Honda Insight Forum banner
21 - 30 of 30 Posts

·
Registered
Joined
·
1,108 Posts
I'm confused, because C2 looks like it only has 600 maH of capacity remaining. Am I misunderstanding?
 

·
Registered
Joined
·
7,746 Posts
Discussion Starter · #22 ·
Right, that's what I said a couple posts up. The pack sat for about a month, C2 self discharged and ultimately only ends up with 600mAh. But because of the faster self discharge, it was also cycled low, and cycling low keeps the voltage from becoming depressed. I'm using this as an example of the ameliorative impact of cycling low. C2 itself is not a good cell - because of the self discharge. Ideally all the cells would be equal, with even self discharge, and then you'd be able to cycle all the cells low and they'd end up with nice, lofty voltage curves, just like the C2 curve, only all the way up the charge state range.
 

·
Registered
Joined
·
7,746 Posts
Discussion Starter · #23 · (Edited)
Figured I'd add a perfunctory update of sorts to the above ideas, as I've been mostly continuing with, well, testing of pretty much the whole 'low charge state usage' theory that started this thread. But most recently it's just focusing on what's going on with this one faster self discharge cell and taking low charge state usage to the ultimate extreme -- near perpetual rock-bottom pack usage... I have to say something because it just blows my mind every time -- just how different, and better, these cells work than you'd think if you had just used them as they're normally used in the car...

Most recently, I've been driving for at least a week or two with the actual state of charge for most usage no higher than 20%, but I estimate more like around 13%, with one cell, the fast self discharge cell, at zero when the pack neg recals. I've basically been resetting nominal state of charge with OBDIIC&C to 40% when pack neg recals, and pack neg recals when this fast self discharge cell is close to empty - while other cells are probably around 13%... Last night I did some tap discharging and subtracted probably about 450mAh from the 13% cells, so they're closer to the fast SD, 'zero' cell...

Suffice it to say that I've been operating the pack really low and trying to get it even lower. It should be around 6% now...

Power Output
The main thing I want to emphasize here is the raw performance capability of the OEM NiMH cells. Even at this super low charge state and at cool temps (say about 50F to 68F), I can easily pound out about 70 amps at about 120-130V. Voltage doesn't plummet, it stays pretty steady, I can get the full 4 seconds or so of full throttle assist. When I do let it charge up higher, 30 amp assist happens at no lower than 140V, I think usually around 140-144V; 20 amp assist is usually above 150V, I'm often seeing something like 150-156V. This is higher than I used to see; it used to be more like 20 amp assist at the greater than 140V level. I have to add that I don't have to charge it very high at all, though, to see this kind of performance. It takes maybe 10-15% above the neg recal point to see these metrics...

Temp
I pound on this pack and the temp change I see is no where close to what I've seen in the past with various packs. I used to think it was 'normal' for the pack to increase in temp quite noticeably after some full assist and regen events. But now, I do see a little temp increase, but it takes quite a bit of usage, for example, probably something like half a dozen full assist and braking regen events and I might see a temp increase of a couple degrees F. In the past I've seen single 'full assist' events make temp increase right away. I never see that now...

Gets rid of Voltage Depression
Another thing I noticed is that, when I do let the pack charge up, pack voltage goes up to about 168-170V fairly quickly and stays around there during most of the charge. I'm pretty sure this low charge state usage does indeed eradicate voltage depression. So, if you go back a couple posts and look at the black and yellow discharge graph I posted, one cell is depressed (black) while the other isn't (yellow). I'm pretty sure that now the depressed cell, which is in this pack, can't be depressed any longer. If I discharged it its curve would look more like the yellow curve, lofty until the end...


Anyway, the performance I'm seeing is... fascinating. I wish I had a better handle on the electro-chemistry. What's interesting is that I just never would have expected these cells could put out so much power at such low charge state. There's this weird thing going on, this incongruity or something, with the difference between charging and discharging: charging is harder, discharging is easier; charging heats the cells more than discharging. Not all chemistries are like that; it's the opposite with my LTO cells and I think my boiler-plate lipo cells, too... I'm thinking that something with this incongruity might have to do with why perpetual high charge state usage causes problems, and on the flip side, why low charge state usage would end up, well, 'fixing' things in the first instance, but then performing better and preserving the performance... It just seems really weird that the cells would have such an easy-er time pulling protons from the negative electrode and pushing them into the positive electrode (discharging) on the one hand, versus pulling them from the positive electrode and pushing them into the negative (charging)... If anyone knows why that'd be the case I'm all ears...
 

·
Registered
Joined
·
7,746 Posts
Discussion Starter · #24 ·
FYI, the different behavior of the A03 BCM I talked about a while back, excerpt pasted below, turned out to be the 'cell difference' thing I describe at the end. Sources say that later Insights use Civic cells, and that this A03 BCM is programmed for those cells. But I was using it with early model, 2002 Insight cells, which need to be charged to a higher voltage to hit the pos recal point. Hence, the A03 BCM was inadvertently under-charging my early model Insight cells, pos recal-ing early. This is consistent with the graph I mention below: with the A03 BCM I was seeing a pos recal at something like 168-170V at ~6 amps. At the stick level that's about 8.45V. When I identify 8.45V on my graph, that corresponds to about 75% charge state for the Civic stick, but only about 40-50% for my Insight stick...

I guess I'll post that graph:
84195




I've been using/testing a later model BCM for the past few weeks, discussed a bit here: IMA Battery will not charge on level road at constant speed

Thought I'd mention here that this BCM - an A03 from something like 2005 and/or 2006 - seems to implement a lot of what I describe in this thread: it doesn't charge nearly as high and appears to concentrate usage below 50%.... I don't know, it's really weird how drastically different this BCM is compared to those others. I'm not sure what to think of the top-end threshold, for instance. In some sense it's too low... I also wonder if there's something different about my cells that causes a lower top-end. I don't think there is, but maybe there were slight differences in the cells used on later Insights that called for different programming, kind of like the purported differences between Insight cells/BCMs and Civic cells/BCMs... I did try to make graphs for Civic and Insight cells at one time, and those graphs do show that the Insight cells need to reach a higher voltage to reach a given high state of charge relative to the Civic cells. But I can't be sure that my test cells were truly representative... Plus, we're talking Civic vs. Insight here, not early Insight vs. later Insight...
 

·
Registered
Joined
·
7,746 Posts
Discussion Starter · #25 ·
Just wanted to post a quick thought/idea/question, I think it's mostly related to the topic in this thread.

I'm wondering just how much 'inefficiency' is the result of energy lost as heat versus energy that goes to 'side reactions' - whatever reactions that don't contribute to useful energy output?

I've continued more or less with this whole efficiency logging/testing thing, and these days, with my various manual management techniques, the 'loss' is so much less than it used to be. In the first post there's a graph that depicts the concept. Basically, back then, after say a few months of usage, net amp-hour input would be like two times the pack capacity to maintain a given charge state; now it's like maybe 25% extra (so '2 times' = 200%, now it's maybe 25%)...

I've wondered similar before, but now it just seems more vivid, real.

A couple other things make me wonder about this - heat vs. 'side reactions'. For instance, recently I did some bench work and it struck me that it doesn't seem like 'our cells' actually lose capacity - like the capacity is all still there (in general, usually), it simply needs to be extracted at super low current. So over time/usage the cells lose the ability to support even modestly high currents, such as on discharge, so the cell effectively loses 'capacity', simply because it can't maintain the voltage that it needs to at the currents that it's supposed to. But, use really low current and I think you'll almost always 'pull out' something close to the stock rated capacity (you can genuinely lose capacity, but I think that's secondary)...

This inability to support high currents actually happens at the bottom and at the top: there seems to be a range of charge state, say between 75% and 100%, and between 0 and 25%, for which the ability to support high current depends on whether these ranges are actually used. Use them and those ranges can be very small, like instead of 75-100 (25% of absolute capacity not used or usable) it can be 95-100 (5% not used or usable), and instead of 0-25 it can be 0-5. I don't think the stock management, though, actually implements a strategy to make you use the bottom and/or top. If it does it doesn't happen very often, I don't see it...
 

·
Registered
Joined
·
7,746 Posts
Discussion Starter · #26 ·
Considering what other experts have written about 'the weaknesses' of our cells, I think I have a rudimentary conceptualization of what might be happening with this 'lost capacity/charge state' in the upper and lower ranges - an essential 'squeezing' of the capacity, an effectively shrunken cell. And I think I have a slightly... deeper insight into the cause, at a slightly finer-grain level, though it's really muddy...

Given that I seem to be able to stretch the capacity (performance) of my cells if I simply use the bottom and top, it makes me think of the 'large crystal growth' idea. Larger nickel oxide crystals can't support high currents/large loads. If you don't use say the top or bottom 25% of absolute capacity, it seems like you'd always be picking the low-hanging fruit. I've written this before.

If you have a lot of large crystals of 'active material' at hand, you can achieve a semblance of performance within the performance demands of the Insight. The crystals are large and can't handle much current on their own, but if you have enough of them they'll do the job. Reactions take place on the surface of crystals - so you're basically depending on a lot of large crystals to do the work, and that work is taking place on the surface - which begets more surface reactions on subsequent charges and discharges. The crystals grow larger. Your capacity is effectively shrinking.

BUT, if/when you actually use top and bottom, you're forcing 'the system' to use harder to react material, to reach into deeper material. This means breaking-up the crystals. I think, when you use high and low charge state, you end up with smaller crystals that can support higher and higher currents... Voltage stays higher, under higher loads, you don't hit the management lower or upper voltage thresholds as soon, temps stay lower, etc etc...

As far as those "slightly deeper insights" go, I'm just thinking this: Our cells have a range of 'stoichiometry', voltage hysteresis - the main reaction doesn't happen at only a single voltage, rather, it can happen within a range. This happens day-in and day-out. If you charge a cell, the discharge voltage just after that charge will be much higher than if you had discharged after letting the cell sit for a day or so. It's amazing I never really noticed this before, it seems so profound. How you're using the cells can impact the efficiency, the power output and input capability.

If, say, you were to charge the pack - leave it charged high when you park at night - over night you'll lose a lot of charge state. When you go to use assist, the voltage will slump a lot... I'm not sure what the range/boundaries are, but it looks like there's a low plateau at around 1.20V to 1.25V, whereas the high, seemingly more correct/normal plateau is 1.31V to 1.34V... So, let's say these are right, so the difference, all else being equal, would be like:

120 cells X 1.20 to 1.25V = 144V to 150V vs.
120 cells X 1.31 to 1.34V = 157.2V to 160.8V

So the 'slumping' lost-charge-state pack will tend to gravitate toward about 144V-150V, while the 'correct' one will gravitate toward 157V-161V... Of course, in the car you've got losses due to resistance, temp, depends on current rate, et al - so the actual voltages will be different. But the 'idealized' voltages might be these.

Let's correct for resistance. Say total resistance is 3.5mΩ per cell, x 120 cells = 0.42 Ω for the pack.
At a 20 A discharge current, that 0.42 Ω resistance will reduce the voltage by: 0.42Ω X 20A = -8.4V.

The 'slump pack' voltage at 20A would be about 138.6V; the 'correct pack' voltage would be about 150.6V...

That's actually pretty close to what I see in the car, when driving - in general... I think that's the way it's calculated.

So... Drive, leave the pack charged high, come back to slumping voltages. Charge and then discharge on the drive and the range of charge state that gets used will return to higher voltages - on that drive. I think I'm saying that leaving the pack charged high isn't a good thing. One, you'll lose that charge over night anyway, so why bother. But two, I'm thinking that something in this process actually ends up 'crudding-up' the cells - it's the difference between my '200% extra charge' over that few month period vs. the '25% extra', that I mentioned in the previous post.

It's like, what happens on the day-to-day usage time scale ends up compounding on the longer, month-to-year time scale. What happens in 'microcosm' ends up happening at 'macrocosm'... I.E The losses I can see within a 24 hour period only - if I were to use the pack one way versus another - would end up compounding over time, were I to continue to use the pack that one way versus the other.
 

·
Registered
Joined
·
7,746 Posts
Discussion Starter · #27 ·
On my short drive this morning I was musing about some concepts/definitions that have really caused me much trouble understanding 'all this'. When faced with Insight NiMH cells and how they seem to work, it becomes doubly confusing.

"Capacity" and "charge state" - two things that, really I'm not sure I understand.

With Insight cells, you can charge to a high - what I've been calling "absolute capacity level" or "absolute charge state." The cells are say 6500mAh capacity. Charge them with 6000mAh current input, and all else being equal, you've charged them to 6000/6500=92.3%, right? Ignore efficiency and all that.

That value - 92.3% - however, I don't think it's technically the same as quote "state of charge" or "charge state."

Say I charge 6000 mAh input, a single cell, the voltage goes to 1.433V at rest. Were I to do some discharge on this cell immediately, with a moderate discharge current, the voltage might fall to about 1.31V under load and, after removing the load, it'd rebound to something like 1.358V. If however I let the cell sit over night, come back next day and do the same type of discharge, the resting voltage upon my return would probably be around 1.40V, the loaded discharge voltage would slump to around 1.25V, and the rebound voltage might be around ... maybe 1.30V to 1.32V...

This is what's so confusing - because this range of potential voltages can happen anywhere within say the 5%-95% 'absolute capacity-charge state level', yet, clearly it seems, one set of voltages reflects a higher 'charge state' than the other.

So, I could charge one cell with 6000mAh input, let it sit over night, come back and end up with 1.25V discharge-loaded voltages; on the flip side, I could charge another cell with only say 3000mAh input (or way less), discharge immediately and see say 1.31V discharge-loaded voltage.

The first cell might rebound to 1.31V; the second cell might rebound to 1.37V - the first has a lower voltage, but it's charged to a much higher absolute capacity level, that 6000mAh versus only 3000mAh. Which cell has the higher "charge state"?

So basically, there's two distinct concepts here that I've all too often conflated. I'm really not sure how that's dealt with, but it seems clear that they're not the same.

I can operate my cells at what seems like a higher 'charge state window' or something like that, even though they can be at a very low absolute charge state or absolute capacity level. And the flip side is true as well: I can operate my cells at what seems like a low charge state window - even though the cells are charged to a high absolute charge state or capacity level. The 'absolute charged capacity level' or whatever doesn't seem to matter, as far as I can tell, at all. Whether the cells operate at the higher charge state window or lower depends totally on how recently you 'charged up' that window...

I think, if I really wanted to understand this, I'd have to study the whole 'oxidation-vario-stoichiometry' concept. That seems to be key, that seems to be what determines the range of potential operating voltages. I just haven't been able to penetrate the...logic of it all, the jargon, etc. It basically requires a firm understanding of some basic chemistry concepts, like oxidation-reduction reactions. I can never even remember whether a compound that's oxidized loses or gains electrons...

NiMH cells are fully discharged when only Ni(OH)2 remains (full reduction); fully charged when only NiOOH exists (fully oxidized). Both have 2 parts 'O', but one, the NiOOH, has only a single hydrogen atom. The ratio between O and H is 2/1 in the charged state - I think that's the 'oxidation state'? - the 'II' you often see associated with it? Or rather, maybe it's the ratio between the Ni and the O, 2 parts O, 1 part Ni, yeah, I think that's right... NiOOH is the 'charged' state because it 'has room' for another hydrogen atom or proton or whatever; when you discharge the cell, a proton will move from the negative electrode, an electron will move through the current collector, that hydrogen ends up at the positive electrode and the NiOOH becomes Ni(OH)2. The addition of that H+ makes the 'NiOOH' less negative, it 'loses' an electron simply by virtue of gaining a proton, a positive charge. I think it is "reduced" - being 'less negative ' is the focus of the term "reduction"...

"Vario-stoichiometry"? I'm thinking the 'text book' case or condition treats the ratio of O to Ni as being 2, 2 parts O, 1 part Ni. But, the 'vario-stoichiometry' aspect means it isn't always or necessarily that perfect ratio. Is it more O or more Ni?... I can look for that in a book I have... Or actually, let me think about this. The 'Huggins' idea, that memory effect is caused by 'HNi2O3' is a similar concept. Here the ratio of O to Ni is 3/2 - so it'd be less O (3/2 vs. 2/1). The vario-stoichiometry I think might be the range from say 3/2 to 2/1... In conjunction with this 'memory effect' idea, the 3/2 is less charged, a lower charge state. You can have a cell or electrode that's, I don't know, in a sort of equilibrium state, when the ratio of O to Ni is within this range, not just at a single value, a single ratio... Ideally, I think, it'd be that 2/1; but when 'whatever' happens, it can be less than that and the cells don't perform as well...

I think I'm getting a little closer...
 

·
Registered
Joined
·
7,746 Posts
Discussion Starter · #28 ·
Kind of just sitting here meditating on stuff above. Wondering what 'highly oxidizing conditions' are and means, what that entails for voltages.

When you continue to charge cells that are nearly full, what happens? I've read a bit here and there but never had much of an understanding from it all. IF hydrogen continues to be withdrawn from the positive electrode, the electrode becomes more negative, it becomes more 'oxidized'. The ratio of O to Ni increases. Meanwhile, the hydrogen, the H+, is going to the negative electrode (eventually). The negative electrode is becoming less negative (reduced). The voltage, therefore, is increasing - positive electrode more negative, negative electrode more positive, the potential difference is getting bigger...

I know I've seen something called 'the proton-deficient limit' - at some point you've extracted all the H+ you can without, I guess, physically destroying the electrode. The electrode structure, the arrangement of the nickel, oxygen, and hydrogen, can hold together well enough with only so much removal of the hydrogen... I'm wondering what the actual voltages are that we'd need to see for, say, having the oxidation state (?) go above 2/1, having the ratio of O to Ni go above 2? And then, what kind of voltages would we/should we be seeing at the 'proton-deficient limit'?

There's an old graph/diagram in the 'Huggins' book that sort of illustrates this. Maybe I'll post that here:

89424


The problem is, I could never quite understand where his arrows are pointing to, exactly, how the values here translate to real-world voltages... The flat equilibrium plateau is supposed to be between 1.327V and 1.367V. I think this means that the normal reaction is supposed to happen within this voltage range - at or near equilibrium conditions, which basically means at really low current.

In the diagram, there's a sloped line toward the top with an arrow pointing to it, with a label that reads, "only NiOOH present." It looks like that has to be happening between the top value of the normal plateau, 1.367V, and what looks like a little above 1.50V. That 1.50V kind of makes sense - our own cells peak at about 1.53V at full, at about a 6.5A current. I think it actually gets close to that at even lower currents...

1.327V to 1.367V: for a pack of 120 cells, those values would be 159.2 to 164.0. That's really quite interesting as I almost always see my pack voltage 'equilibrate' to within this range. I can charge high, for instance, and see a higher voltage, say 170V, but just a little discharge and voltage will drop to around 163V, + or - a bit... Technically, supposedly our cells have an equilibrium voltage of 1.318V, that's a Honda-reported value, plus I've seen almost exactly that in almost every cell in a given pack at one time or another, say at a middling absolute charge state...

So... the question is, what exactly is going on when I see resting voltages above, say 164V? What exactly is present that makes the voltage high? Maybe it's, like, that extra proton removal - the electrode is sort of artificially oxidized, temporarily 'pumped-up' (or really the opposite, since the H+ is removed, not 'pumped' in)... And, perhaps, that makes it especially easy for the electrode to re-equilibrate with the electrolyte: That's another thing I've read, that Ni(OH)2 is not electronically conductive, but NiOOH is. When a cell, or positive electrode, is fully charged, the Ni(OH)2 is gone, and it no longer serves as a barrier layer between the electrode and the electrolyte; it's in contact with the electrolyte. So basically, H+ can migrate into the positive electrode and the charge state is lowered, the voltage drops...

I actually think this is probably close to right, close to what goes on, why we see higher voltages upon relatively high 'absolute charge-state charge', and why they tend to drop pretty quickly... Oh but, actually, I don't think it has to be "high absolute charge-state charge"; rather, it can be anywhere - that's the impact of the vario-stoichiometry (I think). I can see high voltages at low absolute charge state; I can see low voltages at high absolute charge state - it doesn't matter, for the most part.

So now the question is, So what? What would it mean, could it mean, that you have voltages above 164V more often or less often? Does it harm the cells? Is it harm to the cells? Here I'm talking about imaginary, pseudo-voltages, not necessarily the real-world voltages you see, but rather, ones that are adjusted for non-equilibrium conditions. In other words, you can see voltages well above 164V, but most of the increment above 164V is, say, due to resistance, like having a resistor in the cables. The 'actual' voltage in the cell would be that total minus the amount that comes from the resistance...

I'll have to come back to this later.
 

·
Registered
Joined
·
7,746 Posts
Discussion Starter · #29 · (Edited)
You know, it's interesting. I read this book, parts of it, quite a long time ago, reading the most relevant parts many times. But, I haven't really been thinking about it over the past few years, while I've say done my 'low charge state usage' experimentation and such. But now, having more or less banked some of my own real-world observations, and now having revisited a bit of that book, I'm thinking, 'Jeez, what I've been seeing sounds an awful lot like what's described in the book, in this chapter on 'memory effect'.'

I think I need to re-read that stuff.

In general, as I recall, the 'Huggins' idea is that 'overcharging' causes memory effect. That's basically what's being depicted at the top of this diagram, where there's an arrow pointing to a plateau, at a little above 1.50V, with a label that reads, "NiOOH, HNi2O3, O2 triangle plateau." That refers to a 'Gibbs triangle', or something like that, a method to suss-out the potential relations/reactions of different materials.

I think the idea is that, if you keep charging a cell when it's already reached the normal oxidation state, that 2/1 ratio I guess, it can become more oxidized via reactions other than the normal one. You're no longer converting Ni(OH)2 to NiOOH; you're converting NiOOH to something akin to 'HNi2O3' (there's different notation for these things, depending on what your focus is, NiOOH in this latter notation would be HNiO2). So instead of 'HNiO2' we end up with HNi2O3. One thing I'm not getting, however, are all the steps in between: I thought we had more oxidizing conditions, yet, how do we end up with a lower 'oxidation state' - 3/2 vs. 2/1?

In any event, when you take this HNi2O3 in context, it ends up lowering the voltage, and it's so much lower that the cell becomes essentially unusable in whatever equipment you're using, which equipment requires higher, normal voltages.

The remedy is deep discharge: this substance supposedly has a plateau at about 0.78V, once you discharge the cell I guess at least that low, it's supposed to get rid of the substance - and thus the plateau and the cell's overall lower voltage.

Not sure of the precision or accuracy of the voltages, but the general concept seems in line with what I've seen in the real world.

It seems plausible to me that, in general, higher operating voltages - say prolonged, consistent usage above a true, let's call it 1.40V, just above the highest equilibrium voltage on the normal plateau - will tend to produce more 'HNi2O3', or whatever, something akin to it. And that's what ends up shrinking usable capacity, lowering performance, etc. And, I don't think the cell has to be chock-full for you to see this degradation - you can see relatively higher voltages anywhere within the 'absolute charge state' range.

Something like this would go a long way toward explaining my whole 'inefficiency' angle. In the most basic way, my pack input and output was extremely inefficient when I mostly let the car do its thing, which was usually charging high - keeping the absolute charge state high, and keeping the voltage relatively high as a matter of routine. It didn't become more efficient until I started dropping the usage window way down, using low charge state, keeping voltages relatively low...

As I recall, it was pretty normal to, for one, leave the car over night with a pack at a high voltage, say around 168V. I wouldn't think much of using the pack at-will with voltages around 165V+, say doing braking regen. I think it was pretty rare to see resting voltages below around 156V - the BCM would usually be background charging, again as a matter of routine, at or before pack voltage got that low...

The flip side, with manual management techniques, is in effect a general lowering of the voltage window - probably much more between 150V and 160V, rarely leaving voltage above 164V max, usually probably closer to 161-163V...

So, in stock/OEM form, it's probably like 156V to 168V, manual mode is more like 150V to 163V.
Manual mode includes a lot of usage in the 144V-150V range as well, resting voltages.
OEM mode - I don't think it ever lets you use that range, that is, I don't think you ever see resting voltages in that range, the BCM would have background charged by then...
In OEM mode, you'll have pack voltage sitting above 165V a lot of the time; in manual mode, you'll only see this if you purposely charge the pack high...

[later...]
Yesterday I wrote this: "1.327V to 1.367V: for a pack of 120 cells, those values would be 159.2 to 164.0... Technically, supposedly our cells have an equilibrium voltage of 1.318V, that's a Honda-reported value..."

Since I've been using a lot of voltage ranges, it'd probably be good to 'adjust' the theoretical range found in that Huggins diagram to be more consistent with the Honda-reported equilibrium value. The Honda value is 1.318V; the Huggins range is 1.327V to 1.367V. So, a range that'd be consistent with the Honda value might be:

(1.327+1.367)/2=1.347V, 1.318V minus 1.347V= -0.029 -- adjust the range downward.

Instead of 1.327 to 1.367 it'd be 1.298V to 1.338V, or 155.8V to 160.6V for the pack.

No idea if this is even conceptually sound, but figured it'd be good to calculate such an alternative range anyway. I'm thinking different additives, such as cobalt, could be the difference between pure text book values and real-world ones. This adjusted range, though, actually looks pretty low, the voltages I see I think are usually closer to the Huggins range...
 

·
Registered
Joined
·
7,746 Posts
Discussion Starter · #30 ·
I did re-read some 'Huggins', but I can't say it amounted to much. One thing I think I did clarify a bit, though, is that initial question - capacity vs. charge state:

..."Capacity" and "charge state" - two things that, really I'm not sure I understand. With Insight cells, you can charge to a high - what I've been calling "absolute capacity level" or "absolute charge state." The cells are say 6500mAh capacity. Charge them with 6000mAh current input, and all else being equal, you've charged them to 6000/6500=92.3%, right? Ignore efficiency and all that.

That value - 92.3% - however, I don't think it's technically the same as quote "state of charge" or "charge state."
According to the definition of "charge state" in the Huggins, that 92.3% value there is technically the "charge state":

"The state of charge is the present value of the fraction of the maximum capacity that is still available to be supplied."

In a nut shell, I think there's a sort of vernacular, amalgamated usage of quote "charge state" that mushes together voltage and charge, that really doesn't quibble with or parse the concepts of current, power, and energy. Yet, when you're trying to find answers you really need to do the parsing...

"Charge" needs to be thought of like "charge" in a capacitor, I think - you've just got pluses and minuses, + and -, if you've got a lot of + you've got a lot of 'charge' (of course, you've got to think of these + and - in terms of the potential difference between the positive and negative electrodes, not just strictly + and - verbatim).

If you charge an Insight cell with 6000mAh of current input - maybe you have cruddy cells and can't discharge the cell passing another 6000mAh of current on the way out, yet it's simply because the voltage drops too easily, too fast to do it within your voltage window. If you drop the current and/or lower the voltage window, lower the end voltage threshold, say from 1V to 0.8V - you'll get that full 6000mAh (again, ignore minor Coulombic inefficiency for now). The quote "charge state" when you started was indeed 6000mAh/6500mAh=92.3%, it just might not be achieved within the operating parameters that you need or use...

Voltage, on the other hand, is...just a different thing. It makes sense to think of voltage and current together as some sort of master charge state concept, but 'charge' is one thing, and voltage is another. I don't think I know the language though, still, to deal with, say, a high charge state and slumping voltages vs. a low charge state and high voltages. Maybe it's 'energy state'? Or 'energy density' or something like that.

'Despite high charge state, Insight cells can have a variable energy and power density, depending on the oxidation state of the positive electrode', or 'depending on the voltage behavior of the cell'?

'Insight cells can have a higher power output at low charge state versus high charge state, it depends on the oxidation state of the positive electrode, which is variable'?

'The power output of Insight cells can vary solely based on recent usage of the cell, where low charge state can end up putting out more power than high charge state.'

The chemical energy in the cell: the configuration of stuff in the cell is what ultimately determines its energy output. You can 'charge' the cell with a fixed amount of current, say that 6000mAh, but it doesn't necessarily end up as stored chemical energy the same way in the same amount - because the configuration of that stuff in the cell can vary. It's kind of like stacking books in a bookcase: You have a pile of books - the current - and you have a bookcase, the cell. Say the objective is twofold: you need to fit all the books in the bookcase, but you also have to order the books from heavy to light from top to bottom (I know, just the opposite of what you'd really want to do, but bear with me).

If you do it fast you just try to stuff the books in place, with the shelves arrayed as-is. It's hard to get the right order - you get all the books in place (all the current, that 6000mAh), but your organization leaves a lot of heavy books low-down. If you take your time, you see that you need to increase the space of the top shelf - make it bigger for heavier books - so you lower the top shelf. Now, when you place the books you're able to fit the big heavy books on the top shelf, and you do the same moving down the bookcase.

You've fit all the books in both scenarios, but the latter scenario, where you re-configured the shelves, you were able to produce the highest energy configuration... Insight cells can have different shelf arrangements, and some are more conducive to... I think it's high power output, not necessarily higher energy output. So, maybe the bookcase analogy isn't quite right...
 
21 - 30 of 30 Posts
Top