Lindqvist method : DIY BMU Battery Reset

[elm70]

Battery Reset Procedure from Lindqvist:

From Andy video EP248 : https://www.youtube.com/watch?v=ei6OdxynmUk


1 - Drive your battery low
2 - Disconnect 12v battery
3 - Wait 2h-6h
4 - Plug-In the car
5 - Connect 12v battery for 3-5 seconds
6 - Let the car fully charge
7 - Disconnect charger
8 - Reconnect 12v battery

Method above did not really work for me ... success rate for the method above around 30%

What did work for me was the "short" method

0 - Battery should not be fully charged (no need to drive battery low)
1 - Disconnect 12v battery
2 - Plug-In the car (use only the 10A charger included with the car ?)
3 - Connect 12v battery for ~3 seconds (please listen the click from the 10A charger, 12v battery need to be disconnected just before the charger make a click)
4 - Wait ~10 seconds .. check with the Apps the SOH, if it is not 40Ah, please , unplug the car from the charger and repeat from point 2)

If this does not work ... try to:
- Click on the stop button on the 10A charger
- Remove the 220v power from the 10A charger.

In my case ... after repeating few times this did work for my MY2013

Other method
Method D

Apparently there is an error that can trigger BMU reset when 12v battery is low ...
People over discharge the 12v or disconnect the 12v and use an external weak 12v battery ... in connection with charging process
I will get further information, and I will copy here.

D method : http://www.myoutlanderphev.com/forum/viewtopic.php?f=10&t=4581

 

[arnyswart]

Yes. There are reports that this works well.

If you do the software method, you can either do the reset in 2-3 min (same thing) and claim this happened. Or can also do cell smoothing and dbcam to get a more permanent option.

The leaving the previous data seems to reset the battery to where it was. I’m still evaluating how a straight wipe does in comparison such as this method.

 

[MadTechNutter]

Bit astonished people are so quiet about this here.

Could anybody who intends to do this procedure please read with the watchdog the cell voltage when their drive battery shows 'empty' before the procedure and the empty voltage value after a successful reset?

...and of course their trip odometer before and after, not the displayed EV range but the actual kms driven from fully charged in EV mode until the ICE kicks in with light foot driving and then read the voltage on the watchdog, once stopped at a safe location of course.
Please voltages only, no % or Ah.
Thank you

If this really significantly increases the range on older batteries then I can only imagine this is possible by deeper discharge.
If the true SoC goes down to 20%, I have absolutely no concern as this will not degrade the battery and should still have enough reserve for any mishaps.
It seem obvious that the 'improvement" will degrade over time as this seems to be the way Mitsubishi had it designed but then I would actually consider building in a Reset button to discharge over a small resistor the capacitor for the SRAM that stores the BMU data after I disconnect the 12V battery.
We do this the whole time to clear the CMOS for the BIOS in Personal Computers.
The resistor is important in case the reset button is accidentally pressed with 12V battery or charger connected. 470Ω -1KΩ should be OK and then hold the button for a 10 seconds.

What I find a bit questionable is why the battery has to stay disconnected while it is charging after the BMU data is obviously erased.

 

[Lon12]

Just tried it today.
12 volt off for 3:45.
Back on 6:30 later after a 240 volt 15 amp charge.
No difference in capacity.
The DC/DC keeps the car on until full charge then everything shuts off.
Power lift gate can't be opened after 12 volt disconnect. I tried to close it lightly to avoid that but it still latched and I had to climb from back seats to remove 12v lead.
I think my voltage was around 3.85 before charging.

 

[MadTechNutter]

Just tried it today.
12 volt off for 3:45.
Back on 6:30 later after a 240 volt 15 amp charge.
No difference in capacity.
The DC/DC keeps the car on until full charge then everything shuts off.
Power lift gate can't be opened after 12 volt disconnect. I tried to close it lightly to avoid that but it still latched and I had to climb from back seats to remove 12v lead.
I think my voltage was around 3.85 before charging.

Thanks for your feedback.
This is interesting: 3.85V looks like around 50% true SoC, was your battery indicator in the dashboard showing low/ empty?

The proper time to keep the 12V disconnected has not been established yet and in your case looks like is was too short to discharge the SRAM backup capacitor.
The leakage of these capacitors can vary wildly, so I would suggest waiting 8hours(overnight) 12V off before commencing charge if you want to try again .

 

[Lon12]

Was showing empty. Ran the heat with the App until it shut off.
Will try longer disconnect next time.

 

[miles556]

Tried it twice on 2018 PHEV sel US model but it did not work
1st time had 12v battery off for 3.5hours
2nd time 7 hours

Will try again with 14+ hours

 

[MadTechNutter]

Was showing empty. Ran the heat with the App until it shut off.
Will try longer disconnect next time.

3.85V showing empty, that is outrageous !

Below is the typical discharge curve of Li-Ion batteries from a reliable source and we are looking at over 60% SoC here:
https://batteryuniversity.com/index.php/learn/article/lithium_based_batteries

I am having difficulties finding discharge curves for the LEV40 but I assure you in the 25 years I have been dealing with Lithium batteries they all behave fairly similar.

