Drive battery degradation . Houston, do we have a problem?

[STS134]

Just needed to point out the Tesla Model S that was publized to go 400,000 km in point of fact had the battery pack replaced twice and is due for a replacement again.

Apparently the Gen 1 packs in the 90D model had a lot of problems, and Tesla had to go through two revisions before it got better. First pack on that car was a Gen 1 pack, and that's why it was replaced despite having only 6% degradation. Second pack probably died from a lot of use.

 

[STS134]

Also ... very very strange ... storage at 0% SOC .. is what preserve at best the battery capacity .. (ok 0% SOC is consider 3.0v .. for cells that can be discharge down to 2.5v .. so not a real 0% SOC, but quite a low SOC)

Below 3.92V or about 60% charge, you get growth of SEI (solid electrolyte interface) at the anode. This pulls lithium ions out of the electrolyte and coats them onto the anode surface, in the form of lithium oxide and lithium carbonate. Above 3.92V, you start to become prone to the effects of EO (electrolyte oxidation) at the cathode. This becomes particularly problematic above 4.10V, which is why the Outlander cuts off the charge at 4.10V. 3.92V is thought to be the best voltage that keeps both SEI and EO at a low point. That's why whenever I go on vacation, I put the voltage of my battery at exactly 3.92V/cell, which is around 7 out of 16 bars on the gauge.

 

[elm70]

Just seen the Dog card from Anko towing (from FB) :

Lowest voltage per cell : 3.345v
Max delta voltage per cell : 0.095

The pack is looking both weak and abused :shock:

Possibly after this torture towing .. the pack is bad enough for be exchanged by warranty

I believe it should never happen in a PHEV such low voltage .. these are 3.8v nominal cell .. going down to 3.3 can be "hard" for the cell life ... so .. something to think and correct by Mitsubishi engineers

0.095v difference ... is really a huge unbalance ..

 

[anko]

For political reasons, I would say "just used by user, perhaps abused by BMU" :lol:

Note 1: SoC has never fallen below 22%. At that time, the ICE wouls revet to serial mode and SoC wiould quickly go up to 25%. Then parallel mode again. In earlier years, this happened between 20 and 22%. Would voltage not have been even lower then?

Note 2: Immediately after charging (at 6 amps because on campsite), SoC was below 100% but SoH was still reporting 26.1.

 

[elm70]

Pressing charge while towing, would prevent to go below 30% SOC ?

 

[anko]

Pressing charge while towing, would prevent to go below 30% SOC ?

No. In parallel mode the ice does not produce enough power to maintain speed (due to limited rpm) let alone recharge the battery. It needs to revert to serial mode for this. It does so only at 22% (20% before).

Below 65 km/h the car wil maintain minimum of 25% as normal.

 

[anko]

Does we have the definitive answer on this question:

Does the battery have a top side buffer / is the theoretical maximum voltage to which you could charge the individual cells in our battery higher than what the BMU allows?

Different approach to same topic: does a perfectly aligned BMU report the exact physical capacity of our battery or does it report the exact physical capacity minus a top side buffer?

 

[HHL]

Does we have the definitive answer on this question:

Does the battery have a top side buffer / is the theoretical maximum voltage to which you could charge the individual cells in our battery higher than what the BMU allows?

Different approach to same topic: does a perfectly aligned BMU report the exact physical capacity of our battery or does it report the exact physical capacity minus a top side buffer?

It depends what you define as a top buffer. If you assume that the LEV40 cells are in fact a nominal 40Ah capacity cell, but we know that most cars report 38Ah when new, I guess there is one.
The difference between charging to 4.2V and 4.1V may well be 2Ah, but we know the charging voltage is limited to 4.1V/cell in our cars.
I think Mitsubishi are being conservative, as higher charging voltages may result in fewer cycles. I guess we could soon tell if they started to re-program the BMU to charge to a higher voltage to enable a bit more capacity. I don't think that will happen.

 

[anko]

.The difference between charging to 4.2V and 4.1V may well be 2Ah, but we know the charging voltage is limited to 4.1V/cell in our cars.

Mitsu says, as part of a warranty claim the dealer must first determine the exact capacity by performing DBCAM. If the outcome is more than 26.0 Ah (Netherlands), there is no basis for a claim.

My line of thinking is this:

You have the real physical capacity (based on 4.2 volt) and you have the capacity made available to us by the BMU (taking into account SoH and and voltage restrictions programmed into the BMU). I don't see any sort / type of capacity in between.

If that makes sense, then apparently the warrant is related to the capacity made available to us by the BMU. By performing a DBCAM, they have not just performed a measurement, they have performed a repair. If the result of this repair (and earlier repairs) lasts less than 10.000 km, one could say the repair was not sufficient. Right?

 

[HHL]

.The difference between charging to 4.2V and 4.1V may well be 2Ah, but we know the charging voltage is limited to 4.1V/cell in our cars.

Mitsu says, as part of a warranty claim the dealer must first determine the exact capacity by performing DBCAM. If the outcome is more than 26.0 Ah (Netherlands), there is no basis for a claim.

My line of thinking is this:

You have the real physical capacity (based on 4.2 volt) and you have the capacity made available to us by the BMU (taking into account SoH and and voltage restrictions programmed into the BMU). I don't see any sort / type of capacity in between.

If that makes sense, then apparently the warrant is related to the capacity made available to us by the BMU. By performing a DBCAM, they have not just performed a measurement, they have performed a repair. If the result of this repair (and earlier repairs) lasts less than 10.000 km, one could say the repair was not sufficient. Right?

Not sure if it can be called a "repair", I think "adjustment" maybe the correct term.. anyway, you're walking a fine line and at the end of the day you will probably have to rely on their goodwill. We had a few noisy chaps here in Australia about this issue but they've gone a bit quiet.... Maybe Mitsi told them to shut up

 

[anko]

Adjustments. Repairs. Sure, fine line. Does that matter? The bottom line remains the same. Looking at the development for my battery, I would say my battery clearly meets the conditions (26 h at 160.000 km). For different adjustments have only temporarily masked that fact:

Every time DBCAM has been performed, the effect of the procedure has faded away within roughly 10.000 km and the trend has continued where it was interrupted. Maybe not 100% true for the very last one, but that has to do with the fact that last two procedures were only 10.000 km apart and I have not driven much since the last one.

So, when it comes to judging a claim, what is more likely to go by: output of latest DBCAM? Or day to day readings (which determine my experience with the car)? That being said, after adding the DBCAM readings and adding a trend line, it looks like that might drop below 26.0 Ah anyway. Question is: when?

I know, maybe we are all depending on goodwill. But maybe we are not.

A weird observation on the side: after the DBCAM, I got a report from the dealer stating 26.8 Ah (at 153029 km). The BMU reports result of the DBCAM to be 26.5 Ah (at 153030 km). After collecting the car, the current reading is also 26.8 Ah. Looking at my graph you can see that BMU Ah value has been higher than DBCAM output in all occasions.

 

[elm70]

Adjustments. Repairs. Sure, fine line. Does that matter? The bottom line remains the same. Looking at the development for my battery, I would say my battery clearly meets the conditions (26 h at 160.000 km). For different adjustments have only temporarily masked that fact:
.

160k km mark is close
0.8Ah to be permanently lost in 7k km is looking a "challenge"

Since the cost of battery replacement is very high, I'm afraid Mitsubishi may stick to the letter of their warranty

 

[anko]

Since the cost of battery replacement is very high, I'm afraid Mitsubishi may stick to the letter of their warranty

I have some concerns too ;-)

On the other hand, their promise is "26.0 Ah", not "26.0 Ah right after a DBCAM". Executing the DBCAM is only part of their promise, it is part of their complaint handling / analysis procedure. It does / should not effect the promise they make.

Under certain controlled circumstances, like a heated garage, slow charging, laptop hooked up (or should I say 'in a test environment' in the post diesel-gate era :mrgreen: ) the car seems to deliver 26.0 Ah and even a bit more. In real live circumstances it does not.

All wishful thinking, I know.

 

[shadogowah]

Hello all.

Just joined after browsing the forum looking for some answers.

My Phev has 71000 km and I bought it second hand 5 months ago (had 65000 back then).

I had already noticed that the range was shorter than expected but I was blaming myself for not being able to drive better (it is my first EV). Recently I bought a OBDC II tool and used the watchdog app and I was shocked to find out in the battery condition that the max capacity is at 75% already and it does not charge more than 28.6 Ah. I was expecting something close or even above 90%.

I am still trying to gather information about other owners but in any case I have arranged an appointment with the dealer in a few days.

Any link to websites or even posts in this forum related to this topic are more than welcome.

I know there are many Phev owners in the Netherlands (where I live), I am also interested in all the information regarding warranty. The Mitsubishi site mentioned something about a warranty if the capacity drops more than 30% and the car is not older than X years or has less than 120000 something km. I think I am going to be there soon.

 

[anko]

If yours is a MY14, MY15 or MY16, the warranty is at least 26 Ah within 160.000 km / 8 years. If yours is an MY17 or later, it is at least 28 Ah.

Be aware, this is not the number the Dog tells you now, but the number the Dog will tell you after a so called DBCAM procedure (which might lift Ah a bit).

 

[elm70]

Some observations ... which maybe are relevant for battery degradation:

- I'm using, like some of you, the keyfob hack for pre-warm my car in these winters days.
- My car get disconnected form the charger automatically at 6am ... actually based on my power monitor on the charge it is looking the car was almost fully charged at 6am ... 200wh consumed between 5:30 and 6:00 ... so it was at the end of charging it not it may finish the charging process. (my cheap WiFi power switch monitor, does make an history record every 30min)
- At 6:35 I activate the keyfob for warm up the car ... 10m at 4000w .. should be 666wh .. which are around 2Ah at 320v
- At 6:50 I start the car, and the dog reported 33.3Ah (SOH is 33.9, so 0.6Ah less them max ... but as well only 4.042v max)

In summer my car assumed to have 29Ah SOC, after 8h or rest, with a voltage of 4.048v ... so at 4.042v real SOC should be below 30Ah ... I don't expect much voltage increase after a rest from a low power 4kw usage

So ... it is looking to me that the consumption due to the keyfob hack is not tracked by the car.

And for some reason the expected capacity of the car is 2A above the real capacity (this explain, partially, why after 8h rest I can experience over 3Ah battery capacity lost)

Anyhow ... if I would be using my car, till EV range will go down to zero km ... then since the SOC known by the car is initially wrong, I assumed this would allow the car to discharge further the battery then the original design.

This could eventually degradate faster the battery ...

Maybe this issue is one of the reason why keyfob hack is not present by default.

I need to check if using the same function via WiFi App .. can help the BMU to have a more correct state of the real SOC.

 

[anko]

Well, that is interesting. Lately, I have been monitoring the result of each charging session closely. So, checking "SoC%" and "remaining capacity Ah number" after each charge.

Sometimes, it ends up above 100% although not much, sometimes it ends up below 100%, again not much.

Not to long ago I have purposely emptied my battery by running the heater for a while. Ended up with a rather high SoC% and decided not to do that anymore, as I currently do not want to see my SoH overinflated.

Yesterday, my wife took the car to work without preheating. Ended up with 99.13% (0.2 Ah below SoH). This morning I took it after preheating. Ended up at 100,73% (0.2 Ah above SoH).

I need to check if using the same function via WiFi App .. can help the BMU to have a more correct state of the real SOC.

Indeed, very curious what you find out!

Also makes you wonder: DBCAM uses the heater to drain the battery before charging again. And every time with DBCAM, my Ah number is temporarily over inflated :shock:

 

[anko]

Another observation: yesterday, charging took much longer than the 2.5 hours I got used to after my battery capacity went south.

It suggests my pack is badly balanced.
- Longer periods of cell balancing
- Apparently more effect balancing, as charge current actually went up after second cell balance period

This last bit, I have noticed before, in the very recent past.

 

[elm70]

...

Also makes you wonder: DBCAM uses the heater to drain the battery before charging again. And every time with DBCAM, my Ah number is temporarily over inflated :shock:

This would be one of my aim .. find a procedure for inflate the BMU SOH.

This would allow extra EV range at the cost of a bit of over discharge

Discharge down to 20% vs 30% SOC does not look critical to me, especially if by design there is a use case which the car can go below 20% SOC

 

[anko]

Discharge down to 20% vs 30% SOC does not look critical to me, especially if by design there is a use case which the car can go below 20% SOC

You would not only increase DoD under normal circumstances but also the DoD under extreme circumstances. So, you might go further below 20%. Unless you prevent these circumstances.