Honda Insight Forum banner
1 - 6 of 6 Posts

· Registered
Joined
·
8,642 Posts
Hey, eq1. I think you may be on to something. You made a remark in another thread, which I cannot locate, that you had some sticks that had been abandoned in a corner for a couple of years and that on retesting, you found that they had improved significantly....
I've peppered that around in various places, but the main place is the "impact of deep discharge" thread; the linked page below has the most of it:
http://www.insightcentral.net/forum...pact-deep-discharge-prior-grid-charge-10.html

I've been more or less working from the research conducted by some dude (Robert Huggins) at Stanford, who wrote a book titled "Advanced Batteries," debunking in one small section myth and conceptions about 'memory effect' in the positive nickel electrode. So, everything he explains seems to be consistent with what I've found; the problem that a deep (super deep) discharge fixes is 'memory effect' or 'voltage depression' - it's all about the electrochemistry. The voltage depression is caused by the formation of an 'amorphous' phase of nickel, hydrogen and oxygen during overcharge, and the way to get rid of this is to drop the voltage ("well") below that stuff's potential (0.78V). It just disappears upon recharge and the cell's capacity (and behavior) returns to normal.

It's always been a bit unclear to me, though, just how deep is necessary. The sticks sitting in my garage for a year were pretty much totally discharged - 0.63V per cell - so discharged that sticking the voltmeter probes on the stick would drop the voltage [edit: actually, it was the charger leads, not the DVM probes]. At one point I had been thinking that dropping voltage just below 0.78V, at whatever current, would be good enough, but I think it takes a bit more than that. I HAD done that, just below 0.78V, but I didn't see the radical improvement/transformation that I've seen - definitely in these totally self-discharged sticks - but also likely in a full pack I dropped to 22V as well as individual sticks for which I dropped each cell's voltage to something like 0.6V at 1.3A. This latter threshold still might not be enough, as the results on those 20 sticks seem to be a bit mixed...
 

· Registered
Joined
·
8,642 Posts
....I had tried a progressive discharge to 0V in the case of a couple of other sticks, just by placing a 12 ohm resistor across the terminals. One of those sticks showed significant improvement, but the other was killed. I think that probably shows that there are multiple failure modes.
I don't know, I'm not a big bible-banger of the 'multiple-failure-modes' rubric. I mean, sure there's multiple ways that cells can under-perform, fail, etc. Obviously they age and get worn down. But, it seems like 'memory effect' or 'voltage depression' as I have come to know and try to explain it take the largest, most consistent bite out of performance/capacity... Reflecting on the data I've generated with my original pack and a new 'betterbattery' stick, it looks like a little less than half the performance lost since new is simply due to age, normal wear and tear, whereas the other half is voltage depression (2500mAh vs. 4300 vs. 6500; or 1200 vs. 3000 vs. 5200 useable discharge capacity in car, voltage depressed vs. deep discharged vs. new '8Ah' stick, respectively). Granted, I still have a lot of sticks to work through, aspects to check, and real results would require a random sample of used sticks and this and that, blah blah blah etc...
 

· Registered
Joined
·
8,642 Posts
I wanted to discharge sticks to 0.78V/cell in slow yet automated fashion and minimize voltage rebound when the load is removed....
How did you come up with the 0.78V per cell threshold? I thought it was roughly 0.78V, too, but actually now, after reading and re-reading the material that got me started down this path, I think the author was saying 'well below 0.78V per cell' and possibly as low as 0.19V. It's based on a type of diagram (a 'Gibbs triangle') I only marginally understand. BTW, that material, most of it, is linked in the 'impact of deep discharge' thread, if anyone's interested, in one of the first few posts...
 

· Registered
Joined
·
8,642 Posts
....Seems the described memory effect at 0.78v does not look like a big problem for NIMH batteries as my pack did not show much charge between 140v and 1.8v . at 200mA discharge it took it 2 hrs and 45 min total .
So, you're generalizing your one-shot results to all NiMH batteries? Seems a little odd. But in any event, if I understand the rest correctly, I'm not sure it works that way. Sounds like you're expecting a lot of capacity under 0.78V, is that right? If so, it kind of seems like that's the way it should work, that the capacity is 'all there' still; it's simply locked-in at a lower voltage than normally useful. But I haven't seen this in any of my tests, and it may be a misreading, misunderstanding, of the research...
 

