I know there is a way to reset a loco's decoder to its factory default settings although I've never had a reason to do so, at least not on purpose. That's why I was surprised to learn that apparently happened with one of my locos. I have a BLI Hudson#5275 that I had not run since at least last spring, maybe earlier. I lubricated it and set it on the track. I punched its roadnumber into the throttle and tried to start it up. Nothing. Tried the bell and the whistle. Still nothing. I gave it a little nudge to see if that would get it moving and still no luck. It wouldn't respond to any throttle command. I took a shot in the dark and entered the default decoder address of 0003 and it responded. I have no idea how it got reset. That's something I would have had to check the manual to figure out how to do so I think it would be strange if I would have done it inadvertently. I can't think of any reason I would have done so on purpose. Maybe I am losing my marbles and did it and just don't remember doing it. Is it possible that a loco sitting idly for a long time would reset to its factory defaults spontaneously?
John-NYBW Is it possible that a loco sitting idly for a long time would reset to its factory defaults spontaneously?
Not likely. I have had BLI Paragon decoders get "scrambled" usually due to a short circuit or other anomoly on the layout but in order to "factory" reset the button, jumper or reed switch, depending on the decoder version, would need to be closed THEN track or DC power applied to the decoder to complete the reset.
Usually this is accompanied by the "coupler crash" sound to acknowledge success. IIRC some QSI decoders would "talk back" and say reset.
I believe your engine was from the second run. Is it stock #064? I think that was a 2006 issue Paragon decoder.
Have you tried to readdress the engine (and lower some of the volume )
Good Luck, Ed
gmpullman I believe your engine was from the second run. Is it stock #064? I think that was a 2006 issue Paragon decoder.
Are you sure the decoder reset to factory default, or did the decoder simply lose its long address? That would not be unheard of with a dormant QSI decoder.
Whenever I am uncertain about the status of the long address, I put the loco on the programming track to read the long address and confirm that it is the active address, not the short address.
Rich
Alton Junction
I have seen decoders do some odd things. Usually after a short on the layout. My old club used a Lenz system. It had burst programming mode. During a train show a club member inadvertently used burst programming instead of programming on the main and set every decoder on the layout to the address he was programming. That was fun. Sometimes when someone was zero stretching a DC locomotive one of my BLI Blueline locomotives would constantly blow the whistle.
Could you have left the locomotive consisted? The consist address of 03 would do the same thing.
Pete.
I'm not sure what run this would be. 2006 seems to be about the time I bought them, but that is just a guess. I bought three of these Hudsons from Trainworld at a clearance price of $130. Even back then that was a heck of a bargain. Soon after they were being offered elsewhere at a much higher price so I believe mine were from a first run that Trainworld was liquidating prior to the second run.
wrench567 I have seen decoders do some odd things. Usually after a short on the layout. My old club used a Lenz system. It had burst programming mode. During a train show a club member inadvertently used burst programming instead of programming on the main and set every decoder on the layout to the address he was programming. That was fun. Sometimes when someone was zero stretching a DC locomotive one of my BLI Blueline locomotives would constantly blow the whistle. Could you have left the locomotive consisted? The consist address of 03 would do the same thing. Pete.
I do have a Lenz system. I don't think I've ever tried to doublehead my Hudsons so I don't think I would have consisted this one. Maybe I am losing my marbles and did a reset on this one and just don't remember doing it. I can't think of any reason I would have done it but my memory is not what it used to be. My long term memory is still sharp. I can remember details of sporting events or lines from TV shows from 50 years ago. Just don't ask me what I had for breakfast yesterday.
All QSI variants could be reset using the three step process, but some, not all, of BLI's offerings came with the magnetic wand to do the reset. Some required removing the jumper and inserting it again, but reversed. Weird, but it worked the one time I figured I'd see what happened.
Considering all that a decoder must experience to be reset, even if just CV8 = '08', it's odd that it would have reset on its own or due to some inadvertent process. Extremely odd.
selector All QSI variants could be reset using the three step process, but some, not all, of BLI's offerings came with the magnetic wand to do the reset. Some required removing the jumper and inserting it again, but reversed. Weird, but it worked the one time I figured I'd see what happened. Considering all that a decoder must experience to be reset, even if just CV8 = '08', it's odd that it would have reset on its own or due to some inadvertent process. Extremely odd.
I have several of those magnetic wands laying around although I've never tried to use one intentionally. I suppose its possible when straightening out the layout room I could have passed a wand over the Hudson. Is that all it would take to reset the decoder?