I am bed-stricken ATM and don't have any App but when I am fit again I am going to drive the crap out of my battery and see where I am at.

 

[MadTechNutter]

Had my neighbour drive it to --.-- km.
Very tiny bit left on the battery symbol.

Crawled out of bed to measure with a dizzy head.
Watchdog reads with handbrake pulled and selector in D and head lights on: 3.825V
Cell difference is 0.011 and NOT 0.005, so I wonder if the BMU and ECU are 'arguing' about that?
According to the display from 29Ah there is 8.4Ah or 29% left.
...but at that voltage I should have around 60% left.

Pulled the 12V.
After that I also pressed the START button for 20 seconds, just had a hunch. We do that to clear RAM in MacBooks and some other laptops
I read somewhere that by disconnecting the 12V for a longer time the 15l petrol per 3 month counter gets reset !
So it is now obvious to me that the ECU has a volatile memory section that can be cleared too, by removing the 12V.

This might be the reason why it might not work with newer models as they changed something in the ECU to make the stored values resident.
This could simply be a matter of a software change to write the values in the resident EEPROM rather than the volatile SRAM.
Compared to the ECU the BMU would be a relative dumb device where you can't simply change it's Firmware.

I probably can only keep it disconnected for 5 hours before I will have to sleep and put the charger on overnight. Tomorrow at 12 noon I will know more.

 

[luispcorreia]

Been following your posts very closely!

Keep’em coming please!

 

[MadTechNutter]

Will do

Finally somebody posted useful information on the comment section of the Youtube link in the first post just an hour ago, scroll down to sportster999,
who claims after the procedure you can use the battery down to 3.6V and yes I agree, that is perfectly fine.
10kWh charge instead of 8kWh
The obviously fake 38Ah SoH drops 0.1A per day, will have to wait if and where it stabilises.

Here is where I plan to do something different.
If the reset works for me I will try to charge it only up to 4.05V max cell voltage, roughly a 90% charge.
This might leave me with 3km less range but I have gained 15km or more with the procedure.
1 - There are many claims and I know it from experience with laptop batteries that charging only up to 90% will double the cycle life of Li-Ion cells.
2 - Not achieving a full charge might not reduce the SoH. If a charge has never finished and properly terminated the 'counter' in the software might not detect it. Again just a hunch but I will try it.

 

[MadTechNutter]

After charge has started and 12V disconnected again, the green park lock light next to the shifter stays on. Car is locked.
Is that normal?
I never charged at night perhaps I didn't notice.

Alarm LED blinks like normal, hope it doesn't go off in the night because the 12V from the charging circuit might not be as stabile as the battery.
My last car was 22 years old and when the 12V would suddenly drop only 0.05V from a door light that would trigger the alarm. I don't know if modern alarms still work like that or just use the ultrasonic sensors these days.

 

[luispcorreia]

After charge has started and 12V disconnected again, the green park lock light next to the shifter stays on. Car is locked.
Is that normal?
I never charged at night perhaps I didn't notice.

Alarm LED blinks like normal, hope it doesn't go off in the night because the 12V from the charging circuit might not be as stabile as the battery.
My last car was 22 years old and when the 12V would suddenly drop only 0.05V from a door light that would trigger the alarm. I don't know if modern alarms still work like that or just use the ultrasonic sensors these days.

There’s people on the German ev forum performing the same method that also mention the alarm light behavior. One of them had the alarm turning on a couple of times if I remember correctly.

But others had assured him that that behavior is to be expected.

 

[MadTechNutter]

There’s people on the German ev forum performing the same method that also mention the alarm light behavior. One of them had the alarm turning on a couple of times if I remember correctly.

But others had assured him that that behavior is to be expected.

Could you provide the name or link(leaving out the http:/) to that forum?

 

[elm70]

I did try on my 2013 PHEV but it did not work for me

I did read from success "stories" that 12V 3 to 5 seconds reconnect for start the charging process .. in theory it should reconnect only until hearing the first click (in my case I let 3 clicks before disconnect the 12v)

I will try again ... next weekend.

 

[elm70]

Will do

Finally somebody posted useful information on the comment section of the Youtube link in the first post just an hour ago, scroll down to sportster999,
who claims after the procedure you can use the battery down to 3.6V and yes I agree, that is perfectly fine.
10kWh charge instead of 8kWh
The obviously fake 38Ah SoH drops 0.1A per day, will have to wait if and where it stabilises.
...

This does not sound good.

Yes if the SOH is artificial set to 38Ah while the real battery SOH is only 30Ah or less ... yes it is possible to "over discharge" the battery .. like the guy sportster999 is reporting

I think bring regularly down to 3.6v is not good.

Even worst is to make a long trip in hybrid mode with the battery left all time at 3.6v .. instead at around 3.8v as designed

 

[MadTechNutter]

Will do

Finally somebody posted useful information on the comment section of the Youtube link in the first post just an hour ago, scroll down to sportster999,
who claims after the procedure you can use the battery down to 3.6V and yes I agree, that is perfectly fine.
10kWh charge instead of 8kWh
The obviously fake 38Ah SoH drops 0.1A per day, will have to wait if and where it stabilises.
...

