Lindqvist method : DIY BMU Battery Reset

[Pomst]

Guys ...

Let's go back to the main topic ...

I'm still trying to reproduce this Lindqvist method on my PHEV

Even trying right now .. this time I will have the 12v battery disconnected for over 24h before start the charging process

Not sure it will make any difference ... but I don't have to use my PHEV today

It will be nice to know how much success we can have in this procedure .. based on a similar topic in FB, the success rate is around 30/40%

More then the success rate, it will be nice to understand which is the critical part of the process that cause this magic reset ... and for this the experience of who did initially fail and then made it working it is very important.

I am going to try the method tonight, but a variant of it.
I will try to short out the capacitors with a resistor between the Negative and Position terminal of the battery connectors after the 12V battery is detached.

In theory, if there are no diods, this should drain the potential capacitors and clear any memory much quicker then just waiting for self-discharge to take place.

 

[MadTechNutter]

The PHEV has now LIM40 not LEV40 batteries, OK I will leave that fantasy beside, believe what you want.
I promise I will not respond further to all that.

- I have already tried shorting on the jumpstart terminal to no avail. There is no rest voltage measurable.
There are not just diodes but DC converters in between because the BMU logic/MCU will be most likely powered with 5V or 3.3V and any SRAM backup capacitor is always protected against discharge with a diode otherwise it could not work.
There could be versions that have supercapacitors that could last months and others with standard electro caps that hold the charge for a few hours.
Then again this whole clear memory with power loss could be rubbish. Somebody needs to take the BMU apart and reverse engineer it .... hmmm :ugeek:

- It would be interesting if there is a simple way to find out the software version of our cars through the OBD-II port.

- if you got it working even if you had all the latest service done by Mitsubishi

- Maybe those who got it working could post an image of their charger.

- and for me if anybody got it working with the Australian model or any model that does not have CHAdeMO.

 

[Pomst]

The PHEV has now LIM40 not LEV40 batteries, OK I will leave that fantasy beside, believe what you want.
I promise I will not respond further to all that.

- I have already tried shorting on the jumpstart terminal to no avail. There is no rest voltage measurable.
There are not just diodes but DC converters in between because the BMU logic/MCU will be most likely powered with 5V or 3.3V and any SRAM backup capacitor is always protected against discharge with a diode otherwise it could not work.
There could be versions that have supercapacitors that could last months and others with standard electro caps that hold the charge for a few hours.
Then again this whole clear memory with power loss could be rubbish. Somebody needs to take the BMU apart and reverse engineer it .... hmmm :ugeek:

- It would be interesting if there is a simple way to find out the software version of our cars through the OBD-II port.

- if you got it working even if you had all the latest service done by Mitsubishi

- Maybe those who got it working could post an image of their charger.

- and for me if anybody got it working with the Australian model or any model that does not have CHAdeMO.

So apparently you know that they are running an isolated DC/DC converter.
How did you get this knowledge?

 

[elm70]

I'm glad to post ... that I made it :mrgreen: :mrgreen: :mrgreen:

Ok ... I believe the trick is:
- 12v battery disconnected
- Connect the charger
- ~3 seconds reconnect of 12v battery

If lucky ... watchdogapps will show 40Ah ... immediately ... no need to wait

Battery does not need to be empty .. only charging process need to be started

I have to say .. I made a lot of mess:
- I left 12v battey disconnect for over 24h ... but I don't think this is relevant
- I connect the charger
- I connect the 12v battery for to short ... and charging process did not start
- I tryed again ... charging started .... but no increase in the SOH from the dog.
- Then I stop and repeated few times
- At some point I did click the stop button on the charger ... for see the effect ..
- I also believe I did remove the 220v power from the external charger too (my 220v is controlled by a WiFi switch)
- Anyhow ... after few attempts .. bingo ... no error .. the car LCD was showing that trunk was open .. and the dog reported 40Ah or 105%

I think it is all about timing .. or to maybe interrupt the charging process while 12v battery is disconnected

Anyhow ... like said by sommebody else before in FB ... I think the trick is the charging process from 12v disconnected .. and SOH should be seen immediatly updated on the Dog

