Valdemar
Posts: 2487
Joined: Tue May 10, 2011 10:32 pm
Delivery Date: 09 Sep 2011
Location: Oak Park, CA

### Re: Bad Cell in Leaf Spy: What to do about it?

Turbo3 wrote:The formula can be found in the 2011 April version of the Leaf Service Manual on page EVB-67.

If there are any other formulas please post them and where you got them from.

The only formula I know about is Judgement = Avg - (max - avg)*1.5 and the min voltage must first be below 3.713 volts. The Leaf must be parked. So you don't make this measurement while the battery is under load.

For this example the numbers are min=3.645, avg=3.673, max=3.690. Filling in the formula gives:

3.673 - (3.690 - 3.673) *1.5 = 3.673 - (0.017)*1.5 = 3.673 - 0.0255 = 3.6475

Because the minimum cell is 3.645 we know at least one cell is judged as bad.

It is possible that after heavy use the battery needs some time to stabilize. So if you leave the battery sit for a few hours and it still reports a bad cell it would have more significance.

You should definitely keep an eye on cell #28.

Does the service manual specify the SOC at which voltage measurements should be taken? Leaf Spy routinely flags at least 5 cells on my pack as bad around VLBW and below, no DTCs. I usually see this after pulling into garage, so it is possible this doesn't occur after the car sits for some time, need to verify. In any case I suspect it is not an actionable situation.
'11 SL, totaled
-1CB@33k/21mo, -2CB@53k/33mo, -3CB@68k/41mo, -4CB(41.5AHr)@79k/49mo, -5CB(38.85AHr)@87.5k/54mo
-0CB(66.14AHr)@87.5k/54mo (BBB), -1CB(53.92Ahr)@140k/29mo,
53.92AHr, SOH 84.5%, 140k miles

9kW Solar

Turbo3
Gold Member
Posts: 1974
Joined: Mon Jul 19, 2010 8:34 pm
Delivery Date: 12 May 2011
Leaf Number: 002191
Location: San Jose, CA

### Re: Bad Cell in Leaf Spy: What to do about it?

No SOC is not specified only the minimum cell voltage before a test can be run.

Turbo3
Gold Member
Posts: 1974
Joined: Mon Jul 19, 2010 8:34 pm
Delivery Date: 12 May 2011
Leaf Number: 002191
Location: San Jose, CA

### Re: Bad Cell in Leaf Spy: What to do about it?

achewt wrote:After doing my regular commute today, which I logged entirely in Leaf Spy Pro, on the 'battery cells' page I was given a note that I had 4 weak cells. The delta was 40 mV. Dash SOC 15%, 44 GIDs and 3.3 kWh remaining. http://kootenayevfamily.ca/img_1350/

This surprised me, as I haven't seen this message before (although I've only had the app for a few months, and have only fully logged several trips).

A few hours later, my wife took the car to the grocery store, and arrived home with 8 GIDs - I ran it down to turtle at 6 GIDs, then took a screenshot of the 'battery cells' page again - no weak cells anymore, despite the delta of 374 mV! http://kootenayevfamily.ca/img_1357/

Running the numbers through the formula up thread shows that my minimum exceeded the calculated value, so I guess that's why they didn't show up?

Anyways, my SOH has generally been quite high for how old my car is, the mileage and the reasonably hot summers, so I'm not really complaining, and will keep a watch out for more of the same. I can't help wonder though if it is somehow related to the other item I've noticed with 'clunking' sounds from my battery pack and odds 'burning' smells!

viewtopic.php?f=30&t=19883

It would appear that Nissan's formula they originally used has trouble if the cells are well matched at the lower voltages when there is little difference between min and max values. This may be why Nissan dropped this formula from their later Service Manuals.

In additional to the minimum voltage to run the test it looks like there should be a minimum delta as well. So if your cells are well balanced the test is not run. In your case the weak message when the cells have a delta of only 40 mv can be ignored.

I am working on the next release and will probably add a 200 mv delta requirement before I give any weak cell messages.

achewt
Posts: 63
Joined: Mon Mar 31, 2014 11:45 am
Delivery Date: 09 May 2014
Location: Nelson, BC

### Re: Bad Cell in Leaf Spy: What to do about it?

Turbo3, thanks for your input, much appreciated! And double (or triple, or more even!) for your work on Leaf Spy. Well worth the money. Make sure you cash in on your efforts now, since when we are all driving around in 300+ mile Leafs we won't need this type of software anymore.