· Registered
Joined
·
8,642 Posts
....Driving the pack/cell lower isn't about getting more juice out of them, it's about driving the voltage lower, which redistributes the electrolyte throughout the cell thereby increasing its capacity....
I saw Cobb say something similar. That's different than the theory I'm going on. Isn't the electrolyte liquid? If so, why wouldn't it already be distributed evenly throughout the cell?
 

· Registered
Joined
·
8,642 Posts
....Final issue is the juice in the cells.
It is a concentrated Potassium Hydroxide fluid. New cells have several drops of extra liquid, and the inter plate separators are well saturated, but all the old used cells that I have dissected are pretty dry, and show sections of blackened very dry separators....
Great information. I had been wondering what exactly the electrolyte looks like - and imagining, erroneously, that it was a free flowing liquid inside the cells. But then, I had been examining some picts at your website of disassembled cells, and now your summary above: saturated separators with "several drops of extra liquid" in new cells; old cells "pretty dry"... Now I shouldn't have to cut open any batteries, at least for now...

One other thing I wanted to mention, after re-reading a bit of this thread. It has to do with 'how low do you go?', the difference, if any, between a battery drained via prolonged self-discharge vs. a manual 'super' deep discharge, etc. Of course I don't know anything along these lines for sure, and I'm no electrochemistry expert, not even close. BUT, my hunch is that it shouldn't matter what means are used to get cell voltages low, as long as all cells DO get low.

I think that's the critical difference: most of the manual discharges have been at relatively higher rates (i.e. relative to the rate of a self-discharge), on multiple cells in series, and usually not to very low voltages. The result is that not all cells reach the critical low voltage threshold - so not all cells get 'fixed', and performance doesn't bounce back as you'd expect or want. But a prolonged self-discharge DOES drain all the cells to that critical low threshold (I'm thinking 0.19V-0.78V per cell, yet really interpreting the research as indicating 0.19V, all else being equal).

Point is, based on my reading, once you see that critical voltage, it indicates that this or that 'phase' is gone - it's like a chemical, scientific fact: you see this or that potential/voltage because this or that chemistry is going on; if you don't see this or that voltage then this or that chemistry is no longer happening. So, I don't think it matters how the cells get low, so long as they get low. I mean, it matters for other reasons - like you wouldn't use a 50A load to take 120 cells down to 0 volts. But it doesn't, or shouldn't, matter in terms of getting rid of voltage depression, as I understand it...

One aspect that remains a bit unclear to me is the variation one might see in the real-world vs. the technical/theoretical values. For example, I'm using 0.19V-0.78V based on my interpretation of stuff in that book. And then, I saw about 0.63V for each cell in those sticks that sat for a year. And Keith saw similar values. Of course there will be variation based on measuring instruments; that's not really what I'm concerned about.

I'm concerned about the meaning of that range between, say, 0V and 0.78V. In my limited experience it has seemed as though there's very little difference, capacity-wise, performance-wise, functionally, between say 0.63V and zero. For example, you stick a 75w bulb on a pack with all the cells at 0.63V or zero, and in either case the bulb is going to be dark or go dark in seconds. The extent to which it doesn't simply reflects the extent to which some cells are not discharged as much...

Point is, there's wiggle room in these numbers - such as 0.19V per cell - when it comes to the realities of doing the work, discharging the packs or sticks or individual cells, measuring, etc... The technical/theoretical value is cut and dried - or it is in the sense that, if we've picked the right number, by nature it is cut and dried; generalizing that value to the work we have to do is a bit different.

If, for example, that critical threshold IS 0.19V, how do you ensure that each cell in a 120 cell series reaches it (barring 120 DVMs)? It seems like, you can only be sure if you take the pack to zero...

Anyway, I think I mentioned before that the great task seems to just boil down to discharging the pack or stick or whatever more or less completely. Light bulbs, fixed watt resistors - things that complete the task with lower and lower currents the lower the state of charge gets, seem to do it, seem to be the way to go. You try to limit the number and duration of cell reversals as much as possible, by whatever means; the lower the pack gets the less deeply reversed cells will be driven; etc. etc...
 
1 - 6 of 6 Posts
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top