Tomorrow I have a 52km trip ... I can monitor how my PHEV which was 32.4Ah will handle the "long" trip, which after a reset I believe I can do fully in EV mode

:ugeek:

 

[MadTechNutter]

I'm glad to post ... that I made it :mrgreen: :mrgreen: :mrgreen:

Ok ... I believe the trick is:
- 12v battery disconnected
- Connect the charger
- ~3 seconds reconnect of 12v battery

If lucky ... watchdogapps will show 40Ah ... immediately ... no need to wait

Battery does not need to be empty .. only charging process need to be started

I have to say .. I made a lot of mess:
- I left 12v battey disconnect for over 24h ... but I don't think this is relevant
- I connect the charger
- I connect the 12v battery for to short ... and charging process did not start
- I tryed again ... charging started .... but no increase in the SOH from the dog.
- Then I stop and repeated few times
- At some point I did click the stop button on the charger ... for see the effect ..
- I also believe I did remove the 220v power from the external charger too (my 220v is controlled by a WiFi switch)
- Anyhow ... after few attempts .. bingo ... no error .. the car LCD was showing that trunk was open .. and the dog reported 40Ah or 105%

I think it is all about timing .. or to maybe interrupt the charging process while 12v battery is disconnected

Anyhow ... like said by sommebody else before in FB ... I think the trick is the charging process from 12v disconnected .. and SOH should be seen immediatly updated on the Dog

Tomorrow I have a 52km trip ... I can monitor how my PHEV which was 32.4Ah will handle the "long" trip, which after a reset I believe I can do fully in EV mode

:ugeek:

Congratulations elm70

That you can see a successful reset immediately on the Dog is the most important information on this matter to date, well done.

However I tried now probably 100 times but had no success.
Always listening carefully to the relay clicks and disconnecting at various stages but my charger will not go on until all the relays are on and that is after 4 seconds.
I also played with the charger, terminating charge by switching off the 240V mains, rather than pulling the j1772 plug. I found that I have to switch off the charger for at least 5 seconds otherwise charge will continue. Trying to find a sweet spot there didn't work either.

I strongly assume that the time it takes for my charger to respond is the critical issue.

Could you please post an image of your charger or write at least if it looks different to mine:
Thanks

 

[elm70]

I have the same charger

After the charge and some hours rest, SOH did change down to 37.9, I was hoping for 38Ah

No error at all on my PHEV

Battery is now seen as 60days old, with only 16Ah acl/dc charging history

 

[RaZe73]

I'm glad to post ... that I made it :mrgreen: :mrgreen: :mrgreen:

Ok ... I believe the trick is:
- 12v battery disconnected
- Connect the charger
- ~3 seconds reconnect of 12v battery

If lucky ... watchdogapps will show 40Ah ... immediately ... no need to wait

Battery does not need to be empty .. only charging process need to be started

I have to say .. I made a lot of mess:
- I left 12v battey disconnect for over 24h ... but I don't think this is relevant
- I connect the charger
- I connect the 12v battery for to short ... and charging process did not start
- I tryed again ... charging started .... but no increase in the SOH from the dog.
- Then I stop and repeated few times
- At some point I did click the stop button on the charger ... for see the effect ..
- I also believe I did remove the 220v power from the external charger too (my 220v is controlled by a WiFi switch)
- Anyhow ... after few attempts .. bingo ... no error .. the car LCD was showing that trunk was open .. and the dog reported 40Ah or 105%

I think it is all about timing .. or to maybe interrupt the charging process while 12v battery is disconnected

Anyhow ... like said by sommebody else before in FB ... I think the trick is the charging process from 12v disconnected .. and SOH should be seen immediatly updated on the Dog

Tomorrow I have a 52km trip ... I can monitor how my PHEV which was 32.4Ah will handle the "long" trip, which after a reset I believe I can do fully in EV mode

:ugeek:

Congratulations elm70

That you can see a successful reset immediately on the Dog is the most important information on this matter to date, well done.

However I tried now probably 100 times but had no success.
Always listening carefully to the relay clicks and disconnecting at various stages but my charger will not go on until all the relays are on and that is after 4 seconds.
I also played with the charger, terminating charge by switching off the 240V mains, rather than pulling the j1772 plug. I found that I have to switch off the charger for at least 5 seconds otherwise charge will continue. Trying to find a sweet spot there didn't work either.

