Speak EV - Electric Car Forums banner

1 - 11 of 11 Posts

·
I'm not crazy, the attack has begun.
Joined
·
29,503 Posts
Sorry but what is the issue?



Nominal 34Ah pack drops to 26Ah after 150,000km .... and .... the reason that is an unrealistic pace is ... ????

Were owners really expecting better than 75% SOC after 150,000km?

"The traction battery on the Outlander PHEV comes with a warranty of 8 Years / 100,000 miles.
This includes a guarantee that if the traction battery capacity falls below 70% in the first 8 years/100,000 miles"
Hybrid Electric Battery Warranty - Mitsubishi Motors in the UK


Maybe there is a software fault by the look of it, yet still within warranty to date?

[Edit ... kWh has been corrected to Ah]
 

·
Registered
Joined
·
2 Posts
Discussion Starter · #4 ·
Softwsre issue yes. Can be reset to 100 % but we want it fixed so we dont need to worry about it. The real degradation is much less, and we loose EV range.
 

·
Registered
Joined
·
1,232 Posts
No thanks - but ill sign one if it bans them from using rapids!

bloody nuisance they are
But they have Chademo for rapid charging !? Cheers
 

·
Registered
Joined
·
4,781 Posts
Softwsre issue yes. Can be reset to 100 % but we want it fixed so we dont need to worry about it. The real degradation is much less, and we loose EV range.
I'd like to know how people think they have determined that the BMS is over reporting degradation ? Without taking the cells out of the car and individually measuring them on a discharger there is no way to conclusively know as you have to rely on the (supposedly incorrect) BMS for data.

The Ah estimate going back to "normal" after resetting the BMS means nothing if it soon goes back to the figure it was at before.

I have an Ion which has slightly bigger cells than the Outlander but otherwise the same cell type and chemistry (LEV50 instead of LEV40) and in my case the BMS consistently under estimates the degradation rate of the cells and has to make irregular large downwards corrections to the reported Ah capacity to keep up with the true degradation rate.

The degradation rates shown in the graphs are consistent with or perhaps less than what I see on my Ion which is now down to about 75% SoH at 52k miles.

Sorry to be the bearer of bad news but these LEV cells (either 40 or 50) just don't last as well as Mitsubishi originally expected/promised, and degradation rates are relatively high, on a par with the very first generation Leaf which also had high degradation rates.
 

·
Registered
Joined
·
2,039 Posts
Rather ugly graphs but UK seems to come out better than EU and Oz.

Thermal degradation?

Or do furriners actually charge their PHEVs rather than BiK them? ;)
 

·
Registered
Outlander
Joined
·
4,936 Posts
No thanks - but ill sign one if it bans them from using rapids!

bloody nuisance they are
The nuisance cars are the ones that occupy the space for longest, because longer occupancies create longer queues. At rapids an Outlander should finish charging in under half the time of an Ioniq.

If the driver intends to park for hours then they should use slower chargers, which are better for the environment and better for their car; but this says more about the driver than the car.
 

·
Registered
Joined
·
4,781 Posts
Softwsre issue yes. Can be reset to 100 % but we want it fixed so we dont need to worry about it. The real degradation is much less, and we loose EV range.
I've now had a chance to watch both the linked videos.

The first one by Andreas Kirch unfortunately shows that he doesn't know what he's talking about regarding how a BMS works.

He says at around 10:00 into the video that (paraphrasing his peculiar English slightly) "The BMU in the car is not able to see any real degradation of the battery, it calculates degradation, it estimates degradation, but it's not able to determine any degradation, it's not measuring and taking any parameters into account, to make this assumptions and estimations. So it's not like the software can actually determine and compare the state of health of the battery from previous times to now, and make assumptions of the health of the battery. That's not how it works. The BMU in these cars is super dumb, it calculates the state of health of your battery based on cycle and time calculations."

Sorry but this is just not true. It is true that the BMS has an "assumed degradation model" based on cycle count and age that it uses to estimate and interpolate the rate of degradation in between actual measurements, (all EV BMS's do this) however it is incorrect to say that it does not and can not measure the degradation at any time and that it relies solely on this estimate.

It needs a near full discharge/charge cycle from <20% to 100% to take a proper measurement using voltage estimation to determine the state of charge at each end as a reference/starting point. Depending on driving style this could happen very infrequently, perhaps only every few months. In between that it has to estimate based on assumed degradation.

Compare it to a meter reading for power/gas bills - some of your bills are based on estimates as a meter reading wasn't provided, but eventually a meter reading is taken and corrective action is taken if the true reading deviates significantly from the estimates. Likewise the BMS will take corrective action if it takes a measurement after a long period of estimating and the measurement differs significantly from the estimate - this is the source of sudden large jumps in Ah reading.

