DCC - Strange Problem with Quasami Soundtraxx Decoder in 50 ton Shay

[Peter O]

My 50 ton Shay (old #4) has been running on DCC without any issues since I purchased it in 2008. A couple of weeks ago I was ready to run her out of the basement to clean the tracks when I got sound but no movement. I moved her to the RIP track and hostled(?) up the 2 truck to take over the duties. No problem.

So this weekend I took a close look to try and figure out what may have happened. On the plug that goes into the third truck, I noticed that the gray wire (one of eight) was broken off. So I fired up the soldering iron and after some micro surgery (and a good deal of swearing) it was fixed. Put it on the rollers and... still sound (whistle, bell,etc.) but no movement.

So I opened up the coal bunker and started poking around. After much peering and poking, I found that the orange wire connect to the decoder (lower) board was disconnect, and might have been shorting with one o the other contacts on the four screw connector. Fixed that. Still only sound.

So I decided to reset the decoder, which is done by setting CV8 to 8. Keyed in #3 on the MRC Prodigy, sound and movement. Success!

So I program the loco to address #4, sound, but no movement? Back to #3, everything still works fine. Set to #2, everything works fine. Back to #4, no movement, but whistle, bell as before. Set to #24, where I left it, works fine. What's up?

I thought about some weird advanced consist problem, and set CV19 to 0. Made no difference and why would it only be effecting it when the address is #4?

I'm stumped. Any ideas?

Peter.

[Jim Banner]

Take a look at the standard numbering of the bits that make up a data byte
bit #0 = 2⁰  = 1 when turned on
bit #1 = 2¹  = 2 when turned on
bit #2 = 2²  = 4 when turned on
bit #3 = 2³  = 8 when turned on
bit #4 = 2⁴  = 16 when turned on
bit #5 = 2⁵  = 32 when turned on
bit #6 = 2⁶  = 64 when turned on
bit #7 = 2⁷  = 128 when turned on

The two digit address is programmed into CV01 in binary and is the sum of the values of the bits that are turned on.  Bit #2 is turned on in CV01 when you set the address to 4.  It is not turned on when you set the address to 1 (bit #0 on), or to 2 (bit #1 on), or to 3 (bit #1 and bit #0 on) or to 24 (Bit #4 and bit #3 on.)  It almost sounds like the decoder is not decoding bit #2 properly.  If this were the case, then one might expect address 28 would not work either (bit #4, bit #3 and bit #2 all on.)

Extending that, no address which required bit#2 would work.  These would include addresses 4, 5, 6, 7, 12, 13, 14, 15, 20, 21, 22, 23, 28, 29. 30. 31, 36, 37, 38, 39, 44, 45, 46, 47, 52, 53, 54, 55, 60, 61, 62, 63 and so forth.

Addresses that would work would not require bit#2 and would include: 1, 2, 3, 8, 9, 10. 11, 16, 17, 18, 19, 24, 25, 26, 27, 32, 33, 34, 35, 40, 41, 42, 43, 48, 49, 50, 51, 56, 57, 58, 59, 64 and so forth.

If you have time, it would be really interesting to see if at least some of these predictions are true.  Trying some of these addresses will not cure your decoder but will not make it any worse either.

Jim

[Peter O]

That's a brilliant piece of deduction and next time I'm home, I'll test your theory. Note that the sounds, whistle, bell etc. are working fine on address 4, only the motor drive is misbehaving. Don't know if that would fit your theory, but I'll give it a try.

Peter.

[jnichols]

Used use a Prodigy Advance on our HO scale layout at the store, and I would bet you that's where the problem lies. I have had locomotives become unresponsive to certain inputs, and only after deleting the locomotive from the throttles recall stack (press and hold delete once the address has been selected), and then re-adding it back to the throttle will it work. I don't know why this is the case, but I fear it has something to do with the way the Prodigy "looks" at the locomotives stored in the stack in the throttle. The MRC system stores these much differently than most of the other DCC systems, and I think it just gets confused sometimes.

[Peter O]

Well, it was the Prodigy, but it was also the user.

I used a virtual  throttle thru Decoder Pro to prove that the decoder was good and then it occurred to me that there was something I'd read about old style (command station based) consists. So I deleted the consist in the Prodigy and #4 was working.

The only time I'd ever used a consist was with two LGB Moguls which has MTS decoder which don't support decoder based conssiting. Somehow in setting up that consist, #4 had got involved.

Problem solved. Thanks for all the ideas and support.

Peter.

[Jim Banner]

Glad to hear it was only a bit of "finger trouble" and nothing seriously wrong with the locomotive or the decoder.

Jim