Well done Andy and Gerard. Your efforts Andy in producing these videos has greatly helped all Outlander PHEV owners. Gerard's work in understanding the code is brilliant. Mitsubishi should be paying both of you as consultants. The least they can do is give you a new battery. They should also integrate the current into the battery to determine number of Coulombs (a Coulomb is a ampere-second) going into fully charge the battery. The number of Coulombs to fully charge the battery will tell you how many Lithium ions in the battery are being converted. As the battery deteriorates less Lithium ions will be available as they become locked up in the battery. This is why the Ah value is the most important reading in determining battery performance.
Bloody brilliant well done ' team' Australia, This will shape the future of EV management. Not forgetting the input from ' team' Portugal. A worldwide influence on a mega corporation. Keep making the videos Andy, never surrender!
Hi Gerard, I was talking to Andy about my situation and he advised me to talk to you. I have one question for you: How did you managed to read the BMU code? And can I look at it as well? :) Thanks
Most probably using the MUT III. It's around 450 USD on ebay. My question is rather this : what is the equation governing the SOH calculation? Surely it cannot be y=ax+b...
Great work Andy and Gerad. Still more to do with MMC, but loving the new PHEV you convinced me to buy (😉), so glad it’s not a 🍋 🤣 However, as this issue isn’t a safety issue, I’m not surprised MMC don’t roll out a recall. For those who suffer and ‘just drive’ it costs MMC nothing, for those that see 70% ‘assumed’ degradation, this procedure may save millions in battery replacement costs. For the likes of us tinkerers, it will shut us up... hopefully! One question is still bugging me, though: What collaboration has Yuasa, the BMU manufacturer and MMC had to set this system up? If none, maybe the LEV-40 is a better product than the degradation algorithm gives it credit? Then this procedure should give many of us a great solution. Or maybe MMC didn’t realise how their cars were going to be driven and thought this would never be an issue... 🤔
Episode 190 - Should be called the REVELATION. Huge information. Wonder what our batteries really are at? Can't wait for my next service. Should be front page news. Well done.
Andy, thank you for your videos and for the whole team for your efforts! When I bought my PHEV I didn’t anticipate the enthusiasm involved in this technology. However, I am completely hooked with the watchdog and your videos. You should sell your show to Discovery channel!
This is superb news. Gerard and you have done a cracking job. We seem to have at least a work around for now. I'm amazed that it is still Version 1 of the BMU software. I have a service in November. I really hope MMC gets new procedures out to dealers by then.
Thank you so much for all your efforts, I tried to talk to my dealership when I brought the vehicle to get them to do a cell smoothing etc, but they didn't even know what that was. Now I have your videos to back up what I am saying. I have emailed my dealership asking to have this procedure done.
My head hurts a bit! Great findings and thanks to all Australian phev team. But you are the lucky ones! Because in other country's in Europe it is impossible to take the problem to the local Mitsubishi delars! And that's for those cars that are still under warranty. In my country they will never agree that they have a problem with the phev. In other words this year I should make the move to a second-hand PHEV, but which one. Still, I remember that whit the old 🔋 you had a range of over 30km, this is my daily commute, and will should take care of the problem. But the bad software is still there!
That is excellent, the more people know and complain with their dealers the better. Public pressure is the only weapon we have to get this finally fixed.
So 40Ah for 2018 and previous models and 46Ah for the new one. right? We all owners all over the world must thank to Gerard and your team for this precious information and the posibility that MMC solves it through a future recall at the MM workshops.
I suspect that the bmu having the wrong initial input from the beginning continously "think" that the battery degrades much quicker than it should and at every cycle of evaluation tries to "save" it, thus continuous drop in the apparent capacity. This way user should not be able to degrade the battery further but the loop continues in some kind of a vicious cycle and will never stop unless everything is reset and a proper initial value is inserted. The biggest bug in this system is that the BMU does not correct the predictions using, for example, max nominal V of the battery and uniformity of V across the cells.
Very interesting video thanks Andy. It is interesting to see that scientific proof has largely confirmed what I said in the earlier part of last year, and that is that most of these so-called deteriorating battery problems were in fact the BMU artificially strangling the amount of charge admitted into the battery using arbitrary parameters. I was wrong in assuming that this was to prolong the life of the battery, not that it would not have occurred as a result but it is much more the result of incompetence and laziness on the part of Mitsubishi motors. It sounds like you are suggesting that the three procedures should be done to older cars. Mine has just gone over 30,000 km, so possibly the real battery degradation is not all that great.
We have to test the procedures in older cars and see if this makes a difference or not. One viewer has confirmed here it does not make a difference though...
Doing the math on mine only the battery cycle degradation seems to line up roughly. 1969days old, 98600Ah of usage, 85.8% SoH. Thanks for digging into this. I'm very curious on how this all turns out. Almost makes the intro song more manageable.
Not sure if I totally understand what you are saying. We know the battery is a 40Ahr battery, but it is only ever charged to 4.1V. To actually get 40Ahr into the battery you need to charge to 4.2V until current reduces to about 0.05C. So I would have thought a full battery for the PHEV is about 38AHr, because Mitsubishi wants to keep the voltage at 4.1V or less for increased life.
Yes, that is correct. the actual cells charge only to 4.1V regardless what the capacity of the battery is. 4.1V is around 85-90% state of charge of a cell, so more like 36Ah (40Ah-10%). The Dog always showed 38Ah after a reset or even in new cars so assumed this is the max capacity it has. The Dog should show 40Ah. if it shows 38Ah it is not aware of the true capacity and makes false assumptions.
