Speak EV - Electric Car Forums banner
1 - 17 of 17 Posts

·
Registered
Joined
·
3 Posts
Discussion Starter · #1 ·
Morning everyone

Quick question -

If the 88 cell battery has an approximate Ah of 46 when at 100% SoH, would the smaller battery of 80 cells found in the mid 2012 and onwards Ion and Zero be somewhere around 41.8 Ah when at 100% SoH? I assume all the 80 cell packs would have the revised LEV50N cells?

Thanks in advance - Toby.
 

·
Registered
Joined
·
5,232 Posts
Ah doesn't change with number of cells, however usable capacity in kWh will of course since the voltage is different. 100% SoH as best I can find out is 45.8Ah.

This is based on the fact that after running the traction battery replacement procedure in the diagnostic tool but before doing a battery calibration the figure changes to 45.8Ah as a default.
 

·
richi.uk
Joined
·
305 Posts
The Ah figure is the charge per cell, so it doesn't change. The manufacturer, Yuasa, claimed the difference is the N cells degrade more slowly and/or can be discharged a bit deeper. This was the rationale behind removing a module from the PSA clones.

If you can read Japanese, the LEV50N research paper is in my GDrive: PUBLIC richi.uk - Google Drive (i-MiEV/LEV50N paper 009_01_026.pdf)
 

·
Registered
Joined
·
3 Posts
Discussion Starter · #4 ·
Thanks very much for your replies.

My knowledge of batteries in EVs is limited to building e-bikes a few years ago

So if the Ah figure shown on e.g. Canion is per cell, would this then be an average Ah or would all the cells degrade at the exact same rate and so be accurate for all of them? Does any one know the difference in battery voltage when new between the 80 and 88 cell versions?

The reason I ask is that I'm currently looking at two 2013 PSA clones, one with the 16kWh, one with the 14.5. The 16kWh has 28,000 miles and SoH @ 94%, the 14.5 has done 50,000 miles but I'm not sure on the Ah. I assume it's lower as the RR on the dash shows only 56 miles compared to over 70 for the 16.

My average journey is under 15 miles a day so range isn't as important but I would want the car to last for as long as possible.

Do charging times and power input decrease as the battery loses its SoH?
 

·
richi.uk
Joined
·
305 Posts
If you're talking about the state of health estimate (SoH; aka the confusingly-named "Current Capacity"), it's effectively the estimated Ah behaviour of the weakest cell. The 80/88 cells are in a series string, so when the weakest cell discharges to the point where its open-circuit Voltage (or estimate thereof) drops below a safe threshold, the battery management unit (BMU) lights the turtle. Then when it drops below the second threshold, it limits power.

If they're PSA clones, they can't both be MY2013 cars. The one with 88 cells will be a MY2012 that's sat around for a while before being registered—assuming it's a PSA, not a Mitsubishi-badged car.

Assuming the dealer hasn't done a naughty BMU initialise, which would cause the SoH to temporarily over-read, I'd get the 94% car—that's an excellent SoH.

BTW, how are you calculating the percentage? What's the Ah value estimated by the BMU?
 

·
Registered
Joined
·
5,232 Posts
Thanks very much for your replies.

My knowledge of batteries in EVs is limited to building e-bikes a few years ago

So if the Ah figure shown on e.g. Canion is per cell, would this then be an average Ah or would all the cells degrade at the exact same rate and so be accurate for all of them?
Every cell has its own Ah (Ampere/Hour) figure, however the figure reported by the BMS in Canion is the usable Ah of the pack as a whole which as mentioned by @richi is determined by the wost cell in the series string of cells.

If only the cells would degrade equally - in practice that rarely happens, especially if you deeply discharge the cells a lot which tends to increase the spread between cells over time. On my Ion (now sold) there was a large divergence in the cells, at the point where I replaced some failing cells the worst cells were around 32Ah and the best were still around 40Ah. (The dealer diagnostic tool is able to report the capacity of best and worst cells)
Does any one know the difference in battery voltage when new between the 80 and 88 cell versions?
Nominal cell voltage is 3.7 so 3.7 x 80 = 296v, and 3.7 x 88 = 325.6v.
Full charge voltage is 4.1v so 4.1 x 80 = 328v and 4.1 x 88 = 360.8v.

You can tell from Canion which version of the car you have because you can see the number of cells listed on the cell voltage screen has 80 cells or 88 cells.
The reason I ask is that I'm currently looking at two 2013 PSA clones, one with the 16kWh, one with the 14.5. The 16kWh has 28,000 miles and SoH @ 94%, the 14.5 has done 50,000 miles but I'm not sure on the Ah. I assume it's lower as the RR on the dash shows only 56 miles compared to over 70 for the 16.
Forget about the RR remaining figure on the dash - it takes into account how the car has been driven recently, (it will read higher if driven gently in the last 15 miles or lower if driven hard for the same SoH and SoC) and also changes depending on whether the heater is turned on at the time the picture was taken.