No BMS system for Lithium Ion cells can work properly without having an accurate Ah estimate/measurement of the battery - because the BMS must use Coulomb counting to keep track of the state of charge of the battery between about 20-80% where voltage estimation is inaccurate, and for Coulomb counting to work the BMS needs to know what the usable capacity of the battery is as it is counting the charge going in and out of the battery which is a form of dead reckoning.

If if doesn't know the usable Ah capacity then it will estimate the SoC wrong, and this could result in the battery "suddenly" running out when it was still reporting say 20%, or saying that it has run out when really it hasn't.

As for the second video by Gary, his car is down to 34.5Ah after 10.5 months (but how many miles ??) and he is claiming new should be 40Ah, even though as far as I know it should be 37Ah, and his most recent drop was 0.6Ah. Ok so that does seem to be relatively high degradation, but why assume it's a software problem rather than a faulty cell ?

Neither of these two guys have the equipment to diagnose this problem properly - they are looking at one single figure - Ah, and are not looking at the change in voltage of individual cells during a full discharge. It could very easily be a single faulty/weak cell dragging down the usable capacity of the entire pack and causing a high degradation rate, as the usable Ah is always limited by the weakest cell. If one cell is weak and degrades quickly the apparent/usable Ah capacity of the pack will likewise drop rapidly, due to that one cell.

What I don't understand is why when they see a higher than anticipated degradation they jump to the conclusion that the BMU is reporting incorrectly, rather than considering the possibility they have a faulty/defective/out of spec cell.

To diagnose this apparent rapid degradation properly they need to look at individual cell voltages during a full discharge to see if there are any "outlier" cells whose voltage drops a lot quicker than the others towards the end of the discharge.

Lots of blather in the video about performing a "BMU reset" and a "Cell Smoothing" procedure to try to fix this "software problem". A BMU reset will simply cause the BMU to go back to the original pre-programmed capacity for a new cell, or about 37Ah, and assume this is correct until a measurement says otherwise.

The only result of doing this will be that the car will over-estimate the range and health of the battery until you do a deep discharge and then the Ah figure will suddenly drop again after a measurement is taken. A BMU reset will do nothing to help if the battery is actually degraded. Likewise "cell smoothing" won't do a thing.

Cell smoothing is simply cell balancing - which is performed by the BMU automatically at the end of every full 100% charge cycle! So completely placebo.

The cell smoothing feature in the diagnostic tool is there to manually enable the cell balancers after replacement of individual cells to bring the cells into balance with each other as quickly as possible.

Normal charging/discharging will eventually balance the cells but if the difference in SoC of the replacement cells is too great it could take many charge/discharge cycles for them to be balanced, so the cell smoothing diagnostic function is there to quickly balance the cells without going through many charge/discharge cycles.

What I see from the two videos is people that know enough to be dangerous but don't really know what they're talking about, who are over analysing the Ah figure reported by the BMU, misunderstanding how it is calculated, and grasping at straws trying to explain away larger than expected drops in capacity as a software bug.

My gut feeling based on my experiences with the LEV50 in my car (which has a couple of weak cells) and also keeping abreast of other Ion/C-Zero owners cars battery state is simply that the cells in these cars are not very good and that some individual cells degrade a lot faster than Mitsubishi originally stated they would, (perhaps due to quality control issues during manufacturing/testing) and will be struggling to reach their warranty end period while staying above 70%.

It's highly likely that the cars owned by these two youtubers have a weak cell or two, but without diagnostic software like Canion that can show voltage graphs of all the cells, the problem can't be diagnosed correctly.

Keep in mind that a C-Zero/Ion with 50Ah (46Ah usable) LEV50 cells is 16kWh with a small 1100Kg car to push around - the Outlander weighs about 1800Kg, has smaller LEV40 cells that are 40Ah (about 37Ah usable) or about 11kWh usable - driving the car in EV only mode is putting tremendous stress and cycling on those rather small cells. At full acceleration they're running at over 4C discharge rate for cells with very limited passive cooling, with just a small fan into the pack.

Australian cars showing much greater degradation than UK cars on the presented graphs suggests that high cell temperature is probably a factor - high ambient temperatures are not going to create software bugs! But it will increase the cell degradation rate, that's for sure. So my money is on actual cell degradation rather than a software reporting bug.

I'd be annoyed at Mitsubishi for the poor quality of the cells and that they're not holding up, (perhaps due to them just being too small for the job and thus over stressed, or poor quality control of the cells) not chasing them for some imaginary software bug that is under-reporting the state of health of the battery.
 
1 - 11 of 11 Posts
Top