Now it is getting interesting. Could "we" reverse engeneer a working BMU solution from another manufacturer to reflash our cars? We would be the first EV modding crew..
what will it give you? The satisfaction that your SOH number in WatchDog is more accurate? If the battery is discharging to 3.7V and charging 4.1V then you have the full capacity available for your driving - regardless of what the BMU is telling you
On the UK PHEV group FAQ, they state: "Over the next 1,000 miles, your SoH and guess-o-meter range should relearn and hopefully start to increase, all else being equal. It’ll help if you also follow the tips in FAQ#M." It will be very interesting information if someone following the procedure on "A. Fix bad battery health data (without dealer)?" could confirm seeing any increase of SoH. According to the last discoveries, this would be impossible.
Thanks for the info Andy. I'm not sure about those percentages though as it would make my 62 month old MY14 with 49068km on the clock be expecting a SOH of 73.48% when I've actually got 85.3% showing in the Dog. Perhaps the percentage degradation calculation is not exactly 0.3% and 1% ? I have wondered about this - the BMU is old and presumably slow by today's standards. Does it do proper floating point calculations, or is it using some low-precision 8 or 16 bit integer math? Perhaps the calculations are simplistic and give different results compared to us using our scientific calculators and spreadsheets?
It looks like someone in the design office of Mitsubishi has stuffed up and did not follow through there design to work correctly, A back dated trade description would be nice for all the cars to be updated with the new software. I would be willing to pay a small fee for the update. I would have thought at the end it would save Mit- a lot of money in replacing batteries on car up to the 8 years old as this date has not arrived yet as no car is of that age yet. I would not think that MMC would endorse a no action policy in the light of bad press against the car and will have an adverse effect on sales and customer satisfaction.
Mitsubishi was not afraid of bad press when I started the channel. I don't think they are taking this serious enough. The PHEV sales may be only a very small number in comparison to their overall sales. We are talking to the Head of Customer Experience all the time and still did not get the attention we should. Not even if you have a camera as weapon...
My deep respect to you and Gerard ! Let's see if Mitsubishi creates a service procedure for this. My PHEV dropped 2Ah and 5% SOH in 5000 km, so i'm looking forward to such a simple fix. www.phevblog.nl/en My usage = 98640Ah, so would be 15.9% capacity drop And age = 1459 days would be 14.6% Actual SOH = 90.8% (34.5Ah) This does not match the calculation in the video, however the first 125.000 km the car was not charged by the previous owner. I got it with 95.8% SOH at 125000km. When i only use the delta after i got it, charging twice per day (7356Ah and 113 days) the capacity drop using the formula is 2.32%.
Andy, Not so sure about what you are saying. As a matter of fact, Mitsubishi themselves have explained in one of their service bulletins that the BMU can only calibrate itself when it is regularly charged over A/C and even then, only "when certain conditions are met". Are they bluntly lying? For cars that are never externally charged or only via DC chargers, Mitsubishi themselves indeed say that the BMU has noting to go by, other than the preprogrammed deterioration params (we didn't know the actual params, so thanks for that). But we have also seen cases (on the same PHEVWatchDog,net you refer to) of cars where reported SoH spontaneously went up in several small steps when charging pattern changed. If you and Gerard were right about the software, this would be impossible. BTW My battery is 65+ months, so I should be down roughly 20% for this. My Ah in/out is 170.000 Ah, so I should be down another 27% down for this. In total, I should be down approx. 47%. But I am 'only' 30% down. Less than 2/3rd of what to expect based upon preprogrammed params. Yes, I have had several DBCAMs performed, but never a full reset. So, according to you, the DBCAMs should not have made a lot of difference. Also, in 8 years, every battery should be down by at least 8 * 12 * 0.3 = 30%. Subtract some more for usage and about every battery would eventually be eligible for warranty replacement (provided the owner did not more than 100.000 miles).
@@unpluggedEV So the people claiming to see their SoH value going up by regularly charing the car, such as Richi Jennings, are lying? And so are the stats on PHEVWatchDog.net that show SoH on some cars going up, step by step? With all due respect, your explanation was not all that convincing to me (or to yourself, as you did not want to be quoted on your claims ;-) ) and the 'evidence' appears to be restricted to TeamPHEV members only. Tough to sell, if you ask me.
Hi Andy: Obviously your work and that of Gerard plus many others including Gary from North Bay . Canada is so very much appreciated. Amazing effort. What begun as a new car purchase for myself and my wife has turned into a hobby AND possibly a new religion!!! BRAVO to all of you. I have forwarded your episodes # 183 + # 190 to my local Mitsu dealer her in Toronto , Canada.. He is very appreciative and understands all. He in turn has forwarded the links onto his regional Mitsu tech. representative. My dealer is very familiar with the first 2 steps of your "inoculation". He has no knowledege or understanding of the 3rd step..." D.B. CAM". Have we understood this properly??? Can you further describe where to find more about this " D.B. CAM" ?? Thanks so very much. Ron
if Gerard already went through the pain of reverse engineering the firmware, why not modding it right away so that owners or Mitsubishi shops could reflash/reprogram the BMU themselves?
Because it seems like it lacks the necessary sensors to actually work - well perhaps not lack of - but in any case the sensors don't report to the firmware. You could re-flash the BMU, but you still wouldn't know the real capacity because Mitsubishi made capacity a simple calculation rather than real world. The cell smoothing procedure which is basically a cell balance and then DB Cam (whatever the hell that is?) seem to then be able to populate the registers with the correct values (provided you reset the BMU first and clear out all the existing data.) But after that the BMU will do what it has always done and start degrading the battery based on it's inbuilt firmware.
@@mondotv4216 , well maybe that's not needed to improve the way the BMU calculates the state of the battery (although it probably would be ideal if the BMU could access and use data from the real health state of the battery). It may be worthwhile to check if it is possible to overwrite (change) both deterioration constants (charging cycles and age) in the BMU to a less aggressive value. That's not even a programmatic change, just pure configuration. This solution wouldn't fix the problem, but slow down the deterioration which in its stock version seems to be based on worst case assumptions. If programmatic changes would be feasible then it may make sense to look at how exactly the math algorithm works in the BMU. If it is some sort of a recursive function - meaning, it calculates always on the basis of the previous result then it may make sense to investigate if the algorithm could be changed in such a way that it calculates always based on the initial battery health state value provided by the calibration process. So instead (x=x-constant) it would be (x=y-constant) - the X coming from the lower register and the Y from the upper register (see video). Ideally Mitsubishi would change the firmware and introduce a solution that can read out the real state of the battery and update the upper register in the BMU accordingly.
Andrzej P if that is the case then my BMU was always correct. Which I dont think it is because I've witnessed 2% drop in 3 months before doing the procedures...
@@rubencosta3975 you see I believe that the car do not use the SOH reported by BMU for anything. It charges and discharges to true capacity based on cell voltage . Mine shows 88% on WatchDog but charges to 105%. I'm sure your will charge exactly to 100% of reported SOH since you just had DBCAM and your BMU have the correct value. But the point is that it doesn't matter what SOH BMU has - the car will drive the same. At least that's what I think.
That would be insane. I refuse to believe that just because I'm a good person and think the car is great and the batteries are great and everything will be great eventually.......... I know exactly what you're saying and I hear you for a while but I did not want to hear that!!!! Well, once my car goes through the procedures we will find out I guess, right. If the SOH does not change anymore and sticks to 40Ah the only method to measure degradation is the charging energy.
Thanks for this info! Have you got a yellow card in your Dog's Battery Condition tab? How old is your vehicle, km, do you know how much kAh you had on the Dog before the reset?
Do you recommend performing the triple procedure even before getting your new Outlander? Afterwards the system should know the real capacity, shouldn't it? By the way: Thanks for your great work!
Thanks Nicolas. The triple Procedure does not seem to work over time. It looked good at the beginning but SOH is going down again after a while in the same pace as without the procedures. It calibrates the BMU to the correct capacity though, that is correct. It's the best process we have at the moment...
Hmm this is very interesting, I’ve not looked at my figures yet, but if the degradation is purely algorithmic, I’m wondering why are the 2019 batteries on such a steep degrade path, compared to earlier models. Any thoughts? If the code is unchanged the degradation gradient should be the same?
Ian U.K. maybe if the 6200Ah of charge+discharge value has not been changed in the new ones’s BMU, that 1% down is more frequent and shows bigger degradation steps down as well
Alejandro Carrera I understand your point entirely, however taking my own car as example, the battery is 135 days old, I’ve had the car 2 months. The Dog says 41.9Ah SOH and total energy use is just 3734Ah. So, you can see that I’ve not reached the 6200Ah threshold yet, so can discard that, and worst case 4.5 months of 0.3% degradation for time. 46Ah x 4.5 x 0.3% = 0.621Ah maximum degradation. My actual degradation is 46 -41.9 = 4.1Ah ie 8.9%. Way off, so there’s more at play here I believe.
(Mark Walkington, NZ here ) Good work everyone!!! ... however if the BMU has such a direct algorithm can we pin it down precisely. I have applied the the assumed degradation factors to my 146668 Ah usage and 1899 days and this predicts even higher degradation. i.e.1- ((146668/6200)*1% + (1899/(365/12))*0.3%) = 57.6% (my vehicle reports 68.9%). Any other views on this equation?
I don't think it calculates correctly if there is no bearing to the actual capacity to start with. The reported 68.9% are from an uncalibrated BMU, right?
WTF??? :D SO I am already at 83% of SOH? :D After 13000km?? :D Now, how do you explain the 107% SOC? Did the BMU somehow learn or it's the charger that decides that Vmax is not yet reached? Ps. still no global solution because my dealer won't listen to Andy...
So for my 2019 coming in June. Do I need the Dog to see if the dealer actually did the 3 steps? And how can you see it? And of course thanks TeamPHEV for all this info 👌
@@unpluggedEV being sarcastic yes? Otherwise you guys did not know about doing the 3 steps. But I bet it is a bit stressful to see the rate go down so fast.
@@benjaminlibotte8471 Seriously, don't get the Dog. Keep watching Sturm der Liebe and that's it. Don't measure your own car. PS: You can now get a Model 3 for the same price as the Outlander... just saying...
@@unpluggedEV model3 is too small for my needs and waiting list will be very very long and it remains to be seen if it goes for 40k-ish US $ or €. And there is not a nice unplugged EV to watch about them, unless you have plans in that direction Andy 😉 on a site note... it's better to learn how to walk before start running so a PHEV is the in between step for me and who knows in 5 years cars run on water
I think it's a percentage of the available capacity. So the calculated degradation is not linear and also depends on your usage, which is probably not the same every day either.
unplugged EV if it only cares about cycles and time wouldnt that mean that it will always be the same, the only difference is the available capacity of the battery?
@@unpluggedEV My case is that the potential degradation pattern would still be the same if you look at it in a graph just that the values would be a little different. Since its always the same calculation(?) and the only value that actually change from the BMU fix is the total capacity of the battery?
Hi Andy, Gerard Unfortunately the things don't seem so be as simple as explained. The data from my car does not fit in your explanations. Currently I have SoH 83.2%=31.6Ah which means that my car is one of those with 100%=38Ah. In the same time the total current is 110 000Ah which should mean 18% degradation according to your formula. It is also 58 months old which would mean 17% degradation more. According to your degradation formula, the BMU should have calculated 17%+18%= 35% degradation. Lucky, my SoH is 83% instead of 65%. Apparently, there are other parameters that the BMU considers. Also, strangely enough, after the first 3 years and 140 000 km with the previous owner, the SoH was 92%. Now only two years later and 40 000 km more, my SoH is 82%, so the degradation started dropping faster with my usage. Any thoughts on this? How does it fit in your findings?
Recalculate like the percentage you get on leasing in a bank: apply percentages every time she SOH is changed. Do not apply projected degradation to the initial capacity.
I keep asking the same question over and over - does SOH number effects the power the battery can take or the miles the car can drive ? I'm afraid Mitsubishi don't care what SOH BMU is reporting! You keep ignoring this. I'm afraid that if you would do your DBCAM to your old battery, your range would still be the same.
@@unpluggedEV I think we are going back to the problem #1 that was discussed many times - small battery (12kWh) in a large heavy vehicle that operates for months in EV mode. Those batteries must work under very high stress and they just degrade much faster than in other BEV's . Not as fast as BMU reports, but still much faster than we would like. Perhaps Mitsubishi should turn on the ICE much faster or just increase the battery size, lower the EV power etc. Last month I got a 40kWh Nissan Leaf for my wife and we drive it now much more than Mitsubishi. The Leaf is smaller, lighter and has over 3x bigger battery. Also, we didn't have to charge it to full even once during the last month and our SOC didn't drop below 35% neither. In Mitsubishi we would have to do probably 20-25 full cycles during that time (25% through 90%) - I think this is the reason why Mitsubishi batteries suffer premature capacity loss.
@@Specwinclean There is as many Ah capacity as you can charge into the battery - no more no less. If your EVSE or WatchDog shows you that you charged to 36Ah then your battery has 36 Ah capacity available for your drive, regardless of what BMU "calculated" SOH is showing you. Also in all LiIon batteries, the SOC is proportional to voltage if your battery is charging up to 4.1V then it's charging to the full available capacity as designed. It won't suddenly charge to 4.2V if your BMU start showing you 100% (40Ah) SOH - there is only so much Ah that you can pump in the battery until it reaches 4.1V/cell.
Correct, but if the BMU says that you only have a maximum of 25.5 Ah, you won’t be able to put any more in, because it’s the BMU that says that it’s full, but you are correct I think with your summation of this type of EV.
So let me get this straight. Mitsubishi's BMU is really nothing more than a glorified counter that "guestimates" battery degradation (and very aggressively it seems)? It doesn't measure actual cell capacity? Ever? That seems like a very odd engineering decision. And it explains a lot. Like how you can get your car back from a dealer after getting the battery tested and suddenly it's range jumps up to near new only to degrade again very quickly. You would think at the very least Mitsubishi might introduce some kind of per cell amps in and amps out monitoring and compare it to actual voltage. It's not exactly rocket science. At the very least they should be able to do this for the pack as a whole.
Honestly, this all develops to a real nightmare. I just hope more and more owners jump onto the bandwagon and complain about this. Finally this should urge MMC to provide a fix. A real fix.
Another video, another bout of blood pressure issues.... It shouldn't matter how old the Fing cars are, they were screwed over by Mits and their shitty build methods. Tesla is beginning to look better and better at this rate. So, basically, I now have to convince the dealer to do a 3 day plus procedure, for free, because the manufacturer doesn't actually know how to build cars... Hello, can we say class action lawsuit... Better opening again...
Well done Andy and Gerard. Your efforts Andy in producing these videos has greatly helped all Outlander PHEV owners. Gerard's work in understanding the code is brilliant. Mitsubishi should be paying both of you as consultants. The least they can do is give you a new battery. They should also integrate the current into the battery to determine number of Coulombs (a Coulomb is a ampere-second) going into fully charge the battery. The number of Coulombs to fully charge the battery will tell you how many Lithium ions in the battery are being converted. As the battery deteriorates less Lithium ions will be available as they become locked up in the battery. This is why the Ah value is the most important reading in determining battery performance.
Bloody brilliant well done ' team' Australia, This will shape the future of EV management. Not forgetting the input from ' team' Portugal. A worldwide influence on a mega corporation. Keep making the videos Andy, never surrender!
Thanks Les. A team effort from Gerard, Daniel and many others...
Nice video Andy. Such a breakthrough.
Thanks to you!
+1
Hi Gerard, I was talking to Andy about my situation and he advised me to talk to you. I have one question for you: How did you managed to read the BMU code? And can I look at it as well? :) Thanks
I'm sorry. This info is only discussed within our private group.
Most probably using the MUT III. It's around 450 USD on ebay. My question is rather this : what is the equation governing the SOH calculation? Surely it cannot be y=ax+b...