Only go by the Ah figure reported in Canion.
Do charging times and power input decrease as the battery loses its SoH?
Charging time should in theory decrease as the cells degrade, but as a lot of the end of the charge session is spent balancing (where almost no power is drawn) you probably wouldn't notice.

Power input for AC charging won't decrease, it just won't charge as long. For rapid charging you may notice a drop in charging speed/power if the cells degrade - I certainly did. When I first bought my Ion it would charge from 20 to 80% in just under 20 minutes. By the time the battery had degraded the smaller capacity remaining was actually taking over 30 minutes.

This was due to a few cells that developed high internal resistance as rapid charging speeds are greatly affected by the internal resistance of cells, while the slower AC charging is not. I replaced the cells with high internal resistance and rapid charging speeds came back to near normal.
 

·
Registered
Joined
·
6,161 Posts
Thanks very much for your replies.

My knowledge of batteries in EVs is limited to building e-bikes a few years ago

So if the Ah figure shown on e.g. Canion is per cell, would this then be an average Ah or would all the cells degrade at the exact same rate and so be accurate for all of them? Does any one know the difference in battery voltage when new between the 80 and 88 cell versions?

The reason I ask is that I'm currently looking at two 2013 PSA clones, one with the 16kWh, one with the 14.5. The 16kWh has 28,000 miles and SoH @ 94%, the 14.5 has done 50,000 miles but I'm not sure on the Ah. I assume it's lower as the RR on the dash shows only 56 miles compared to over 70 for the 16.

My average journey is under 15 miles a day so range isn't as important but I would want the car to last for as long as possible.

Do charging times and power input decrease as the battery loses its SoH?
The 14.5 kWh battery will be a better long term bet: the earlier battery chemistry suffers random cell deterioration, which drags down the entire pack.
The SOH is actually a random number generator because it goes up and down. You need to drive the car over at least two cycles from 100 percent down to 10 percentish and see how the car performs. Watch for cell imbalance across these cycles.
 

·
Registered
Joined
·
6,161 Posts
The 14.5 kWh battery will be a better long term bet: the earlier battery chemistry suffers random cell deterioration, which drags down the entire pack.
The SOH is actually a random number generator because it goes up and down. You need to drive the car over at least two cycles from 100 percent down to 10 percentish and see how the car performs. Watch for cell imbalance across these cycles.
Your 2013 car will also have some subtle improvements compared to the 16kWh car which as DBMandrake has suggested, is probably an earlier car registered as new after sitting in stock. These improvements might run to a seat heater but also changes to the OBC, BMS programming, redesign of the inverter subboards, different guessometer algorithms etc.
 

·
Registered
Joined
·
5,232 Posts
The 14.5 kWh battery will be a better long term bet: the earlier battery chemistry suffers random cell deterioration, which drags down the entire pack.
The SOH is actually a random number generator because it goes up and down. You need to drive the car over at least two cycles from 100 percent down to 10 percentish and see how the car performs. Watch for cell imbalance across these cycles.
Actually the SoH figure on these cars is pretty dependable and doesn't suffer from random number generator syndrome like the SoH figure on the Leaf.

The figure almost never goes up, and it takes many cycles (two cycles isn't enough) for the figure to shift at all, with a minimum of 0.1Ah increments. I used to do 1000 miles a month and until I started having cell problems the figure would change at most once every month (1000 miles and about 25 full cycles) and always went down and never up.

I would have fairly good confidence in the reported Ah figure from one of these cars unless it's been tampered with.
 

·
richi.uk
Joined
·
305 Posts
I hate to disagree with m'learned friend, but the estimate does fluctuate up/down on my car—by as much as 2 Ah. I agree that it shouldn't, and the logic on my Outlander PHEV doesn't seem to.

But if older cells' behaviour diverges from the ideal model in the BMU, it causes confusion as to the actual SoC (or at least that's my theory). An alternative explanation is that my pack doesn't have a significantly "weak" cell per se, but it does have a cell that's notably stronger than the other 87—perhaps this is confusing the BMU?

Whatever the reason, if it's unsure of the SoC, it can't accurately measure the SoH, obvs. I see it most clearly after a lot of childish hooning around—the SoH estimate drops soon after. It then recovers in stages after more normal driving.
 

·
Registered
Joined
·
5,232 Posts
I hate to disagree with m'learned friend, but the estimate does fluctuate up/down on my car—by as much as 2 Ah. I agree that it shouldn't, and the logic on my Outlander PHEV doesn't seem to.

But if older cells' behaviour diverges from the ideal model in the BMU, it causes confusion as to the actual SoC (or at least that's my theory). An alternative explanation is that my pack doesn't have a significantly "weak" cell per se, but it does have a cell that's notably stronger than the other 87—perhaps this is confusing the BMU?

