Decoder fried itself - what did I miss?

[tbdanny]

Hi all,

I completed the construction of an On30 kitbash, and put it on my program track to program.  However, when I did so, I got no response from the decoder (TCS WOW sound decoder - steam).  Upon testing it with the body removed from the chassis, I found that the decoder was heating up, so quickly that I almost burnt my finger on it.

However, I had tested the decoder on the chassis previously, with no issues.  It worked with the motor and power pickups.  The difference this time was that the body has the lights and speaker on it.

Following this, I did some testing.  There was no short-circuit between the power pickups or the motor, and the LEDs were working without too much resistance.  The only odd thing I found was that the speaker (an 8-ohm Soundtraxx speaker) was giving me 9 ohms of resistance.  Would that be enough to blow the decoder?

Given that I couldn't find a short circuit, I'm just trying to figure out what happened in order to spare another decoder the same fate.  Any advice would be appreciated.

[rrinker]

 Something on the body shell is making contact, if it worked fine with the body off. Or you have an uninsulated wire connection or two that get pushed together when the shell is put on. Or you have one of the lights connected to the wrong terminal. Speaker impedence rating is at AC, audio frequencies. When tested with a DC ohm meter, you'll never get the exact value, just close. And in general, higher impedence would REDUCE the load on the amplifier at the expense of lower volume. But getting 9 ohms on a DC meter is fine for what should be an 8 ohm speaker.

 What DCC? The program track shouldn't have enough current too make the decoder get burning hot. Something rarely mentioned, for systems that don;t have a seperate program track output, you should put a resistor in series on the program track to limit the current. Digitrax used to include the resistor with the Super Empire Builder sets which was the one without a dedicated program track output.

                     --Randy

 

[richhotrain]

I want to follow up on what Randy said.

In trying to read your post in a literal sense, after kitbashing the locomotive, you replaced the shell on the chassis and then tested it on the programming track without the lights or speaker in place. The result was that the decoder tested fine.

Then, you removed the shell and added the lights and speaker. You put the shell back on and re-tested the locomotive on the programming track. This time, the decoder fried and was extremely hot to the touch.

So, as Randy concluded, somehow you grounded the lights and/or the speaker to the chassis.

Let me ask you this.  Are you saying that the decoder is fried but that the LEDs still function properly when tested?  That would surprise me because the LEDs are fragile enough to blow as the decoder fries itself.

Can you restate your testing procedures?

Rich

[rrinker]

 Shorting an LED to the motor leads - hmm, that probably can;t be good. Shorting an LED to the track pickup though - that's actually not a problem (unless it's the blue wire that got shorted - that's very bad). Track pickup to LED to function (with resistor) is actually a normal way of doing things where you might not be able to get 2 wires to the LED. It's called half-wave lighting. Blinking and flashing don't work very well but for an on/off light it's an alternative.

 If spproriate resistors were used (1K at least), shorting something THROUGH the LED shouldn;t get anything warm - unless the resistor is bypassed. With a 1K resistor you have less than 10ma in the function line.

 I don;t see a frying decoder blowing LEDs - unless it somehow managed to triple the voltage going to the LED, again assuming a 1K resistor - double the voltage would still be under 20ma and the LED would be safe. Worst case would MAYBE be a couple of volts over the track voltage, voltage doubling even only occurs at the gaps between two different boosters if they aren't wired correctly. However the oppsite is very true - an LED or bulb blowing has been known to blow the function output. That flashbulb effect is the result of far too much current flowing through the LED which also tends to be more current than the function output can handle, however brief it may be.

 All that still comes down to there are wires touch where they shouldn't, once the shell is put on. Could be the speaker wires - or the speaker temrinals touching something. Or any uninsulated lead poking into the decoder.

                                  --Randy

 

[RR_Mel]

I have a 27Ω 2 watt resistor paralleled with a SPST toggle in series between my power pack and test track for just such an occasion when I’m doing maintenance or mods to my locomotives.  I also have a 12 volt GOW across the resistor to get my attention, a bright light is a no go a dim or no light says do it to it.

 

It’s saved me more than once in my more then 50 year model railroad career.

 

 

Mel

 

Modeling the early to mid 1950s SP in HO scale since 1951

 

My Model Railroad   

http://melvineperry.blogspot.com/

 

Bakersfield, California

 

I'm beginning to realize that aging is not for wimps.

[MisterBeasley]

Say "thank you" to TCS:

All decoders are covered by a one year, goof proof, no questions asked warranty. Decoders outside of the warranty which cannot be repaired can be upgraded for a nominal replacement fee.