This does not sound good.

Yes if the SOH is artificial set to 38Ah while the real battery SOH is only 30Ah or less ... yes it is possible to "over discharge" the battery .. like the guy sportster999 is reporting

I think bring regularly down to 3.6v is not good.

Even worst is to make a long trip in hybrid mode with the battery left all time at 3.6v .. instead at around 3.8v as designed

Charging it too often so it stays more often at 4.1V is a lot worse.
It is possible that at 3.6V a lead footed driver will bring the voltage momentarily down to 2.75 but the ICE will kick in before that.

Not sure why hybrid mode is a problem, the ICE is generating electricity and keeping the voltages up. I trust Mitsubishi has that sorted properly.
But artificially 'degrading' batteries much to early as it is reported everywhere is wrong.

More about that in my next post.

 

[elm70]

Charging it too often so it stays more often at 4.1V is a lot worse.
It is possible that at 3.6V a lead footed driver will bring the voltage momentarily down to 2.75 but the ICE will kick in before that.

Not sure why hybrid mode is a problem, the ICE is generating electricity and keeping the voltages up. I trust Mitsubishi has that sorted properly.
But artificially 'degrading' batteries much to early as it is reported everywhere is wrong.

More about that in my next post.

Charging is never above 4.1v .. and has nothing to do with SOH

About hybrid mode ... if you start a long trip, the hybrid mode will kick in at around 30% SOC, now if SOH was wrong and trip start with SOC around 100% ...
Then in automatic mode the car will run in EV mode until 30% SOC, which would equate to around 3.82v ... but since SOH is "fake" and too big .. then the hybrid mode will kick in keeping the battery below the ideal voltage ... and in a long trip it can equate to tons of charge and discharge outside the ideal voltage.

 

[richr]

After charge has started and 12V disconnected again, the green park lock light next to the shifter stays on. Car is locked.
Is that normal?
I never charged at night perhaps I didn't notice.

Yep - totally normal. The green Park LED and the red plug symbol on the instrument panel light up whenever it's plugged in and charging. I think it's because it locks the transmission in Park so it can't be moved whilst charging (so you don't pull the cable out!).

 

[MadTechNutter]

I did try on my 2013 PHEV but it did not work for me

I did read from success "stories" that 12V 3 to 5 seconds reconnect for start the charging process .. in theory it should reconnect only until hearing the first click (in my case I let 3 clicks before disconnect the 12v)

I will try again ... next weekend.

No change for me either, ...well not quite, more about that below

I want be as detailed and specific about this as possible so we can get to the ground of all this, ... hope I don't bore you guys,
... aah come on, geeky details are cool

My 2014 PHEV has around 62000km, which I consider low for the age and the reason why I bought this one a short time ago.
Assuming average 35Km per cycle(I already did better in the first week I owned it) that would be about 1900 cycles if it was only pure electric but I think it is realistic to reduce 20% for petrol use.
So 1500 cycles sounds about right.

According to some website "data sheet" claims up to 5500 cycles for the LEV40 before 80% SoH.
I find that questionable.
In the 25 years(on and off) experience with Li_ion batteries at work and hobby and I have NEVER found a battery that survives 5000 cycles from 0%-100% DoD and still has 80% SoH, ... arrrgh forget it, never survived such cycle count in any way at all !

I can not find that confirmed in the Yuasa technical reports either, actually the LEV40 is so old it is not on their archive, which starts in 2006.
However I found interesting info for the LEV50 from them here:
https://www.gs-yuasa.com/en/technic/vol9/pdf/009_01_026.pdf

the retained discharge capacity is 90% after 700 cycles under the condition of D.O.D. 100% at 45 ℃．

Assuming the LEV40 behaves similar, my battery is now 'finished' because it would be below 80% and it is:

Before AND after the reset 29Ah 76% SoH :idea: sounds about right...
...Well not quite :ugeek:

It was never used under the 100% DoD condition.
At 3.83V the cycle was terminated.
And this is where this whole SoC prediction becomes rubbish. How does the car know many km I can drive until I reach a better use of the battery at 3.65V for example?
This is totally load dependent but what ever it is calculating it is doing something wrong because we have people who are still driving with 180,000km yet my 60,000km battery is showing near dead.

So I can assume that either the former owner had the battery always connected overnight on the charger keeping my poor battery at the 4.1V nightmare and it really has degraded down to 76% or the "reset" revealed a secret:

My range went up to 45km!

All the driver data had been deleted, trip odos are zero and the alarm went haywire when I plugged the battery back in.
I pulled the 12V precisely when I heard the charger click, didn't wait for other clicks.
5 hours dead before charger activated, left it for 15hours after, so definitely longer the 8hours without 12V afterwards.

It was always showing 36km before this procedure and I only managed over 40km driving like a granny, never using brakes, only paddles with a lot of distance from the car in front and a lot of "B0 flow driving" as I call it, tip lightly on the throttle a few seconds until a bit over the speed limit and coast until a bit under, I get a duty cycle of about 10% on perfectly flat road.

Don't forget when you try to recover energy you also lose some of it!
The second law of thermodynamics can be a bitch :mrgreen: