Z
z_TravisE
Guest
Hi all! I came across this forum a while back and have been reading lots of interesting stuff for quite a while, learning quite a bit in the process. Now I think it's finally time for me to start posting. 
I've been having fun experimenting with some of my old "crap" NiMH cells, running them through a battery of tests (no pun intended) that was inspired by some of the threads I read, so I can get a clearer picture on their performance and health as well as collect lots of interesting data.
This must have been the first time I've ever had my C9000 (0G0C01; 1.47v termination; has Slot 1 cycle mode "repeating digits" bug) running for several weeks at a stretch. I've had it for close to three years now and recently encountered a curious occurrence that I've never seen before. I did some searching both here and on the Web and learned about various glitches but couldn't find any discussion about this particular quirk. The unit had been busy for about two or so weeks nonstop (estimated; I didn't track the exact period). At some point I went over to check on things and saw that the elapsed time indicator on all the current programs had gone a little haywire.
Of course, none of those cells were actually in the unit for anywhere close to that; it was more like a few hours. Other than that, everything appeared to function normally and the other results still looked correct. After another couple of weeks of continuous operation, the same thing happened a second time.
My guess is that the clock register the firmware uses is a bit too small and rolls over back to zero after a week or two, screwing up the elapsed time calculation. For the most part it's only a very minor issue in most modes, except that I would imagine that timed operations such as the break-in charge could get terminated early, and if you aren't around to notice, you'd get a bogus capacity measurement due to an incomplete charge and never even know it.
As long as I plan ahead and find a point every few days to remove all the cells and let the unit reset itself, it doesn't seem to happen. But if I have a very large set of cells to test and always replace them one by one, so that there is always at least one slot occupied at all times, then it appears that this glitch will crop up every couple of weeks or so.
Has anyone ever encountered this? It would be interesting to see if the other firmware revisions have the same behavior. It might be something to look out for if you're running continuous tests with the C9000 so you don't end up with messed up results. If my hunch on this behavior is correct, then the following should be able to reproduce the glitch if it exists on a particular unit:
I've been having fun experimenting with some of my old "crap" NiMH cells, running them through a battery of tests (no pun intended) that was inspired by some of the threads I read, so I can get a clearer picture on their performance and health as well as collect lots of interesting data.This must have been the first time I've ever had my C9000 (0G0C01; 1.47v termination; has Slot 1 cycle mode "repeating digits" bug) running for several weeks at a stretch. I've had it for close to three years now and recently encountered a curious occurrence that I've never seen before. I did some searching both here and on the Web and learned about various glitches but couldn't find any discussion about this particular quirk. The unit had been busy for about two or so weeks nonstop (estimated; I didn't track the exact period). At some point I went over to check on things and saw that the elapsed time indicator on all the current programs had gone a little haywire.
Of course, none of those cells were actually in the unit for anywhere close to that; it was more like a few hours. Other than that, everything appeared to function normally and the other results still looked correct. After another couple of weeks of continuous operation, the same thing happened a second time.
My guess is that the clock register the firmware uses is a bit too small and rolls over back to zero after a week or two, screwing up the elapsed time calculation. For the most part it's only a very minor issue in most modes, except that I would imagine that timed operations such as the break-in charge could get terminated early, and if you aren't around to notice, you'd get a bogus capacity measurement due to an incomplete charge and never even know it.
As long as I plan ahead and find a point every few days to remove all the cells and let the unit reset itself, it doesn't seem to happen. But if I have a very large set of cells to test and always replace them one by one, so that there is always at least one slot occupied at all times, then it appears that this glitch will crop up every couple of weeks or so.
Has anyone ever encountered this? It would be interesting to see if the other firmware revisions have the same behavior. It might be something to look out for if you're running continuous tests with the C9000 so you don't end up with messed up results. If my hunch on this behavior is correct, then the following should be able to reproduce the glitch if it exists on a particular unit:
- The unit must be operating for at least a week or two straight without being given a chance to go into sleep mode (where all slots are empty and the backlight goes out). Cells can be changed and new programs can be started as long as they aren't all removed at the same time.
- There must be at least one program in progress at the point where the glitch occurs (it only seems to affect the current operation; after that everything returns to normal again).