I strongly assume that the time it takes for my charger to respond is the critical issue.

Could you please post an image of your charger or write at least if it looks different to mine:
Thanks

Hi

I'm fom Finland and this is what my charger look like (Last picture) and it worked at the second attempt. First time I did not react fast enough when I heard the EVSE (Original 2015) relay click, but second try I removed 12V almost simultaneously as the click, since I remembered the time in my head. I did second attempt directly after without waiting time between only reconnected EVSE.
I had the 12V disconnected for 12,5h over night.

I have the boot open and locked it with a big nail between the look, also I have WIFI and 2015 model PHEV 62000Km SOH 85,1% before, and I DONT CARE ABOUT WARRENTY since that's never gonna happen here :lol:

A EV service needed is now in the display, which I don't care about since I heard several people ac-countered this.
Will follow the charge on the dog during the day. Range went straight from 0%->11% on the dog and on the SOC from around 9Ah to 15.1Ah.

Another funny thing was when I reconnected the Dog the second time after success it did not show 40Ah anymorem, but 12Kw.

Cell Voltage at @ 50% (20Ah) SOC = 3,943V Average, Diff 0,003V.

 

[MadTechNutter]

Thank you elm70 and RaZe73.
This rules out the charger then, which simplifies the search for a solution.

The only difference now is that you both have a ChaDeMo port and I don't.
I can't believe they would have designed a different BMU just for us handful of Aussies but it might be time to get my hands dirty and try to access that thing and get it on my bench.

 

[michael8554]

Recently acquired a UK 2014 PHEV 20K Miles

Have just bought an OBDII dongle and loaded PHEV Watchdog, currently (ha !) shows 30.9Ah at 81.3%.

So I'm interested in this reset.

1) I get the impression from a couple of the recent posts that the Dog will give readings whilst charging ?

eg elm70 on Tue Jul 30, 2019 9:49 pm

"watchdogapps will show 40Ah ... immediately ... no need to wait"

I've only got Watchdog to connect when the car is powered and Ready - so is it safe to power up whilst charging ?

2) The instruction to charge for 8 hours ?

Mine charges in about 2.5 hours with a 16Amp EVSE, do you still leave it connected for 8hours?

Or should I use the 10Amp Granny charger (same as elm70's), again leave on for 8 hours ?

3) The 16Amp charger gives a loud Clunk when charging starts, counting 1-2-3 and disconnecting 12V before the Clunk results in charging not starting.

So should I just anticipate the Clunk and ignore the counting and noises from the car?

Loving the car and this forum.

I have 3.2KW of Solar EV panels and Economy 10 Electricity Tariff, which gives 2 hours of cheap electricity in the afternoon.

So almost free charging at home :->

Michael

 

[jaapv]

Now if this method works and allows the car to use a larger part of the battery capacity, the lifespan of the battery will surely be shortened. For somebody like me, who plans to drive the car for a long time, it will be wiser to accept Mitsubishi's programming, which appears to be aimed at longevity.

 

[Mitch]

I'm absolut with jaapv.

We will see at lot of people crying with dying batteries.
This reset does nothing except dropping an error code and you will lose you "safety buffer"!!

The only working and sinful solution, if needed, is the DBCAM, which is done in 5 minutes, rest is done by the car alone (just discharge and then fully charge).

 

[mellwaters100]

I have been pondering this 'fix' for a few days now and I still come back to the same questions.

1) What mechanism tells the charger to stop charging?
2) How can the actual capacity of a battery be actually known?

So, I have done some digging through academic papers (as a mature student, I can access journals through Google Scholar). This is one I have been reading today (https://www.mdpi.com/1996-1073/4/11/1840) [can't seem to attach the file, but it is accessible under free distribution licence].

A couple of days ago on this thread I ventured that the BMS was programmed to reduce SOH of the battery over time within pre-determined safety margins. In other words, through testing Mitsubishi determined that an average battery will degrade by 0.1Ah every 6,000kms (for example). To account for variations in environmental conditions, driving styles, charging practices, etc, they built a cushion into this and decreased the SOH regularly through software in the BMS. Net result is that, as the 'estimated' SOH reduces, less charge is requested from the charger.

It seems that, as per the attached paper, manufacturers don't have a reliable way of measuring actual State of Charge, Health or Life of a battery and therefore implement systems to manage the various trade-offs in EV's and Hybrid EV's.

Through all of this, I keep coming back to a basic idea that it is not possible to put more charge into a battery than it is able to hold (in the same way you can't put 5 gallons of water into a 4 gallon bucket!!). So, back to my original questions:

1) What mechanism tells the charger to stop charging? I reckon that this procedure wipes current SOH capacity but retains the manufactured SOH capacity of 42Ah. It is this setting that is still telling the charger to stop sending current once that threshold is reached. However, once the car has gone through a drive cycle, the BMS is settling itself down to the 40Ah level, retaining a top end safety cushion of 2Ah (note, in Tesla Bjorn's videos on YouTube he talks about this 'hidden' capacity in new Tesla batteries).

2) How can the actual capacity of a battery be known? I don't think, with the current BMS's, they can be. But, through this process, I suspect that the charger is being stopped once it is detected the battery cannot take any more charge. I think this is why some people are not seeing an improvement with this procedure, their batteries have naturally degraded to that point. But similarly, it could be that those batteries that initially report 40Ah / 42Ah and then quickly return back to their previous settings have degraded as well and, through subsequent charge cycles the BMS is detecting unstable cells and reducing capacity accordingly? Similarly, it could also be that Mitsubishi have upgraded the BMS to fix this back door in earlier implementations.

To Mitch's previous comment - I suspect that running this procedure on every charge would have negative results quite quickly, however, as an occasional procedure, I don't see how it can (coming back to my suggestion that you can't put more charge into battery than it can take). Yes, continually overcharging a battery kills the cells meaning it can't hold charge once you unplug it, thus suggestion that repeatedly doing this could be an issue.

Anyway, just some thoughts to through into the mix! Just thought it was interesting that research shows that BMS's are not 100% truth tellers!

 

[MadTechNutter]

I can't believe they would have designed a different BMU just for us handful of Aussies but it might be time to get my hands dirty and try to access that thing and get it on my bench.

Didn't need to get dirty hands. Lifting off the drink holder in the back next to the charge port revealed a little black box with the part number 9499D054.
Plenty on Ebay, used for $100-200 from the UK.
So far it seems at least the 2014-2017 UK models have the same battery control module as mine.

Anyhow ordering a spare to mess around with is cheaper than this so called "smoothing procedure" that costs about $300 where I live and does not stay resident either.

 

[MadTechNutter]

Recently acquired a UK 2014 PHEV 20K Miles
Have just bought an OBDII dongle and loaded PHEV Watchdog, currently (ha !) shows 30.9Ah at 81.3%.

Well either that is a faulty battery,(you might see weak/unbalanced cells in the Dog) or confirmation that the information from battery management system, including the voltages that are reported through the Dog, is utter B.S.


The OBD- dongle is powered up once the charge process has started, so you can see after the App has synched the result of you activities.

The charger both members above who had success with is the 9.6A one.

 

[Phevy]

I have been pondering this 'fix' for a few days now and I still come back to the same questions.

1) What mechanism tells the charger to stop charging?
2) How can the actual capacity of a battery be actually known?

The charge is stopped when the cells reach 4.1V.
The BMU does not mess with the top level charge.
The 'capacity' determines the bottom level of voltage, when the BMU deemed the battery 'empty' and request the ISE to start the generator.
The difficulty with determining that bottom level is hard to do as the voltage varies with the load. Thus the usage in Amps is measured in stead. For this a calculation needs to be performed to determine how many Amps can be drawn, before considering the battery empty, and this is where the BMU comes in and does this calculation; factoring the age and the cycles of use.

To know the actual capacity, you need to 'fully deplete' the battery and then measure how many Amps are loaded till full charge. This is what the DBCAM procedure is doing.

I may be wrong, but that is what I've learned thus far.
Phevy

 

[elm70]

I have been pondering this 'fix' for a few days now and I still come back to the same questions.

1) What mechanism tells the charger to stop charging?
2) How can the actual capacity of a battery be actually known?