Whatever the reason, if it's unsure of the SoC, it can't accurately measure the SoH, obvs. I see it most clearly after a lot of childish hooning around—the SoH estimate drops soon after. It then recovers in stages after more normal driving.
Maybe a different firmware version ? I never saw anything like that.

If you have Excel (Google Sheets makes a bit of a mess of previewing it) have a look at the detailed battery health spreadsheet (and graph) that I kept:


The only upwards jumps I ever saw (about 1Ah) were soon after doing a forced battery calibration with Diagbox or after replacing cells. Otherwise it was a reliable downwards trend especially before the car started seeing cell problems around 40k miles.
 

·
Registered
Joined
·
6,161 Posts
Actually the SoH figure on these cars is pretty dependable and doesn't suffer from random number generator syndrome like the SoH figure on the Leaf.

The figure almost never goes up, and it takes many cycles (two cycles isn't enough) for the figure to shift at all, with a minimum of 0.1Ah increments. I used to do 1000 miles a month and until I started having cell problems the figure would change at most once every month (1000 miles and about 25 full cycles) and always went down and never up.

I would have fairly good confidence in the reported Ah figure from one of these cars unless it's been tampered with.
Sorry, SOH is a random number on the 80 cell batteries. It goes down incrementally, based on an algorithm that has nothing to do with actual degredation or measureable capacity, then it jumps up, to different values.

Now for a prospective purchaser, they won't see a series, they will see a single value, which is a RANDOM number because no one can be sure if it's near the bottom of the decline series, good recovery value, an exceptional recovery value etc.
 

·
Registered
Joined
·
5,232 Posts
Sorry, SOH is a random number on the 80 cell batteries. It goes down incrementally, based on an algorithm that has nothing to do with actual degredation or measureable capacity, then it jumps up, to different values.
Just stating my experience. I've posted my detailed log book of SoH over time, wheres yours ? :)

Trust me, the BMS on the Leaf is way, way worse than a triplet when it comes to "guessing". The value literally fluctuates up and down every day, on my Ion it changed once a month if that. (I took a lot more readings than the ones shown in the spreadsheet, for a long period of time I checked it a couple of times a week - I just didn't bother to log all the readings that were the same as the previous value)
 

·
Registered
Joined
·
6,161 Posts
Just stating my experience. I've posted my detailed log book of SoH over time, wheres yours ? :)

Trust me, the BMS on the Leaf is way, way worse than a triplet when it comes to "guessing". The value literally fluctuates up and down every day, on my Ion it changed once a month if that. (I took a lot more readings than the ones shown in the spreadsheet, for a long period of time I checked it a couple of times a week - I just didn't bother to log all the readings that were the same as the previous value)
I can only report on the 80 cell BMS. Frankly, a one off reading could be out by 3Ah compared to the best recovered value, which makes it random unless the previous owner has logged frequent values so one can see the series.
 

·
Registered
Joined
·
3 Posts
Discussion Starter · #15 ·
Thanks for all the replies and for linking your Miev pack Richi, and Excel data Simon, looks like I could use this as an average SoH for a 2013 14.5 kWh with 50,000 miles as there' s not much other info available, although it may be out by up to 3Ah with the newer cells as reported above...

My OBDLink LX just arrived today so I'll be able to dig a bit deeper when I view. In terms of individual cell voltages, what would be a red flag to look out for? As I understand, a larger difference in high and low voltage is a warning sign.

I might just let my gut instinct decide on the day... or flip a coin :)

What about a long test drive - would this be the best option to test battery health?
 

·
Registered
Joined
·
6,161 Posts
Thanks for all the replies and for linking your Miev pack Richi, and Excel data Simon, looks like I could use this as an average SoH for a 2013 14.5 kWh with 50,000 miles as there' s not much other info available, although it may be out by up to 3Ah with the newer cells as reported above...

My OBDLink LX just arrived today so I'll be able to dig a bit deeper when I view. In terms of individual cell voltages, what would be a red flag to look out for? As I understand, a larger difference in high and low voltage is a warning sign.

I might just let my gut instinct decide on the day... or flip a coin :)

What about a long test drive - would this be the best option to test battery health?
It will be a short test drive, probably no more than 50 miles. You can use the cabin heater to finish the battery discharge cycle.
 

·
richi.uk
Joined
·
305 Posts
I've not collected data religiously, but I do have enough readings recorded to say: my car's estimate has bounced around between 34.5 Ah and 37 Ah over the past six months.

Resolution of the data seems to be 0.5 Ah—not the 0.1 Simon's seeing. So it could well be there's different BMU firmware at work, or possibly two different PIDs (like there are two SoC values).

(MY2012 i-MiEV, 55,000 miles, realistic range driving carefully is 65 miles, data from OBDZero)
 
1 - 17 of 17 Posts
Top