if this loco has a bulb or an LED and no idea how to get at it!
Not sure when I'll be getting to this, the working week beckons and I think I should upgrade my soldering iron from the cheapo Home depot thing too.
The circuit boars houses capacitors and inductors to cut RF interference (and incidently mess with smooth DCC operation), as well as a bunch of diodes for directional lighting (not needed for DCC - they get bypassed when you plug the decoder in anyway), and possibly also dropping resistors for the LED headlight and backup light, which are simple enough to add yourself when ahrdwiring the decoder.
Uness thre is a wierd design, such as the circuit board being the only obvious way to keep things away fromt he moving parts, I withotu fail completely strip out the factory boards from any loco I put a decoder in and wire directly to the track and motor. Sometimes factory boards have a mistake, which causes no problems on DC but can fry a decoder. Sometimes the factory boards need lines cut anyway to work with DCC, so why bother? And sometimes the facotry boards are full of needlesly complex junk that isn;t even used in a aprticualr model (the P2K E units and FAs with the transistor circuit to alternate the dual filament Mars light - even on models that do not even have a Mars light, somes to mind here). Also those hokey plastic clips to hold the wires on - even if the board stays, I always solder those, the clips either loosen up or have stray bits of wire hanging out which may or may not short things. I generally replace all incandescent lights with LEDs, so there's relalyno reason to make the lighting connection removeable - the LED will outlve my grandchildren (should I even have any). The decoders I use all have a 9 pin plug so if somethign should happen to the decoder I can swap it without cutting any wires. Or even connect a dummy plug to convert the loco back to DC.
I have to agree with Randy on this. When I do an instal I normally remove the locos factory boards and wire in the 9 pin JST harness directly to the decoder and replace the lights with LEDs and resistors. Sometimes the extra circuits on the factory boards can have adverse effects on decoder functions. This becomes especially true if the factory board is set for directional lighting and I want the lighting to work in either direction. It also has the advantage of adding additional space to an already cramped space.
Interesting idea Randy, so what is the purpose of the PCB anyway? Here's showing my electronic ignorance! There's a bunch of black things, 2 little copper coils and a few brown/silver looking things. I can identify the wires leading to it, rail left, rail right (conveniently labelled L and R), then I'm guessing M+ and M- are the + and - motor terminals. The last 2 must be the headlight then I'm guessing..
I've had as close as look as I can and I can't find any more potential crosses on the PCB,and I've checked the resistance where I can and can't find anything obvious.
You could just rip out the board and connect the TSU 1000 directly to the motor and track wires and avoid any chanceof there being a problem with the circuit board. I suppose it's possible that if there is only liek one strand of wire bridging the connections then not enough current woudl get through to actually fry the decoder but would cause problems. Since a test shows it doesn't draw excessive current then it's not a motor issue - although a motor with a bad winding can cause spikes in the current draw that you might not see with a digital meter, and unless your stall test catches iton the bad winding, it won't show an excessive draw there either. But seeing as how there already was an issue with the circut board, I'd start there. The decoder's been proven to be ok, so no problem there, either.
Hi Pete
It's an 8 ohm speaker, 22mm x 58mm x 9.3mm so fairly large. I did try turning the volume down yesterday and that didn't seem to help so I discouted that idea. At your suggestion I have just muted the sound completely and it did over heat.
I have also just tried running this decoder, sound on in another loco and it went round and round quite happily with no problems, put it back in the K4 and it over heated. Hmm, loco issue??! Now back in the first post I did mention before I could run on DCC I had to fix a solder oops on the underside of the PC board. It was pins 4 and 5 where the solder joints were merged so left rail was going directly to the left motor, glad I found that before I smoked the thing! Anyway, could some continued voltage leakage be causing the excessive heating up? It's certainly a messy solder job on this board.
What size and ohm speaker did you use? How about muting the sound and see if it still overheats?
Hi MrB, if I don't get any joy today I will certainly contact them tomorrow, thanks.
At 12V DC it draws ~0.1A. I gently pressed on the loco and got a stall current of 0.7A.
I don't see any capacitors on the PC board.
When running on DCC with the Soundtraxx I haven't had it above speed step 50/128 yet.
Thanks again. I've been reading a lot and there are many people saying a substantial metal heat sink may be required, anythoughts on that?
Send an e-mail to Soundtraxx support. I had a bad Tsunami, basically bad out of the box. We swapped e-mails and they suggested things to try, but finally they said to send it back and they replaced it. They're very responsive.
It takes an iron man to play with a toy iron horse.