First Enginer PHEV install in the Portland Or area, Gen2

Jan 20th 2010
Hello Everyone,
So today we have the good, bad, and ugly.

Yesterday I drove a long ways. 54 miles total with more than 75mpg for most of that. About 8 miles from the end, the red light started blinking but the green light did not go off. So I turned off the system and drove on home.

At home it got ugly. I had 2 dead cells. I put a charge on for a little while and watched it closely but could not resurrect the dead cells. The 2 cells were buddied together. They were #7 on the bottom half of the string. My bad balancer is on the top half so these cell failures are independent for the balancer failure. This happened while I was driving. I believe I was at about 30%-40% left on the battery charge when I saw the red light blinking but did not hear any alarms. I then shut off the system. When I got home the bottom half balancers were only reporting 6 cells because the voltage on #7 was below the threshold for the balancer to recognize that it was there. The voltage on all other cells were at 3.1v (at rest) at that time. So I am assuming that I had 15 pairs of buddy cells giving me 46.5 volts when the pair #7 dropped below the lower limit. Because the DC/DC converter still was seeing 46v, it kept going. I do not know why there was no alarm. Because the DC/DC converter did not shut down, it continued to pull 55amps out of the pack driving these 2 cells to Zero.

At home, I immediately put the charger onto the system and got the bad cells up to 2.7v but they would not hold a charge. They were at that point about 10degF hotter than the other cells. I did not want to charge further because that could damage other cells with overcharging.

So I have sent an Email to Jack and he is super responsive as usual. I have a tracking number for 2 cells and a balancer.

Thanks,
Dan

 

Hi Banshee08,
So it cost you $60 to have 2 of these. Is that correct? Then I assume you pull wires to the front so that you can see a low cell light wile you drive. But it still takes you to take action to turn off the system if you get a light.

Am I guessing this correctly.

Why do you disconect the Balancers? Will this system not work with the balancers in place?

Thanks,
Dan

 

So what I need to do is to run the cells down to 3v or so and then let it set with the balancers working over night, Then in the morning charge until my first cell gets to some value but no balancers during charging or while charged, then I drive until the converter shuts down and then I plug in the balancers and let is set over night.

This means that I need to plan to start my charge so that it will end shortly before my drive because I want the time for the balancers to work to be after the drive and before I start charging.

Am I grasping your concept correctly?

Thanks,
Dan

 

Hi Banshee08,
Yes, I understand the balance at the bottom theory and agree that it is good.

I just do not see how you can continue it over time from a practical standpoint. You do a drive and the batteries may be 50% or 30% or 40% you can guesstamate but really do not know. If you are really at the bottom putting on the balancers for 2 hours brings you back to balance at the bottom. But if you are at 50% then putting on the balancers messes up the balance at the bottom.

I went 15 days before I really discharged to the bottom. Only 2 other of the 15 days were close to the bottom. The rest of the days I was between 40% and 70%. In this situation I feel that I cannot rely on staying balanced on the bottom no matter what I do with the balancers. If you need to take a day to disconnect all the cells and redo that accurate balance at the bottom with cells in parallel then that is not practical.

So that leads to needing a low cell alarm. With a good low cell alarm and overcharge protection, balanceing throughout the whole process will work fine and all the work to balance at the bottom is not needed.

Thanks,
Dan

 

Hello Banshee08,
So you do have me thinking.
I agree with much of your logic. But I also see flaws in some of that logic.

This is a deep discussion and I think it is worth having. However I think this thread is not the place to have a detailed balancing discussion.

Do you have a "balancing at the botom" Thread? If not, we could start one. Then we can direct interested people to go there.

Thanks,
Dan

 

Jan 22nd 2010
Hello everyone,
I am still working at it. I am making good progress and learning a lot.

I did find the rootcause for my 2 dead cells. My converter was not shutting down soon enough. I was able to get the converter to run with the green light on even when the total battery voltage had sagged down to as low as 43v. This is not within the Engineer spec for the converter. I do not believe that Jack had ever seen this before. If this happens you will, sooner or later, kill cells.

So I opened my converter and, with Jacks help, I was able to adjust it.

The no load output had been 245v but now it was 248v. So I think it had drifted. I set it for 241v and put nail polish on the pot.

The shutdown was much harder. I made several adjustments and of course the voltage changes with the battery discharge current. So I ended up with a setting that gave me a start up and run for some small amount of time and then shut down as the battery droops to between 45v-46v. At this 45v-46v condition the converter runs for 15-20 seconds with 45amps coming in from the batteries and 10 amps going out to the Prius. Then the currents slowly drop and the after a few more seconds the currents are below 10 amps input and 1 amp to the Prius and the converter shuts down and the battery voltage recovers to 48v.

I do not recommend that anyone do this process as I did it. If you think you may need your converter adjusted you need to contact Jack or your installer.

Now I am ready for my new cells.

Thanks,
Dan

 

Hi Doax,
No, To check this you have to catch the converter shutting down while you have your instramentation hooked to it. This is very difficult. It is also dangerous in terms of doing damage to a cell.

What I would suggest would be to check and record all the balancer readings at the end of a drive.

Drive the car as you normally would with the enginer system on. At the end of the drive, leave the car in READY. Leave the Enginer system on with the green light on. Hit the EV button so that the ICE will not come on. Go to the kit and check and record all the balancer data. Then turn off the Enginer kit and car and check and record the balancer readings again after 5-10 minutes of recovery time.

If you do this 5 to 10 times after different drives, you will get some good history to have an idea of how low the battery pack is at the end of your longest run. Be sure to check for the green light and no red light before and after each time you take data with the system on. If the red light comes on then the data will be differnt.

You should be able to add the total voltage from the bottom half balancer and the top half balancer. If this total gets down to 46v then you should be getting a red light. If it goes below 46v with the green light still on then you need to be very careful that the individual cell voltages are OK. Also if the "d" number is high then you need to make an adjustment to get a better balance at the bottom.

Hope this helps,
Thanks,
Dan

 

Jan 23rd 2010
Hello everyone,
Today I got notification that my new cells are in the truck for local delivery.

Yesterday I opened one of the dead Mottcells.

I found 3 round cells in parallel inside. The cells are connected to the output nuts with thin bus bars. There are 2 redundant bus bars for each output connection. Each round cell has each bus bar spot welded to it with 3 redundant spotwelds. So there are a total of 6 redundant connections to each end of each round cell.

The round cells are assembled to the output nuts and cover. Then this is dropped into the case and sealed with a solvent glue. Then a latex or silicon liquid is filled into the cavity to protect from vibration and reduce noise.

I inspected all this construction. It was all done very well. All 36 spot welds were good. I had no over heating anywhere in the Mottcell. I am confident that the Mottcell did not contribute to my failure.

I do not recommend that you do this type of disassembly. If you have a cell that you would like to disassemble. I suggest that you return it to Jack or send it to me.

Thanks,
Dan

 

Hi Andres,
Yes, I think I saw somewhere that the Mottcell 40ah cell was really 39ah.

I have seen somewhere that internal resistances on LiFePo4 cells is about an Ohm. I do not have any data from Mottcell.

atfsi took data while charging and discharging a cell. Here is a hysteresis charge generated from his data.

I think the difference between the charge curve and discharge curve is a function of the internal resistance in the cell.

Thanks,
Dan