The charge is stopped when the cells reach 4.1V.
The BMU does not mess with the top level charge.
The 'capacity' determines the bottom level of voltage, when the BMU deemed the battery 'empty' and request the ISE to start the generator.
The difficulty with determining that bottom level is hard to do as the voltage varies with the load. Thus the usage in Amps is measured in stead. For this a calculation needs to be performed to determine how many Amps can be drawn, before considering the battery empty, and this is where the BMU comes in and does this calculation; factoring the age and the cycles of use.

To know the actual capacity, you need to 'fully deplete' the battery and then measure how many Amps are loaded till full charge. This is what the DBCAM procedure is doing.

I may be wrong, but that is what I've learned thus far.
Phevy

Actually you are partially wrong.

Yes, in order to measure the battery capacity it is needed to bring down the weakest cell of the pack down to 2.75v, something that even DBCAM procedure is not doing, and the result is that batter capacity is still a guess from the BMU

For properly handle the SOC, it would be needed to keep up to date the mapping between voltage (at rest) and SOC, something that appear to be a per-configured mapping which get adapted with battery age ... again I believe DBCAM does not help in this important aspect

On top .. it is even questionable how much is the "damage" to discharge the battery down to 30% SOC, vs going down to 20% SOC ... in my experience with Lithium cell, I only notice battery damage/degradation when battery has been bring below 3v or 2.5v .. so below 2% SOC

I believe the PHEV must have a protection that force the ICE to start immediately when any of the cell in the battery go below 3.0v ... at least I hope this is the case ... else they would have done a quite questionable firmware in my view ... soon ... thanks to many people that did rest the battery, we will know how low the voltage at rest and at load is allowed to go down.

 

[elm70]

I can't believe they would have designed a different BMU just for us handful of Aussies but it might be time to get my hands dirty and try to access that thing and get it on my bench.

Didn't need to get dirty hands. Lifting off the drink holder in the back next to the charge port revealed a little black box with the part number 9499D054.
Plenty on Ebay, used for $100-200 from the UK.
So far it seems at least the 2014-2017 UK models have the same battery control module as mine.

Anyhow ordering a spare to mess around with is cheaper than this so called "smoothing procedure" that costs about $300 where I live and does not stay resident either.

If Lindqvist method does not work for you ... I think the best way to reset the battery is with: http://www.myoutlanderphev.com/forum/viewtopic.php?f=10&t=4533

 

[elm70]

Recently acquired a UK 2014 PHEV 20K Miles

Have just bought an OBDII dongle and loaded PHEV Watchdog, currently (ha !) shows 30.9Ah at 81.3%.

So I'm interested in this reset.

1) I get the impression from a couple of the recent posts that the Dog will give readings whilst charging ?

eg elm70 on Tue Jul 30, 2019 9:49 pm

"watchdogapps will show 40Ah ... immediately ... no need to wait"

I've only got Watchdog to connect when the car is powered and Ready - so is it safe to power up whilst charging ?

2) The instruction to charge for 8 hours ?

Mine charges in about 2.5 hours with a 16Amp EVSE, do you still leave it connected for 8hours?

Or should I use the 10Amp Granny charger (same as elm70's), again leave on for 8 hours ?

3) The 16Amp charger gives a loud Clunk when charging starts, counting 1-2-3 and disconnecting 12V before the Clunk results in charging not starting.

So should I just anticipate the Clunk and ignore the counting and noises from the car?

Loving the car and this forum.

I have 3.2KW of Solar EV panels and Economy 10 Electricity Tariff, which gives 2 hours of cheap electricity in the afternoon.

So almost free charging at home :->

Michael

1) OBD2 port is available every time the car is charging, or when remote climate control is active ... no need to have the car in ready mode for use Watchdog App
2) No need to wait until the car is fully charged, reset procedure can be verified few seconds after the charging process has been started. No need to have the car multiple hours without 12v once the charging process is finish.
3) It is looking like the reset happen while 12v power line is restored and quickly removed with a charger connect to the car. It might work with different charger, but I believe you have more chance using the standard charger included with the car

 

[elm70]

Here is my experience post Lindqvist method

My PHEV is a 2013 with 143.000km ... got used 3 years ago when was having already 120.000km .. I'm mainly using this car for daily commute, which did decline recently since I have more home office in these days