Great work Andy and Gerad. Still more to do with MMC, but loving the new PHEV you convinced me to buy (😉), so glad it’s not a 🍋 🤣
However, as this issue isn’t a safety issue, I’m not surprised MMC don’t roll out a recall. For those who suffer and ‘just drive’ it costs MMC nothing, for those that see 70% ‘assumed’ degradation, this procedure may save millions in battery replacement costs. For the likes of us tinkerers, it will shut us up... hopefully!
One question is still bugging me, though: What collaboration has Yuasa, the BMU manufacturer and MMC had to set this system up? If none, maybe the LEV-40 is a better product than the degradation algorithm gives it credit? Then this procedure should give many of us a great solution.
Or maybe MMC didn’t realise how their cars were going to be driven and thought this would never be an issue... 🤔
Episode 190 - Should be called the REVELATION. Huge information. Wonder what our batteries really are at?
Can't wait for my next service. Should be front page news. Well done.
Thanks Barry. It get's more and more complicated unfortunately.
Andy, thank you for your videos and for the whole team for your efforts!
When I bought my PHEV I didn’t anticipate the enthusiasm involved in this technology. However, I am completely hooked with the watchdog and your videos.
You should sell your show to Discovery channel!
This is superb news. Gerard and you have done a cracking job. We seem to have at least a work around for now. I'm amazed that it is still Version 1 of the BMU software. I have a service in November. I really hope MMC gets new procedures out to dealers by then.
Thank you so much for all your efforts, I tried to talk to my dealership when I brought the vehicle to get them to do a cell smoothing etc, but they didn't even know what that was. Now I have your videos to back up what I am saying. I have emailed my dealership asking to have this procedure done.
My head hurts a bit! Great findings and thanks to all Australian phev team. But you are the lucky ones! Because in other country's in Europe it is impossible to take the problem to the local Mitsubishi delars! And that's for those cars that are still under warranty. In my country they will never agree that they have a problem with the phev. In other words this year I should make the move to a second-hand PHEV, but which one. Still, I remember that whit the old 🔋 you had a range of over 30km, this is my daily commute, and will should take care of the problem. But the bad software is still there!
Word is spreading fast andy. People in norway has picked UP on the issue on their Facebook forum
That is excellent, the more people know and complain with their dealers the better. Public pressure is the only weapon we have to get this finally fixed.
well, small steps are better than none Andy, good job to all involved "thumbs up" & stay charged m8's
So 40Ah for 2018 and previous models and 46Ah for the new one. right?
We all owners all over the world must thank to Gerard and your team for this precious information and the posibility that MMC solves it through a future recall at the MM workshops.
I'm skeptic if we ever see and update for the BMU.
I suspect that the bmu having the wrong initial input from the beginning continously "think" that the battery degrades much quicker than it should and at every cycle of evaluation tries to "save" it, thus continuous drop in the apparent capacity. This way user should not be able to degrade the battery further but the loop continues in some kind of a vicious cycle and will never stop unless everything is reset and a proper initial value is inserted. The biggest bug in this system is that the BMU does not correct the predictions using, for example, max nominal V of the battery and uniformity of V across the cells.
I like your theory, J B!
Very interesting video thanks Andy. It is interesting to see that scientific proof has largely confirmed what I said in the earlier part of last year, and that is that most of these so-called deteriorating battery problems were in fact the BMU artificially strangling the amount of charge admitted into the battery using arbitrary parameters. I was wrong in assuming that this was to prolong the life of the battery, not that it would not have occurred as a result but it is much more the result of incompetence and laziness on the part of Mitsubishi motors.
It sounds like you are suggesting that the three procedures should be done to older cars. Mine has just gone over 30,000 km, so possibly the real battery degradation is not all that great.
We have to test the procedures in older cars and see if this makes a difference or not. One viewer has confirmed here it does not make a difference though...
unplugged EV
Andy, I would be quite happy to make my car available. Barry’s DOG recently said the battery was at 79%.
Richard
Doing the math on mine only the battery cycle degradation seems to line up roughly. 1969days old, 98600Ah of usage, 85.8% SoH.
Thanks for digging into this. I'm very curious on how this all turns out.
Almost makes the intro song more manageable.
Unless you had a reset at some stage...
True, true. I almost had it for a month. Could be the previous owner. But yeah, keep on digging! :D
But why no degradation at all after the triplet procedure? Those two factors should still affect the battery SOH after triplet procedure.
I'll show this in one of the next videos. It is still a bit mysterious...
Not sure if I totally understand what you are saying. We know the battery is a 40Ahr battery, but it is only ever charged to 4.1V. To actually get 40Ahr into the battery you need to charge to 4.2V until current reduces to about 0.05C. So I would have thought a full battery for the PHEV is about 38AHr, because Mitsubishi wants to keep the voltage at 4.1V or less for increased life.
Yes, that is correct. the actual cells charge only to 4.1V regardless what the capacity of the battery is. 4.1V is around 85-90% state of charge of a cell, so more like 36Ah (40Ah-10%).
The Dog always showed 38Ah after a reset or even in new cars so assumed this is the max capacity it has. The Dog should show 40Ah. if it shows 38Ah it is not aware of the true capacity and makes false assumptions.
Now it is getting interesting.
Could "we" reverse engeneer a working BMU solution from another manufacturer to reflash our cars?
We would be the first EV modding crew..
A question for Gerard Dunning
what will it give you? The satisfaction that your SOH number in WatchDog is more accurate? If the battery is discharging to 3.7V and charging 4.1V then you have the full capacity available for your driving - regardless of what the BMU is telling you
@@andrzejp8782 I am an engeneer. To be able to do it is the reward.
@@andrzejp8782 Are you sure about this? Is the car allowing the whole charge in and out when the BMU is reporting the wrong capacity?
On the UK PHEV group FAQ, they state: "Over the next 1,000 miles, your SoH and guess-o-meter range should relearn and hopefully start to increase, all else being equal. It’ll help if you also follow the tips in FAQ#M." It will be very interesting information if someone following the procedure on "A. Fix bad battery health data (without dealer)?" could confirm seeing any increase of SoH. According to the last discoveries, this would be impossible.
Thanks for the info Andy. I'm not sure about those percentages though as it would make my 62 month old MY14 with 49068km on the clock be expecting a SOH of 73.48% when I've actually got 85.3% showing in the Dog. Perhaps the percentage degradation calculation is not exactly 0.3% and 1% ?
I have wondered about this - the BMU is old and presumably slow by today's standards. Does it do proper floating point calculations, or is it using some low-precision 8 or 16 bit integer math?
Perhaps the calculations are simplistic and give different results compared to us using our scientific calculators and spreadsheets?
All possible, you may want to ask Gerard. I pinned his commend at the top here...
It looks like someone in the design office of Mitsubishi has stuffed up and did not follow through there design to work correctly, A back dated trade description would be nice for all the cars to be updated with the new software. I would be willing to pay a small fee for the update. I would have thought at the end it would save Mit- a lot of money in replacing batteries on car up to the 8 years old as this date has not arrived yet as no car is of that age yet. I would not think that MMC would endorse a no action policy in the light of bad press against the car and will have an adverse effect on sales and customer satisfaction.
Mitsubishi was not afraid of bad press when I started the channel. I don't think they are taking this serious enough. The PHEV sales may be only a very small number in comparison to their overall sales. We are talking to the Head of Customer Experience all the time and still did not get the attention we should. Not even if you have a camera as weapon...
Thanks for the video. However, you didn’t mention what the triple procedure is ... in what order?
My deep respect to you and Gerard ! Let's see if Mitsubishi creates a service procedure for this.
My PHEV dropped 2Ah and 5% SOH in 5000 km, so i'm looking forward to such a simple fix.
www.phevblog.nl/en
My usage = 98640Ah, so would be 15.9% capacity drop
And age = 1459 days would be 14.6%
Actual SOH = 90.8% (34.5Ah)
This does not match the calculation in the video, however the first 125.000 km the car was not charged by the previous owner. I got it with 95.8% SOH at 125000km.
When i only use the delta after i got it, charging twice per day (7356Ah and 113 days) the capacity drop using the formula is 2.32%.
Andy,
Not so sure about what you are saying.
As a matter of fact, Mitsubishi themselves have explained in one of their service bulletins that the BMU can only calibrate itself when it is regularly charged over A/C and even then, only "when certain conditions are met". Are they bluntly lying? For cars that are never externally charged or only via DC chargers, Mitsubishi themselves indeed say that the BMU has noting to go by, other than the preprogrammed deterioration params (we didn't know the actual params, so thanks for that). But we have also seen cases (on the same PHEVWatchDog,net you refer to) of cars where reported SoH spontaneously went up in several small steps when charging pattern changed. If you and Gerard were right about the software, this would be impossible.
BTW
My battery is 65+ months, so I should be down roughly 20% for this.
My Ah in/out is 170.000 Ah, so I should be down another 27% down for this.
In total, I should be down approx. 47%. But I am 'only' 30% down. Less than 2/3rd of what to expect based upon preprogrammed params. Yes, I have had several DBCAMs performed, but never a full reset. So, according to you, the DBCAMs should not have made a lot of difference.
Also, in 8 years, every battery should be down by at least 8 * 12 * 0.3 = 30%. Subtract some more for usage and about every battery would eventually be eligible for warranty replacement (provided the owner did not more than 100.000 miles).
Yes, not true, the BMU cannot calibrate itself without DBCAM. Obviously MMC is not aware of that and the workshop manuals are wrong and incomplete.
@@unpluggedEV So the people claiming to see their SoH value going up by regularly charing the car, such as Richi Jennings, are lying? And so are the stats on PHEVWatchDog.net that show SoH on some cars going up, step by step?
With all due respect, your explanation was not all that convincing to me (or to yourself, as you did not want to be quoted on your claims ;-) ) and the 'evidence' appears to be restricted to TeamPHEV members only. Tough to sell, if you ask me.
Terrific video. Many thanks.
Hi Andy: Obviously your work and that of Gerard plus many others including Gary from North Bay . Canada is so very much appreciated. Amazing effort. What begun as a new car purchase for myself and my wife has turned into a hobby AND possibly a new religion!!! BRAVO to all of you.
I have forwarded your episodes # 183 + # 190 to my local Mitsu dealer her in Toronto , Canada.. He is very appreciative and understands all. He in turn has forwarded the links onto his regional Mitsu tech. representative. My dealer is very familiar with the first 2 steps of your "inoculation". He has no knowledege or understanding of the 3rd step..." D.B. CAM". Have we understood this properly??? Can you further describe where to find more about this " D.B. CAM" ?? Thanks so very much. Ron
if Gerard already went through the pain of reverse engineering the firmware, why not modding it right away so that owners or Mitsubishi shops could reflash/reprogram the BMU themselves?
Because it seems like it lacks the necessary sensors to actually work - well perhaps not lack of - but in any case the sensors don't report to the firmware. You could re-flash the BMU, but you still wouldn't know the real capacity because Mitsubishi made capacity a simple calculation rather than real world. The cell smoothing procedure which is basically a cell balance and then DB Cam (whatever the hell that is?) seem to then be able to populate the registers with the correct values (provided you reset the BMU first and clear out all the existing data.) But after that the BMU will do what it has always done and start degrading the battery based on it's inbuilt firmware.