As I recall, many decoders respect the convention of resetting CV8 to a value of 8 to get a full 'factory reset'. IIRC at least some Digitrax decoders factory-reset everything but speed-step tables if you entered 9 instead of 8.
That is different from providing a hardware reset, like the reed switch on those QSIs does. I have long been of the opinion that you could easily desolder or remove that thing and substitute one of the little button contacts used for 'resets' on things like calculators that have the little hole for a pen tip or toothpick (the nonconducting alternative to the unbent paperclip) -- just remote the thing to where an unobtrusive hole accesses it. CSX Robert will correct me if necessary, but I believe this would be a NO and not NC momentary contact.
Small electronic aside: some of us remember the old days of 'grounding' things like computer parameter RAM/BIOS options to reset them. Those worked that way because they used battery power to maintain settings. Decoders almost all use NVRAM of some kind, which has to be actively reset or power-blanked (with reversion to hard-coded defaults somewhere).
John-NYBW ... I have several of those magnetic wands laying around although I've never tried to use one intentionally. I suppose its possible when straightening out the layout room I could have passed a wand over the Hudson. Is that all it would take to reset the decoder?
John, I don't think so. I could see the inadvertent proximity, but I believe the locomotive must be drawing power when the reed switch is closed via the wand. That's what I recall, the loco had to be 'on', at which you waved the magnet over a specific location on the tender's deck. Then, power had to be cut and restored to the rails, at which the loco would jerk and the decoder would state 'reset'.
John-NYBW I took a shot in the dark and entered the default decoder address of 0003 and it responded. I have no idea how it got reset.
Are you sure that it got reset and that you didn't just reprogram the address at some point? Also, could you have programmed CV29 (maybe to disable analog conversion or enable the speed table) and accidentally switched it back to the primary (2-digit) address?
As if this story wasn't strange enough, I discovered a few more oddities. Even though this loco's idling sounds were working, it wouldn't respond to the bell or whistle function keys. When I put the loco in motion, it would start up and the chuffing was in sync with the speed of the loco but after several seconds the loco would gradually slow to a stop. My first thought was dirty wheels so after cleaning them I tried again but the same thing kept happening. I was using my LH100 throttle which has push button speed controls. Normally I just press the acceleration buttons until I reach the speed I want and it holds that speed. I then discovered I could keep the loco running if I kept pressing the acceleration button but eventually it would reach top speed but to keep it going I had to keep rapidly pressing the button. As soon as I stopped, it would do a gradual deceleration to zero.
Eventually I put it back on my programming track and reentered the road number 5275 as the loco's long address. It accepted the address and when I put it back on the layout track, all the gremlins had disappeared. The bell, the whistle, the chuffing were all as they should be and the loco responded to the throttle commands for acceleration and deceleration as it should. I have no idea what went wrong with the decoder that would make it operate so strangely, but for now all is well. If anybody has any ideas as to what might have happened, I'd be interested in hearing them.
CSX Robert John-NYBW I took a shot in the dark and entered the default decoder address of 0003 and it responded. I have no idea how it got reset. Are you sure that it got reset and that you didn't just reprogram the address at some point? Also, could you have programmed CV29 (maybe to disable analog conversion or enable the speed table) and accidentally switched it back to the primary (2-digit) address?
See above explanation.
I just recently began working CV modifications on other locos and I am quite sure I never did it for this loco. I had to read the manual to figure out how to do it and I wouldn't have known how to do that the last time I had this loco on the layout.
The CV reset sequence for QSI is: CV 49 to 128, then CV 50 to 255, then CV 56 to 113, upon which you should hear a digital voice "reset".
I don't think there is any other way to reset a QSI decoder to factory defaults using CVs other than that specific sequence.
And the Reed switch will do it mechanically by flipping it via magnetic wand.
As Mike said, its possible the reed switch failed while in storage and got stuck to a reset position.
Also, check what Rich said. The loco might have just lost its long address and that not every other CV has been reset to facotry default (provided that you changed any other than long address)
BTW, if you try to reset it via the three step CV sequence and it doesn't reset the decoder, there is a good chance that the reed switch has failed or is stuck. You can gently flick it with a finger and it might unstick, or they can be removed via nippers.
IIRC, there are no consequences to removing a reed switch from a QSI decoder, but you might want to Google past comments on the subject if you get to that point.
- Douglas
"If anybody has any ideas as to what might have happened, I'd be interested in hearing them."
I do know that the very first thing I would do is fire up JMRI (or locate a friend who has it running with the right equipment) and get a good clear look at the way all the CVs read as configured.
Without that, you're trying to read a schematic with your elbow in Braille.