Cheers.
Ocean Blue 2014 SL, 166,000km as of Sep 2018, 80% SOH
1 bar lost Aug 31/17 @ 54.5Ahr 83%SOH 79.72%Hx 127,413km
Commuting 142km round-trip; level 2 at home, level 1 at work
http://kootenayEVfamily.ca

Twizyflu
Posts: 2
Joined: Sat Jan 02, 2016 11:45 am
Delivery Date: 24 Aug 2015
Leaf Number: 011870

### Re: Bad Cell in Leaf Spy: What to do about it?

My LEAF gets 3 years old soon - in april 2016.
Only 19k km run. But sometimes, while driving, I see Cell #49 is weak - but not often. Most times it says: ALL CELLS ARE OK.

When the car is standing still over night and shunt order is right, it looks like one line - all cells.

SOH was 90% when i got the leaf (an old man was using it, no QC CHAdeMO done before, only L1/L2 charging at home), no high speed.
Now SOH is 93% and i did a few QC already. Many often, i am driving a 160 km trip - with one QC at half of the way.

At home only charging with 10A ICCB - 2.2 kW

Leaf has warranty until april 2018 on battery ... do i need to do some action because of this cell or is it ok?!

drees
Moderator
Posts: 6221
Joined: Fri Apr 23, 2010 3:51 pm
Location: San Diego

### Re: Bad Cell in Leaf Spy: What to do about it?

Looks normal to me. Perhaps a bit out of balance. Do you charge to 80% or 100%?

Twizyflu
Posts: 2
Joined: Sat Jan 02, 2016 11:45 am
Delivery Date: 24 Aug 2015
Leaf Number: 011870

### Re: Bad Cell in Leaf Spy: What to do about it?

Normally i charge to 80% but when i need the car the next days he gehts balanced to 100% always.
Ohne when using CHAdeMO QC he is not always balanced - but that is because the leaf stops charging at 91-98%...

BUT today - after driving 100 km (was on a trip the whole day with -0,6 °C Temperature of Cells in the morning) the SOH went from 93% to 94% (Hui...). I got the car in august with 90% SOH. The pre owner was never charging with CHAdeMO (!) and always charged to 80% he said. The car is nearly 3 years old. 19,5 k km run.

But: Usable kWh is 19,1 kWh - a few days ago (not as cold as today with up to -7 °C!) it was 19,4 kWh (with 93% SOH).

TimLee
Posts: 2811
Joined: Thu Jan 20, 2011 9:40 am
Delivery Date: 17 May 2011
Leaf Number: 2026
Location: Chattanooga, TN

### Re: Bad Cell in Leaf Spy: What to do about it?

When I first got to low voltage required for CVLI about two years ago it showed about eight bad cells early on when min to max delta was small.

By the time delta got above something like 65mV none showed bad.

As pack has gotten older and lost three capacity bars and now always charge to 100% and deep cycle a lot more, only one or two cells show "bad" and it is for a very brief time.
Last edited by TimLee on Thu Feb 18, 2016 11:41 am, edited 2 times in total.

Tim Lee
Chattanooga, TN

Man. Date: 03/10/11, VIN # 2026
Delivered 05-17-2011
Blue Ocean, 2011 SL-eTec

garygid
Gold Member
Posts: 12443
Joined: Wed Apr 21, 2010 8:10 am
Delivery Date: 29 Mar 2011
Leaf Number: 000855
Location: Laguna Hills, Orange Co, CA

### Re: Bad Cell in Leaf Spy: What to do about it?

It appears to me that the Cell Voltage data that the car has available to read
is sometimes from two different scans of the cells, part when the cells were
loaded (all at a lower voltage) and the other part when the cells were unloaded,
or possibly even during Regen (so the cells were at a higher voltage).

Thus, it might be important to let the car sit for a little while, or perhaps
operate at a constant load for a while, so that this split-scan data is not
a possible cause of the apparent two "different" clumps of data.
See SOC/GID-Meter and CAN-Do Info
2011 LEAF, sold in 2015
2010 Prius, 2014 silver Tesla S
Nissan EVSE, mod to 240/120v 16A
PU: SDG&E
Solar PV: 33 x 225W -> 7 kW max AC
To Sell: X-treme 5000Li EV motorcycle

Turbo3
Gold Member
Posts: 1974
Joined: Mon Jul 19, 2010 8:34 pm
Delivery Date: 12 May 2011
Leaf Number: 002191
Location: San Jose, CA

### Re: Bad Cell in Leaf Spy: What to do about it?

garygid wrote:It appears to me that the Cell Voltage data that the car has available to read
is sometimes from two different scans of the cells, part when the cells were
loaded (all at a lower voltage) and the other part when the cells were unloaded,
or possibly even during Regen (so the cells were at a higher voltage).

Thus, it might be important to let the car sit for a little while, or perhaps
operate at a constant load for a while, so that this split-scan data is not
a possible cause of the apparent two "different" clumps of data.

The original cell loss calculation was done with the Leaf in park so it is not under load.

It takes the BMS almost a second to scan all 96 cell voltages. It is done with a serial link running at 19,200 baud.