My 1st SOH reading back in Feb 2017 was 36Ah or 94.7% .. previous owner was mainly using the car in Hybrid mode, plus he did damage his charger (that was left exposed to elements in NL)

From Feb 2017, to July 2019, and only ~20.000km, SOH drop massively down to 32.4AH or 85.3%

After reset ... I got back to 37.9Ah or 99.7%

Interesting I have two card from the dog at 26.1% SOC, both car got after a car left to rest

Most recent one:
26.1% / 9.9Ah / 302.8V - 3.785 cell voltage
1 month ago:
26.1% / 8.6Ah / 304.6V - 3.808 cell voltage

There is over 0.02v difference for the same SOC ... that means the BMU has a "special" punishing logic for battery which look old per the configuration.

Anyhow ... just did a "motorway" round trip , with almost 2h rest in the middle .. 24+24km .. which I never managed to make in full EV mode in the past .. my best was ICE kick in 3km before end of 2nd lag

Anyhow ...
1st section
Battery from 37.6Ah to 25.3Ah
Voltage min/max 307.2/327.3V
Cell min/max 3.829/4.093
Max delta cell voltage 0.021

2nd section
Battery from 23.9Ah to 11.9Ah
Voltage min/max 293.4/316.6V
Cell min/max 3.657/3.963
Max delta cell voltage 0.029

After 2h rest ... SOC did lost 2Ah ... 7h later ... the lost was only 1.8Ah ... next day ... even less 1.7Ah ... yes bouncing back to proper voltage take many hours on my old battery

So ... SOC lost between the 2 trip was 1.4Ah + 1.7Ah lost after the 2nd leg

So total SOC lost was 3.1Ah .. which in my book it means my real SOH in my PHEV is 37.9-3.1=34.8Ah ... but for my BMU before the rest was only 32.4Ah with 1Ah lost in the last month alone (so huge downhill SOH trajectory)

Did I damage my car in my trip which would have need fuel without a reset ?

I don't think so .. since I did not gone extreme ... since trip was in 2 legs, 2nd legs SOC was already reduced .. and 2nd leg end with still 2km of EV range

Could I potentially over discharge my car ?

Yes ... but only in extreme situations
Assuming I will make a long trip from 100% SOC .. I am allowed to consume 2.1Ah more then intended (3.1Ah is current over estimated SOH, time 70%) ... this means I can bring down my SOC to real : 9.3Ah or 24.5% ... which is not far from the 25% SOC which is the second limit for kick in the ICE when driving at low speed (something that I hardly manage to use) .. so for motorway or long trip ... no issue at all with current "error"

Anyhow
... before the BMU was naturally wrong ... with -2.4Ah error
... now the BMU is "reset" wrong ... with +3.1Ah error

The 3.1Ah error will be quickly corrected by the BMU ... I expect to have in few weeks -1Ah in SOH .. and possibly in 6 months / 1 year back to the real SOH

So ... when the SOH will go below 35Ah .. I think I will do another reset

 

[MadTechNutter]

I can't believe they would have designed a different BMU just for us handful of Aussies but it might be time to get my hands dirty and try to access that thing and get it on my bench.

Didn't need to get dirty hands. Lifting off the drink holder in the back next to the charge port revealed a little black box with the part number 9499D054.
Plenty on Ebay, used for $100-200 from the UK.
So far it seems at least the 2014-2017 UK models have the same battery control module as mine.

Anyhow ordering a spare to mess around with is cheaper than this so called "smoothing procedure" that costs about $300 where I live and does not stay resident either.

If Lindqvist method does not work for you ... I think the best way to reset the battery is with: http://www.myoutlanderphev.com/forum/viewtopic.php?f=10&t=4533

Thanks for reminding me
...and that detailed experience post.

It is just that the Lyndqvist method seems the least invasive way and I would like to see if it would work with a replacement BCM that might have older firmware in it or something.
+ I can keep things as they are if I don't touch the original BCM in case something goes wrong when I do try that DIY triple procedure.

It has alway been handy to have a back up and 70UK pounds for the used BCM won't break the bank with the current exchange rate, thanks Boris Johnson