Canion Battery Readings

[Malm]

First the facts:

- it's usual for me to see changes of exactly 5% standing still and off for some time (2, 3, 4 hours). Sometimes i lost 5%, sometimes i gain 5%.

- I gain regularly 5% when i charge at night to 94% until 5:00 am. Then i take the car from the garage (where temperature is 17º C) to the street (where there are 4º C), to let them lose temperature. When i start the car, 8:30 am, the car shows 99%. If i charge 95%, 96%, 97%, it will show 100%.

- I lose regularly 5% after driving some distance and let it stand still for some time. I end the trip with 60% and when i start the car after some hours it will show 55%.

So i think that: the car standing still and off withdraws 5% of SoC when battery temperature rise (i don' know how much or to where), because the energy stored is the same but the battery increases capacity. The car adds 5% of SoC when battery temperature goes down, because the energy stored is the same but the battery reduces capacity.

 

[cometboy]

Thanks guys for the comments and suggestions. Yes, I am in Adelaide South Australia.

Checked a few things today and here is an update:

The difference does settle back to around 40 to 45mv on the 2 offending cells.

Re the internal resistance, I see an 11 volt sag at the full acceleration of 150 amps. So happy with that - but then it is virtually a new pack.

I also stopped in to see the service manager just to give him the heads up. He assures me the vehicle had a complete charge once a month as instructed by Mitsi. This dealer has only sold one other i-MiEV but the company trained two staff members on i-MiEV servicing! They seem to know the product. They also think the cell balance will come good but have noted my concerns about the 2 cells.

So I guess it is now a waiting game and time will tell.

Thanks again

Bruce

 

[Malm]

After the 175 km mRR (shared wits Mrannen), after driving 201,4 km with still 3 bars left, after charging 3 bars going downhill, i have a new galactic record - The lowest SoC and voltage ever.

Nothing is impossible for Malm :lol: .

 

[Zelenec]

Scary!
There's something wrong with app or terribly wrong with the battery. Or are you just kidding us?
If you have 88 cells averaged at 3,986V, it totals in 350,8V.

 

[Malm]

I´m not scared at all. I´ve seen it so much times since i have canion, but only for fractions of a second. The difficult part was to catch the moment. Yesterday canion freezed in the right moment, so much luck. In the reality it has about 70% of SoC and 350,8V. Nothing wrong with the app, maybe this is a syntom of my degradation, that stands over 5% i think, or just addapting to new temperatures :? .

This is a fraction of a second, then all became normal. Happening many times.

 

[Zelenec]

That's better news.

After first few days I have monitored my battery I would say it's in good shape.

 

[Zelenec]

After full recharge SOC was 100% but the cells reached less then 4.1 volts. Is this normal?

Cells voltage remains in range of 0,03 volt, sometimes 0,04 volt when discharging (driving). In couple of minutes after pluged to recharge, they come close in range of 0,015 volt. Batt temps are in range of 6 or 7 C. I noticed that discharge (driving) and recharge with 2,9kW (from wall) doesnt rise the average temperature more than 2 degrees C. Even after being exposed all night to subzero C all night, cells temp in the morning are from 6 to 12 degrees C.

I think regarding this observations I have no reason yet to be consered about the battery health.

 

[Malm]

I don't see any problem with that. Cells all very similar, 100% of SoC, that's what matters to me. Very healthy.

 

[Zelenec]

Weird. While charging, at only 91% of SOC, the cells were charged over 4.1 V. Is this happening to you also?

 

[Zelenec]

One more observation. When the SOC approaches 0, the difference between the cell voltage drastically increases.

At 20% SOC difference varies 5 to 30 mV
At 10% SOC ... 0.1V
At 4% SOC (car stops) difference is already above 0.2 V

Cell #47 droped to dangerous low 2,86V

Evidently going to very low SOC is really risky.

 

[Malm]

"Weird. While charging, at only 91% of SOC, the cells were charged over 4.1 V. Is this happening to you also?"

Yes, to me. 91% and 361 V. Stop charging and gave me just 11 bars. Happened to me three times. Gave me an extra bar latter. Welcome to my world Zelenec. My I-MiEV don't stops to surprise me...

 

[Malm]

One more observation. When the SOC approaches 0, the difference between the cell voltage drastically increases.

At 20% SOC difference varies 5 to 30 mV
At 10% SOC ... 0.1V
At 4% SOC (car stops) difference is already above 0.2 V

Cell #47 droped to dangerous low 2,86V

Evidently going to very low SOC is really risky.

I can see that in mine at 14,5% SoC. It can stop at 14.5%. Number 73 is my worst. This happens when the blanket is covering the head. I have seen my 73 at 2,75 V. But with blanket covering the feet, i can go under 10% SoC.

 

[JoeS]

Malm and Zelenec, thank you for showing what happens at low SoC, but I'm confused: if you value your battery pack, why are you operating down in that potentially dangerous (for the cells) zone? Surely, with pre-planning, you rarely need to drive your EV at its range limits. I can see you having this discussion if the cells were bottom-balanced, but with our top-balancing it behooves us to stay away from the bottom end and not solely rely on the car's BMS to keep us out of trouble. For myself, it is very rare that I get down below two bars (I've never seen turtle), and especially now that CaniOn has shown that cells can start diverging at low SoC I'm even more keen on avoiding that zone.

 

[Zelenec]

Zelenec,... why are you operating down in that potentially dangerous (for the cells) zone?

You see JoeS, from 4% -100% SOC battery recieves less then 13kWh of electricity (about 80% of total 16kWh). That means BMS prevents to discharge the pack too deep even when car stops. It's still in the safe zone, if:

- you charge the battery immediately after full discharge (no additional selfdischarging)
- pack temperatures are in safe zone (e.g. between 10 - 25C)
- you don't push amperes to much in positive or negative direction (acceleration, regeneration) when in low SOC

People, who really know a lot about the Li-Ion tech, have confirmed that there is no risk if the above conditions are met in our battery pack. At the beginning of charging, the cells immediately balance, those most depleted are charged first. So, bottom balancing is present here when charging of course.

The greater danger for significant degradation would result if the battery pack would be left long-term in very cold or hot weather, or in case of strong acceleration or regeneration at extreme cells temperatures at full or empty SOC.

Why I do it? Because I need to. My trips are often far beyond 150 km, sometimes more than 400km. Public charging stations in Slovenia are not very wisely located so I'm forced to squeeze all the juice from the battery maybe once in every three months. In first year it happened even more often (once every week in average). However, I do not see any signs of degradation of the battery yet. All cells are very good balanced in all SOC values when discharging, except in very low zone (less then 6%). Temps of cells are never in greater differences as 5 degrees C. I believe we don't have to be affraid of full (dis)charge at mentioned conditions.

Of course, I stay in even safer zone between 20-80% SOC when possible but not affraid to use every electron when needed.

 

[JoeS]

Zelenec, thank you for the explanation. Yes, being extremely light-footed when the pack is nearly discharged is wise.

At the beginning of charging, the cells immediately balance, those most depleted are charged first. So, bottom balancing is present here when charging of course.

Sorry, but I don't understand this, as I am under the impression that the current is being fed equally into all the cells. The only current shunting I'm aware of is during the final top-balancing. Staying on-topic, using CaniON to look at the cell voltages and when starting to charge a deeply-discharged pack, do you see the voltage of the low cells immediately coming up to match the others before the others start going up?

 

[Zelenec]

using CaniON to look at the cell voltages and when starting to charge a deeply-discharged pack, do you see the voltage of the low cells immediately coming up to match the others before the others start going up?

That's correct. All cells join at almost same voltage values first (5-15 mV difference) and then during charging process they stay in close range. I'll take screenshots next time.

I even believe, when the car is just stopped for some time, the pack self-balance all cells, although not charging. For example, when I stop at home, the voltage is in the range of 20mV. When I come back in an hour or two, cells are balanced in the range of 5mV.

 

[JoeS]

using CaniON to look at the cell voltages and when starting to charge a deeply-discharged pack, do you see the voltage of the low cells immediately coming up to match the others before the others start going up?

That's correct. All cells join at almost same voltage values first (5-15 mV difference) and then during charging process they stay in close range. I'll take screenshots next time.

Zelenec, thank you for your explanation. Very interesting, and would imply that our BMS is more sophisticated than I thought. Since our iMiEV input power is incredibly constant until the very end when top-balancing commences, this would imply that there is some internal mechanism for actually increasing current into the lower-capacity cells while lowering the current to the others, all happening as the overall pack voltage increases. I'm not familiar with this type of BMS.

What prompted me was this post on the Australian EV Association site, as It evidently took repetitive top-balancing to bring the cells back into line:
http://forums.aeva.asn.au/imiev-measured-data_topic3870_post47032.html#47032
This would perhaps indicate that neither bottom-balancing nor intermediate-balancing is taking place?

I even believe, when the car is just stopped for some time, the pack self-balance all cells, although not charging. For example, when I stop at home, the voltage is in the range of 20mV. When I come back in an hour or two, cells are balanced in the range of 5mV.

I wonder if this is a no-load characteristic and does this change as soon as some current is drawn from the pack?

Interesting technical discussion. Anyone else care to chime in? :ugeek:

 

[Barbagris]

... At the beginning of charging, the cells immediately balance, those most depleted are charged first. So, bottom balancing is present here when charging of course...

I don't think so.

Our cells, at the lower limit, decrease its voltage very quickly as they are at the limit. Inversely, when we begin to recharge, increase their voltage much faster than those with more remaining capacity.

I don't think there is any "bottom balancing" system in our cars.

 

[Zelenec]

I can not say why and how this happens, it is true, however, that the cells voltage equalize after pluging in, although they were at the end of driving in a fairly wide range. Could be, there is no BMS low balancing and this phenomenon could be attributed to the charging characteristics of Li-Ion cells. I have no deep knowledge about li-ion tech, so I just made a conclusion from what I've seen with Canion.

 

[Malm]

Ok, let me show you three screenshots at a low SoC. In the first, the car stopped. In the second and third, maybe I could do a very few km.

The same car, the same driver. In all cell 73 is the weakest. In all the car or stopped or nearly stopped, but SoC one time was 14,5%, other 10% and other 25%. What do I think of this? My i-MiEV can't evaluate it SoC precisely.

If i don't let it rest for some time without driving or charging partially, it's SoC becomes more and more unreal. If I let it rest, then it shows a SoC that is more precise. And now i know that i can stop having three bars left :shock: :shock: :shock: , or even more. But i know how to avoid this strange behavior. Stop driving and charging for a while and it takes or give 5% of SoC.

Another thing, when charging beginning at low SoC, I don't see cells becoming immediately balanced, but at 50% SoC they are perfectly balanced (like always).

A degradation of more then 5% may play a role in this, maybe temperature variations too. But maybe Mitsubishi missed something.