Honda Insight Forum banner

21 - 26 of 26 Posts

·
Moderator
Joined
·
7,181 Posts
What can you say about the negative affects of heat on the battery? I have so far been able to keep it under 100. When does heat become an issue?
It's always good to keep your battery as cool as possible. Ideally, you would keep it at around 72F. Anecdotally, it's easy to "see" that the batteries don't last as long in the extremes - Texas, SoCal, Arizona, while they tend to last the longest in the northern latitudes. But individually, you still have packs that have lasted 14 years in AZ and packs that only lasted 3 years in Maine. There are a lot of variables, but given a big enough sample size, that's the trend.

It sounds like you have this much more under control than most people. If you can keep your battery at under 100F during summer, that's a good thing.

The Ford Escape Hybrid implemented an AC System for the battery, and I have no doubt this has greatly contributed to their on-going excellent lifespans.

Also when using IMAC&C for Regen is there an amp range to stay under. I’m worry about hitting it with more than 30-40 amps for more than a few seconds. I try to mimic patterns I see the car naturally doing.
Hmm. I've always taken advantage of the manual nature to regen more than the car would at any given moment. I guess I've never thought too deeply about the potential side effects of this, mostly just focusing on that it allows you to maintain a higher SoC - even in the face of higher assist utilization.

How you use the battery is all cumulative, it all matters and definitely has an effect on things.. ie: cycling a battery with a square waveform will cause different wear than a triangle, etc.

It's hard to weigh these miniscule effects though. From my point of view based on what I know right now, higher regen is good because that's when you get equalization. Imagine each burst of 50A of regen as a mini pack equalization, if you're at high SoC. If you're at low SoC, it's basically all absorbed.
 

·
Moderator
Joined
·
7,181 Posts
Thanks... That would make some difference, about -0.073V, so my 2.4V at 6.5A would be about 2.33V were it an 18.6A discharge.



I only charged these sticks to ~3400mAh and then did those discharges. I had done charges to about 1500 before and saw no difference. I've been using my in-car pack between zero and about 10-30% SoC for ages (so max of about 2000mAh charged) and have only seen the voltage increase at lower charge states, though sounds like you're suggesting that it'd slump, sag... I would see saggage - if I charged up to ~75% and then came back down. The only way I can avoid saggage at low charge state is to use low charge state. Even one cycle up will result in saggage down low. Part of that I think is because I have old cells with less than new capacity...

Basically, whatever Richard's sticks have/had - call it voltage depression or 'crud' and 'high IR', etc. - those voltages should have been higher. The inflection point (where the curves start to flatten-out after start of discharge) for instance is about 6.9V - so 1.15V per cell. If you go back to my graphs and look for the voltage at about +1500 or so, it's about 2.52V for a cell-pair, subtract 0.07 to adjust for your higher discharge rate, and the comparable voltage is 2.45V, or 1.225V per cell. On a pack level that'd be 120 X 1.15V=138V vs. 120 X 1.225V=147V... That 138V is nearing assist-throttling territory... Actually, the lowest curve hits 6.6V after only about -250mAh discharge, so that's about how much unmolested usable capacity Richard would've been getting out of these sticks...

^This stuff applies to your following comment, too. Whatever we call it, sagging voltages such as those in your first graph might not cause a pack to fail out-right, but it definitely stunts performance, and I suspect it has compounding effects over time that at least lead to failure:





Probably. I use the term loosely ("voltage depression"). Any voltage sagging is 'voltage depression' to me. I don't know how to differentiate between, say, 'high IR' - which accompanies voltage depression, or just 'voltage depression/memory effect' in isolation, or what have you. I've seen just 'high IR' cells, without being VD, but I don't think I've ever seen VD cells without higher than normal IR.
The only thing wrong with Richard's sticks in the first graph is that they are at low SoC(well, I suppose balance could be better...). They aren't suffering from high IR, their IR is only about 10% more than when the sticks were new. I can't stress enough how much every single thing that a cell experiences through it's life can effect it's future going forward. Every single one. Not so much with healthy cells, but particularly when you are trying to recondition the batteries - operating at the extremes. It really sucks when a Reaktor messes up and say, terminates early(or fails to terminate) during a process. You can often see that anomaly going forward, forever..... It can actually cause sticks to not match each other any more.