@@mondotv4216 , well maybe that's not needed to improve the way the BMU calculates the state of the battery (although it probably would be ideal if the BMU could access and use data from the real health state of the battery).
It may be worthwhile to check if it is possible to overwrite (change) both deterioration constants (charging cycles and age) in the BMU to a less aggressive value. That's not even a programmatic change, just pure configuration. This solution wouldn't fix the problem, but slow down the deterioration which in its stock version seems to be based on worst case assumptions.
If programmatic changes would be feasible then it may make sense to look at how exactly the math algorithm works in the BMU. If it is some sort of a recursive function - meaning, it calculates always on the basis of the previous result then it may make sense to investigate if the algorithm could be changed in such a way that it calculates always based on the initial battery health state value provided by the calibration process. So instead (x=x-constant) it would be (x=y-constant) - the X coming from the lower register and the Y from the upper register (see video).
Ideally Mitsubishi would change the firmware and introduce a solution that can read out the real state of the battery and update the upper register in the BMU accordingly.
Daniel D. You could do that but then you would risk getting incorrect readings on the GOM. I suppose that’s not such a big deal on a PHEV.
I've done the triple procedure (reset, smoothing and DBCAM) but my soh didn't went up.. It stayed at 73%... What does this mean?
That means that your battery is at 73%
Andrzej P if that is the case then my BMU was always correct. Which I dont think it is because I've witnessed 2% drop in 3 months before doing the procedures...
@@rubencosta3975 you see I believe that the car do not use the SOH reported by BMU for anything. It charges and discharges to true capacity based on cell voltage . Mine shows 88% on WatchDog but charges to 105%. I'm sure your will charge exactly to 100% of reported SOH since you just had DBCAM and your BMU have the correct value. But the point is that it doesn't matter what SOH BMU has - the car will drive the same. At least that's what I think.
That would be insane. I refuse to believe that just because I'm a good person and think the car is great and the batteries are great and everything will be great eventually..........
I know exactly what you're saying and I hear you for a while but I did not want to hear that!!!!
Well, once my car goes through the procedures we will find out I guess, right. If the SOH does not change anymore and sticks to 40Ah the only method to measure degradation is the charging energy.
Thanks for this info!
Have you got a yellow card in your Dog's Battery Condition tab? How old is your vehicle, km, do you know how much kAh you had on the Dog before the reset?
Do you recommend performing the triple procedure even before getting your new Outlander? Afterwards the system should know the real capacity, shouldn't it?
By the way: Thanks for your great work!
Thanks Nicolas. The triple Procedure does not seem to work over time. It looked good at the beginning but SOH is going down again after a while in the same pace as without the procedures.
It calibrates the BMU to the correct capacity though, that is correct. It's the best process we have at the moment...
They know the problem and they won't fix it? Then they won't ever sell me a vehicle. Bye bye Mitsubishi sales team.
How could Mitsubishi Australia "fix" anything ?
They'll ALWAYS have to go past Mitsubishi Japan to even consider fixing anything.
Hmm this is very interesting, I’ve not looked at my figures yet, but if the degradation is purely algorithmic, I’m wondering why are the 2019 batteries on such a steep degrade path, compared to earlier models. Any thoughts? If the code is unchanged the degradation gradient should be the same?
Ian U.K. maybe if the 6200Ah of charge+discharge value has not been changed in the new ones’s BMU, that 1% down is more frequent and shows bigger degradation steps down as well
Alejandro Carrera I understand your point entirely, however taking my own car as example, the battery is 135 days old, I’ve had the car 2 months. The Dog says 41.9Ah SOH and total energy use is just 3734Ah. So, you can see that I’ve not reached the 6200Ah threshold yet, so can discard that, and worst case 4.5 months of 0.3% degradation for time. 46Ah x 4.5 x 0.3% = 0.621Ah maximum degradation. My actual degradation is 46 -41.9 = 4.1Ah ie 8.9%. Way off, so there’s more at play here I believe.
(Mark Walkington, NZ here ) Good work everyone!!! ... however if the BMU has such a direct algorithm can we pin it down precisely. I have applied the the assumed degradation factors to my 146668 Ah usage and 1899 days and this predicts even higher degradation. i.e.1- ((146668/6200)*1% + (1899/(365/12))*0.3%) = 57.6% (my vehicle reports 68.9%). Any other views on this equation?
Upper Hutt Cricket To be honest I was also thinking the same thing... However that is what the 'software' reckons, not the actual degradation...
I don't think it calculates correctly if there is no bearing to the actual capacity to start with. The reported 68.9% are from an uncalibrated BMU, right?
WTF??? :D SO I am already at 83% of SOH? :D After 13000km?? :D Now, how do you explain the 107% SOC? Did the BMU somehow learn or it's the charger that decides that Vmax is not yet reached?
Ps. still no global solution because my dealer won't listen to Andy...
The BMU does decide when to stop with the charging but does not learn from it.
So for my 2019 coming in June. Do I need the Dog to see if the dealer actually did the 3 steps? And how can you see it? And of course thanks TeamPHEV for all this info 👌
Just drive it, DON'T get the Dog!!!!!!!!!!!!!!!!! Ever!
@@unpluggedEV being sarcastic yes? Otherwise you guys did not know about doing the 3 steps. But I bet it is a bit stressful to see the rate go down so fast.
@@benjaminlibotte8471 Seriously, don't get the Dog. Keep watching Sturm der Liebe and that's it. Don't measure your own car.
PS: You can now get a Model 3 for the same price as the Outlander... just saying...
@@unpluggedEV model3 is too small for my needs and waiting list will be very very long and it remains to be seen if it goes for 40k-ish US $ or €. And there is not a nice unplugged EV to watch about them, unless you have plans in that direction Andy 😉 on a site note... it's better to learn how to walk before start running so a PHEV is the in between step for me and who knows in 5 years cars run on water
I'm gest curios why don't you replace lithium cells whith lithium titanate the price of a diy conversion is about the same as a new battery .
The battery was free under warranty.
Wont the degradation pattern stay the same then? If it is just based on percentage and usage?
I think it's a percentage of the available capacity. So the calculated degradation is not linear and also depends on your usage, which is probably not the same every day either.
unplugged EV if it only cares about cycles and time wouldnt that mean that it will always be the same, the only difference is the available capacity of the battery?
But cycles are heavily dependant on your driving style. Remember even the regen putting energy back in the battery counts towards this 6200Ah=1%...
@@unpluggedEV My case is that the potential degradation pattern would still be the same if you look at it in a graph just that the values would be a little different. Since its always the same calculation(?) and the only value that actually change from the BMU fix is the total capacity of the battery?
Hi Andy, Gerard
Unfortunately the things don't seem so be as simple as explained. The data from my car does not fit in your explanations.
Currently I have SoH 83.2%=31.6Ah which means that my car is one of those with 100%=38Ah.
In the same time the total current is 110 000Ah which should mean 18% degradation according to your formula. It is also 58 months old which would mean 17% degradation more.
According to your degradation formula, the BMU should have calculated 17%+18%= 35% degradation.
Lucky, my SoH is 83% instead of 65%.
Apparently, there are other parameters that the BMU considers.
Also, strangely enough, after the first 3 years and 140 000 km with the previous owner, the SoH was 92%. Now only two years later and 40 000 km more, my SoH is 82%, so the degradation started dropping faster with my usage.
Any thoughts on this? How does it fit in your findings?
Maybe your had a reset before you got it...?
You can ask Gerard directly, I pinned his comment here at the top.
Recalculate like the percentage you get on leasing in a bank: apply percentages every time she SOH is changed. Do not apply projected degradation to the initial capacity.
I keep asking the same question over and over - does SOH number effects the power the battery can take or the miles the car can drive ? I'm afraid Mitsubishi don't care what SOH BMU is reporting! You keep ignoring this. I'm afraid that if you would do your DBCAM to your old battery, your range would still be the same.
Andrzej P As there world be physically more ah, then yes, it would affect range...
I'm not ignoring this. I'm hesitating to reply... I'm scared you could be right.
@@unpluggedEV I think we are going back to the problem #1 that was discussed many times - small battery (12kWh) in a large heavy vehicle that operates for months in EV mode. Those batteries must work under very high stress and they just degrade much faster than in other BEV's . Not as fast as BMU reports, but still much faster than we would like. Perhaps Mitsubishi should turn on the ICE much faster or just increase the battery size, lower the EV power etc. Last month I got a 40kWh Nissan Leaf for my wife and we drive it now much more than Mitsubishi. The Leaf is smaller, lighter and has over 3x bigger battery. Also, we didn't have to charge it to full even once during the last month and our SOC didn't drop below 35% neither. In Mitsubishi we would have to do probably 20-25 full cycles during that time (25% through 90%) - I think this is the reason why Mitsubishi batteries suffer premature capacity loss.
@@Specwinclean There is as many Ah capacity as you can charge into the battery - no more no less. If your EVSE or WatchDog shows you that you charged to 36Ah then your battery has 36 Ah capacity available for your drive, regardless of what BMU "calculated" SOH is showing you. Also in all LiIon batteries, the SOC is proportional to voltage if your battery is charging up to 4.1V then it's charging to the full available capacity as designed. It won't suddenly charge to 4.2V if your BMU start showing you 100% (40Ah) SOH - there is only so much Ah that you can pump in the battery until it reaches 4.1V/cell.
Correct, but if the BMU says that you only have a maximum of 25.5 Ah, you won’t be able to put any more in, because it’s the BMU that says that it’s full, but you are correct I think with your summation of this type of EV.
So let me get this straight. Mitsubishi's BMU is really nothing more than a glorified counter that "guestimates" battery degradation (and very aggressively it seems)? It doesn't measure actual cell capacity? Ever? That seems like a very odd engineering decision. And it explains a lot. Like how you can get your car back from a dealer after getting the battery tested and suddenly it's range jumps up to near new only to degrade again very quickly. You would think at the very least Mitsubishi might introduce some kind of per cell amps in and amps out monitoring and compare it to actual voltage. It's not exactly rocket science. At the very least they should be able to do this for the pack as a whole.
Honestly, this all develops to a real nightmare. I just hope more and more owners jump onto the bandwagon and complain about this. Finally this should urge MMC to provide a fix. A real fix.
Another video, another bout of blood pressure issues.... It shouldn't matter how old the Fing cars are, they were screwed over by Mits and their shitty build methods. Tesla is beginning to look better and better at this rate. So, basically, I now have to convince the dealer to do a 3 day plus procedure, for free, because the manufacturer doesn't actually know how to build cars...
Hello, can we say class action lawsuit...
Better opening again...