TheRailwire
General Discussion => DCC / Electronics => Topic started by: Philip H on January 15, 2014, 10:35:56 PM
-
So thanks to two of my favorite etailers, I got a GP60M in the red and silver scheme, and a Digitrax DZ125IN decoder. Pried off the shell, carefully removed the FVM plug in board, inserted the decoder and set it on the programming track. My NCE Power Cab read it back just fine - ID"d the manufacturer and version properly and then rolled to asking "1=STD 2=CV 3=REG" so I figured I was good. Went into 1 and set the cab address (both short and long) made sure the speed step option was set to 28, and exited programming.
Set the thing on the main - headlights are on. Hit the headlight button to shut it off, and the loco starts moving at constant speed. hit the button again, it shuts down. The actual speed controls do nothing.
Move it back to the programming track, log back in, and do a reset to factory values by setting CV 08 to 009 (Which preserves the 28 speed steps). reassign the cab number, but once back on the programming track - it still behaves as before.
So, since I'm . . . challenged by DDC decoders and their manuals . . . anyone got an idea what's up?
-
You might have installed the decoder backwards. The way the pins are assigned, the power from the track will still go to the decoder (but with polarity reversed - which is not a problem). But the motor outputs will be hooked up to the headlights and the headlight outputs will be hooked up to the motor. Since the headlight outputs can only sink current (provide path to the decoder's ground), they will not activate the motor. But the motor outputs can make the headlights light up. :)
-
Thanks Pete , everytime I have installed ( 2 ) this type of decoders I was allways afraid to burn something as there no clear way to install it .
So I guess I was lucky 2 times to get it right !!!
-
On the decoders I have installed one of the ends in marked with a gold coloured 1 and same with the chip you take out. Make sure you have it the right way before you take the old one out. It sounds like you have it in the wrong way like Pete said.
-
We'll find out when I get home . . . but the little white boxes on the one side with the black dots are facing up which ishow the bad B/W pic in the decoder paperwork looked. :facepalm:
-
That's exactly how I installed mine, Philip. I compared it to the picture and went by the bits and pieces on the PC board. I hope reversing the chip works for you.
DFF
-
Thanks Dave . . . and since they were all black and white pieces neither of us can claim color issues.
-
My diagnosis was based strictly based on the described symptoms. I'm a bit puzzled that the motor gets powered when the headlights are turned on, but is till seems to point to backwards installation. If it is installed correctly, and it still behaves the way it does then I would say that the decoder is defective.
Of course, there might be some problem with the locomotive circuit board itself. Did you try the model on DC with the dummy decoder board installed?
-
I tried it and there was a note from my vendor indicating he had done the same. I gently pulled the decoder out, put a piece of laptop tape on the bottom the loco circuit board above the plug for good measure and gently reinserted. Now my NCE says it can't read any of the cv's and its flashing the headlights every time it tried.
Ugh.
-
And when placed on the main I can turn the head lights on and aff and switch ends with the direction switch
-
That's strange. At least I would say that it is now installed correctly. Let me ask you again: does the loco work fine in DC (with the dummy decoder installed)?
-
Let me respond again - yes.
-
Let me respond again - yes.
Sorry - missed that. :facepalm:
I'm still not 100% sure if you reinstalled in the original orientation, or flipped over. It is now behaving differently from the initial problem description.
Sounds like a damaged decoder. If you can dial in the address and control the headlights then at least part of the decoder is still working.
My Digitrax manual doesn't mention anything about setting CV8 to 009 preserving 28 speeds. It does a reset without resetting the speed tables. I would set CV8 to 008 to do a full reset.
Make sure the CV29 is either 006, and try it using a 2-digit address.
-
For clarity I flipped it over.
I emailed Digitrax' tech support with a lot of stuff cut and pasted from this thread, and they said it sounded like a short to the frame ausing the decoder to essentially shut down to protect itself.
So, anyone got an idea where such a short would be? There's factory installed Kapton tape under the motor-frame wipers, and i added a piece under the decoder between it and the board above. But other then perhaps a wheel wiper getting pushed sideways (which I haven't seen yet) I'm at a loss as to where to look for a short.
-
Philip,
But what is shorted to the frame? Digitrax wasn't very clear, were they? :facepalm
Do they mean that some components on the decoder are touching the frame? Isn't it isolated by Kapton tape?
Before blaming the problem on a short, why not do some checking? A quick test to check if both connections to the motor are shorted would be to try running the loco on DC with the decoder and the dummy plug removed. If it doesn't run, there is still a possibility that one side is shorted.
You would need an ohmmeter or a continuity tested to check for that. This is to be done with nothing plugged into the decoder socket. Check for open circuit between the center 2 pins and the outside 2 pins of the 6-pin socket. Since I don't know how the socket is oriented, I'm instructing you to check both sides. either way, they all should show open circuit.
So, you would test for open circuit between pin 3 and 1, 3 and 2, 3 and 5, 3 and 6. Then pin 4 and 1, 4 and 2, 4 and 5, 4 and 6. All should show open circuit. If that is the case, the locomotive has no shorts.
-
Philip - did you ever resolve this problem? If yes, I'm sure that all of us who contributed to this thread would want to find out what the solution was.
-
No resolution yet. I've been on the road for both work and helping my folks successfully through some medical issues. I'm home tonight and tomorrow night to reload the suitcase, so it may be a while til I get back at it.
-
Thanks for replying, and I'm sorry to hear about your problems.
-
With the discussion here (https://www.therailwire.net/forum/index.php?topic=32343.0 (https://www.therailwire.net/forum/index.php?topic=32343.0)) I pulled out the Kapton tape and went to town following Mike Fifer's suggestions. I now have a loco that doesn't show a short on the decoder, and will accept both loco numbering and operations commands - e.g. I can take it off the programming track and run it back and forth under proper control on the layout by its assigned cab number. Interestingly, I've lost the headlights . . . :facepalm: . . . which now no longer illuminate at all. Frankly, just to see the thing run is enough since its a beautiful model.
-
So, what is the diagnosis? Was there a short between the uninsulated chassis and components on the decoder? Was the original problem (where motor was activated by turning on the headlights) caused by the decoder installed upside down? Inquiry mind wants to know. :D
With the current headlight problem, if you reinstall the dummy decoder plug, do the headlights work ok?
If yes, then the problem is with the decoder. It can be hardware-related (blown headlight function outputs) or possibly software-based (some CV setting which reprogrammed the headlight functionality).
Another clue might be to (in a darkened room) remove and apply power to the decoder (by tilting one side of the loco off the track and putting it back on the track). When the power to the decoder is initially applied, most Digitrax decoders will produce a very short and dim pulse to the headlights. If you see that flash then there is a good chance that the decoder hardware is ok (the problem is most likely with the CV programming).
Welcome to the wonderful world of DCC. This is what happens when we start installing programmable computers (that is what a DCC decoder is) in our tiny toys. :)
-
Another clue might be to (in a darkened room) remove and apply power to the decoder (by tilting one side of the loco off the track and putting it back on the track). When the power to the decoder is initially applied, most Digitrax decoders will produce a very short and dim pulse to the headlights. If you see that flash then there is a good chance that the decoder hardware is ok (the problem is most likely with the CV programming).
I saw similar behavior when I put the loco on the layout after programming. SO I concur from my seriously limited experience that its a CV issue.
As to a diagnosis - I have yet to visually see anything that would have created a short between the loco frame halves or the motor or anything else. But adding the next round of Kapton - and replacing some I did earlier but applied poorly seems to have cleared it up.
-
I saw similar behavior when I put the loco on the layout after programming. SO I concur from my seriously limited experience that its a CV issue.
As to a diagnosis - I have yet to visually see anything that would have created a short between the loco frame halves or the motor or anything else. But adding the next round of Kapton - and replacing some I did earlier but applied poorly seems to have cleared it up.
So, you are able to see the headlights flash (even if they do not work in the conventional sense). That does seem to indicate a programming (CV) problem. Try resetting the decoder (on programming track, write a value of 8 to CV08 http://www.digitrax.com/tsd/KB675/resetting-a-mobile-decoder-to-factory-settings/ (http://www.digitrax.com/tsd/KB675/resetting-a-mobile-decoder-to-factory-settings/) ) . That should fix the problem.
The lights flash in the programming track during programming (align with the motor "twitching") because the decoder needs to create an electrical load to acknowledge the programming commands. So yes, that is also a good way to check the decoders hardware functionality.
As far as the other question goes, I guess that since the problem went away when you made sure that the chassis was properly isolated from the decoder (by applying more caption tape), the logic would dictate that a short between exposed decoder circuitry and the chassis was causing the strange behavior.