You can't compare your data to that initial discharge straight out of the car unless you put your sticks through a very similar regimen. Let's say the previous owner let the batteries sit for 24 hours a day 3 days a week followed by 20 minutes of driving, and then another 3 days of sitting. That's 167 hours, 40 minutes of sitting a week.. for 2.5 years. It's just a completely different regime.

If the previous owner had rev charged the battery every 3 months, the outcome would have been different. I can guarantee it. I highly recommend people who don't drive at least 12,000 miles a year do this; it's well worth the few pennies in gas. It's actually not a bad idea in any case....

I do agree that voltage depression stunts performance, that is a given. But again, it's just exceptionally rare that I see a 1st generation Insight battery with what I call voltage depression. I'm having trouble recalling a single one...... That's pretty profound in my mind. I'll occasionally see a voltage depressed HCH1, HCH2 more rare, Accords are more frequent and Insight2s are nearly every one.

The hallmark of voltage depression is being able to hold current, but at a lower potential than "normal". Once you "use up" the lower potential material, the cell is ready to be recharged to normal(or close to it).The pack failed because it never got charged to full, I promise. I was able to bring it back, which is somewhat rare. It took a lot of effort though. A lot of effort on the charging side, not the discharge side. When the cells are operated at low SoC, the material that hasn't been used in a long time can become incredibly resistant to input. It takes special procedures to "break through" it and make it behave normally again, and often it's not possible.

I've never worked with 2nd gen cells before... But, first, what's the discharge rate here? If we compare my graphs to even the second one, which you say isn't voltage depressed, we see that my voltage/s still are much higher at comparable points. Mine doesn't hit 2.4V per cell pair (14.4V were it a stick-pair) until there's only about 400mAh of capacity left. If we find that point in your 2nd graph (about -5500mAh), we see that the voltage is only about 13.4V, so 1V lower than mine...

What do we call that? Surely higher voltage is better, right?
The pairs were discharged at 15A.

Your voltage observation is actually very interesting, I glossed over that fact. To me, the "fixed" part is in the overall look of the curve. Notice how the first discharge is very linear with no knee, straight down to 12V. After the first reconditioning cycle, the curve is back to "normal" visually. But you are right, there is still some element of voltage depression present because it continues to improve on subsequent cycles. That's what I had in mind when I threw out the "90% fixed after a single cycle" statistic. I think that's pretty close, maybe 85%.

Yeah, I do agree that higher voltage is better. It's all actually shockingly complicated. You can have sticks that have all of the same vitals.. IR, capacity.. but their voltage curves do not match. It's because of the things that are much harder to measure, the different "types" of IR.. polarization and such.... The differences arise due to the differences in use the sticks have seen....
 

·
Registered
Joined
·
135 Posts
Discussion Starter #23
I will Negative Recall anytime SOC gets around 50% or voltage drops to the 140’s. Always trying to Regen to keep the voltage up.
 

·
Moderator
Joined
·
7,181 Posts
Do you have a BCM that positive recals to 75% or 81%? Regardless, that puts your capacity at around 1600mAh... certainly not great, but still quite a ways from throwing the light.

One of the biggest problems with Honda's programming allowing a wide SoC window is that it causes a much wider range of experiences... something we're all familiar with.

It's very common to hear from people that declare "Everything was fine, then bam! The IMA light came on..". These tend to be people who don't use very much of the battery capacity at any given moment. On the other side of the coin, you have people who are acutely aware of every negative recal. These tend to be people in hilly areas, where battery utilization is higher...
 

·
Registered
Joined
·
1 Posts
Buying a Bumblebee and having it installed this week. 2001 Insight automatic with 187,000 miles on it. Had it since 2012. Tried charger but only got 6 months out of it. Keep you posted.
 
21 - 26 of 26 Posts
Top