So much has happened. Where to start? Here goes:
A couple of months ago I decided this year's Good Weather Project would be to unravel the Mitsubishi I-Miev battery secrets or to at least chip away at them. My car was fine since it got a new factory battery pack four years ago under warranty. I still have 38Ah left. I found a 2014 I-Miev up in Dallas 8 years ago and a good friend bought it as his first EV. He loves it as I love mine. They are great cars for getting things done locally.
So, fast forward to current times. There is a local 2012 I-Miev on Craigslist that has severe degradation. The main guy at the car lot where it is for sale was a former customer of mine so he has been interested in letting me try to figure out what can be done to save the car. I thought that the car could be saved with some undamaged cells. I found some in Canada but shipping battery cells cross border was a bigger deal than I wanted to deal with. And in looking further at this pack when at a very low SOC it became apparent that there were more than just a few cells that needed replacing. Car has 19Ah remaining and knowing what I know now it is smoked. Needs a full cell replacement.
Now the guy who so graciously has let me poke and prod his car has a vendor that travels from lot to lot doing outside inventory for small lots. She drives an I-Miev or used to. SO, he said she talked about getting rid of it since it has a battery problem. She drives a Model 3 now. I was put in touch with her. She still has the I-Miev. (It has sat for two years.) She and let me come over to assess what I would give her for it. I could tell she really didn't want to sell it. So I proposed that I try to fix it for her instead. She agreed and we are off to the <turtle>races. I'm retired so it seems like a good use of time. My friend's 2014 car mentioned earlier had an anomaly in the last Canion scan that I didn't dig too deep into. I thought that he was likely gonna need some work done soon so what I learn on this woman's 2012 car will allow me to fix his car when the time comes. And fix mine when the time comes back around.
So the first thing I did on her car was to scan it to identify all of the low cells. If I did an exhaustive study on which cells were causing turtle at the 22ish mile mark then I could simply replace those cells and all would be swell, right? After several scanning sessions with 4 or 5 different apps and a pro level scanner I had a list to hang my hat on. I had become acquainted with another user here who just so happened to have a nearly full set of 26Ah cells which would be perfect for replacing her ailing cells. I bought 16 of those cells from him for Round 1. I dropped the 2012 pack for the first time with the help of the friend who owns the 2014 car.
I went about testing all of the 16 replacement cells to weed out any bad ones, if any. I charged each cell up to 4.1V with my iCharger and then discharged them to 3.5V at an 8 amp charge rate which is roughly what the car charges at in Level 2 charging. I made a spreadsheet with the serial numbers of each cell and the capacity of each in milliamp hours. The best were just under 30Ah and worst one was just over 25Ah. Most fell around 27Ah. Her scans showed a 19Ah cell capacity.
Then I started pulling the old bad cells from their plastic cages. They were puffed so it took making a special tool to place against my chest to push out all four cells in one lump since they are stuck together with butyl rubber pads. This took a lot of effort! I shot a video with my son demonstrating the procedure but I don't know how to put it here without uploading it to YouTube:
Well, as it turned out, I tested the old cells and they failed the same capacity test at an average of 11Ah. Pretty bad! I bought a $70 milliohm meter on Amazon to try to use it to predict which cells were bad but it just pointed to a smallish difference between "good" cells and bad ones. I ended up returning it as the readings were different every time I tested a cell. I would love to buy a Hioki milliohm meter some day just to satisfy that curiosity. Toyota techs use them.
The really unfortunate part is I got the bright idea to test a cell that was not deemed bad and would remain in the pack. It came out at 12Ah. Damn. That's not good. I tested more and found the same thing. It was then that I kinda knew this would be a longer process than I first thought. I proceeded to complete this first stage and replaced all 16 cells but decided to replace 2 more that were on the edge in the scans. Did that and reinstalled the pack. All of the cells that I replaced could not balance out at the high end of charge due to their much greater capacity:
They were more closely matched after some miles (shown under light load):
And very obvious further on in SOC (shown at rest):
I found that there were new laggards at the end of SOC. The range had not improved but was instead slightly worse:
I've found a different kind of locally-sourced NMC cell to try in the last month and bought some. I made a module (more details later) of eight cells and bought two more used LEV50's and replaced 10 more low cells. I put all of those in the pack and found that even though I had replaced 30 cells at this point there were still more laggards that would show up after the last batch were replaced thus confirming that playing whack a mole doesn't work of a fully depleted pack like this one. But the overriding issue on Round 3 was that there were now three distinct quality groups of cells that limited the amount of miles I could drive on the pack to 12.3. So, in conclusion, I can tell you that when you have a 12 year old Texas pack that is puffed and has 82K miles you are better off starting over. More to come on that front.
A couple of months ago I decided this year's Good Weather Project would be to unravel the Mitsubishi I-Miev battery secrets or to at least chip away at them. My car was fine since it got a new factory battery pack four years ago under warranty. I still have 38Ah left. I found a 2014 I-Miev up in Dallas 8 years ago and a good friend bought it as his first EV. He loves it as I love mine. They are great cars for getting things done locally.
So, fast forward to current times. There is a local 2012 I-Miev on Craigslist that has severe degradation. The main guy at the car lot where it is for sale was a former customer of mine so he has been interested in letting me try to figure out what can be done to save the car. I thought that the car could be saved with some undamaged cells. I found some in Canada but shipping battery cells cross border was a bigger deal than I wanted to deal with. And in looking further at this pack when at a very low SOC it became apparent that there were more than just a few cells that needed replacing. Car has 19Ah remaining and knowing what I know now it is smoked. Needs a full cell replacement.
Now the guy who so graciously has let me poke and prod his car has a vendor that travels from lot to lot doing outside inventory for small lots. She drives an I-Miev or used to. SO, he said she talked about getting rid of it since it has a battery problem. She drives a Model 3 now. I was put in touch with her. She still has the I-Miev. (It has sat for two years.) She and let me come over to assess what I would give her for it. I could tell she really didn't want to sell it. So I proposed that I try to fix it for her instead. She agreed and we are off to the <turtle>races. I'm retired so it seems like a good use of time. My friend's 2014 car mentioned earlier had an anomaly in the last Canion scan that I didn't dig too deep into. I thought that he was likely gonna need some work done soon so what I learn on this woman's 2012 car will allow me to fix his car when the time comes. And fix mine when the time comes back around.
So the first thing I did on her car was to scan it to identify all of the low cells. If I did an exhaustive study on which cells were causing turtle at the 22ish mile mark then I could simply replace those cells and all would be swell, right? After several scanning sessions with 4 or 5 different apps and a pro level scanner I had a list to hang my hat on. I had become acquainted with another user here who just so happened to have a nearly full set of 26Ah cells which would be perfect for replacing her ailing cells. I bought 16 of those cells from him for Round 1. I dropped the 2012 pack for the first time with the help of the friend who owns the 2014 car.
I went about testing all of the 16 replacement cells to weed out any bad ones, if any. I charged each cell up to 4.1V with my iCharger and then discharged them to 3.5V at an 8 amp charge rate which is roughly what the car charges at in Level 2 charging. I made a spreadsheet with the serial numbers of each cell and the capacity of each in milliamp hours. The best were just under 30Ah and worst one was just over 25Ah. Most fell around 27Ah. Her scans showed a 19Ah cell capacity.
Then I started pulling the old bad cells from their plastic cages. They were puffed so it took making a special tool to place against my chest to push out all four cells in one lump since they are stuck together with butyl rubber pads. This took a lot of effort! I shot a video with my son demonstrating the procedure but I don't know how to put it here without uploading it to YouTube:
Well, as it turned out, I tested the old cells and they failed the same capacity test at an average of 11Ah. Pretty bad! I bought a $70 milliohm meter on Amazon to try to use it to predict which cells were bad but it just pointed to a smallish difference between "good" cells and bad ones. I ended up returning it as the readings were different every time I tested a cell. I would love to buy a Hioki milliohm meter some day just to satisfy that curiosity. Toyota techs use them.
The really unfortunate part is I got the bright idea to test a cell that was not deemed bad and would remain in the pack. It came out at 12Ah. Damn. That's not good. I tested more and found the same thing. It was then that I kinda knew this would be a longer process than I first thought. I proceeded to complete this first stage and replaced all 16 cells but decided to replace 2 more that were on the edge in the scans. Did that and reinstalled the pack. All of the cells that I replaced could not balance out at the high end of charge due to their much greater capacity:
They were more closely matched after some miles (shown under light load):
And very obvious further on in SOC (shown at rest):
I found that there were new laggards at the end of SOC. The range had not improved but was instead slightly worse:
I've found a different kind of locally-sourced NMC cell to try in the last month and bought some. I made a module (more details later) of eight cells and bought two more used LEV50's and replaced 10 more low cells. I put all of those in the pack and found that even though I had replaced 30 cells at this point there were still more laggards that would show up after the last batch were replaced thus confirming that playing whack a mole doesn't work of a fully depleted pack like this one. But the overriding issue on Round 3 was that there were now three distinct quality groups of cells that limited the amount of miles I could drive on the pack to 12.3. So, in conclusion, I can tell you that when you have a 12 year old Texas pack that is puffed and has 82K miles you are better off starting over. More to come on that front.
Last edited: