Cell Voltage Loss Inspection test (cell-pair test), P33E6

My Nissan Leaf Forum

Help Support My Nissan Leaf Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
Marktm said:
Will set up a "test", take data and get back. Fun project for a retired guy! Just hope I don't have to push my leaf for a block to get to the front of my house after I went too far with battery depletion :shock: I was definitely a newbie at that time!

Interesting concept of pushing both "pedals" - it would seem regen and accel at the same time would cancel, or is the CAN system smart enough to only use the disks brakes when accelerating?
^^
Test with the car standing still.
 
Marktm said:
Quite happy with my 2012 leaf, especially with the "Lizard" replacement battery. However, it does seem that there are two weak cell assemblies - # 75,76. Do you have a recommended procedure that might help rebalance - without "peeling" to the individual cells?

Any success with a replacement business? A company called Fenix Power is working on it in the US - no updates for sometime however :roll:

Only non invasive thing you can do is to keep the car in ON position as much as possible, and charge using the slowest method possible (L1, 110V 6A). That will allow for as much cell balancing as possible.

However, if the cells balance out good at the top SOC, then the situation will not improve. Bottom balancing an EV is inherently much better than top balancing, but the BMS doesn't work that way.
 
Dala said:
Only non invasive thing you can do is to keep the car in ON position as much as possible, and charge using the slowest method possible (L1, 110V 6A). That will allow for as much cell balancing as possible.

However, if the cells balance out good at the top SOC, then the situation will not improve. Bottom balancing an EV is inherently much better than top balancing, but the BMS doesn't work that way.

I will drive to "turtle and beyond" , leaving the Leaf sitting in the on position as much as possible, then charge using my L1 charger (or my OpenEVSE set at 6 amps) - I have the time. Generally, the weak cells #75/76 will be the lowest voltages when at 100% charge, but the differential may only be 20 mv.

Thanks for the advice
 
This just occurred to me, and it may be nonsense. That said, charging to the EQ stage (all 3 charging lights lit) and then running the heater on Low just might keep the voltage in the EQ stage a bit longer, giving the BMS more time to bring up the weak cells. You might have to turn the heater on and off several times.
 
Depleted battery totally today with mix of highway, suburban, sitting with drive on. The final SOC was 2.4% with 4 GIDS. Full power was 80,000 watts right up until turtle. When turtle kicked in, the max was around 7000 watts (LeafSpy battery bar readings). At present L1 charging (for the next 12 hours of so!).

The battery values at shutdown:

Turtle mode capture 7.22.19.jpg

Quite a surprise that the weakest cells are quite different this time - almost a year later than the last "depletion" and that the message is "all cells OK". The only DTCs are the A/C system - it's so hot, it seems 82785 and 6 are thrown regularly.
 
Interesting and good data.
Looks like 3.1 volts -> Turtle

To elicit turtle at a higher SoC due to weak cells you have to stress them, meaning sustained maximum load.
 
Marktm said:
SageBrush said:
To elicit turtle at a higher SoC due to weak cells you have to stress them, meaning sustained maximum load.

I actually did the hard accelerations almost until turtle
Then the weak cells passed the test ... for now

The good news for you is that when those weak cells eventually hobble the pack power, it will be a pack defect warranty rather than a degradation issue
 
Yes, now the decisions about how to make that happen quicker vrs extending battery life - within the confines of acceptable usage and care. I believe the "defective" is 10 years. However, a replacement "Lizard" start date might be in question??
 
Marktm said:
Depleted battery totally today with mix of highway, suburban, sitting with drive on. The final SOC was 2.4% with 4 GIDS. Full power was 80,000 watts right up until turtle. When turtle kicked in, the max was around 7000 watts (LeafSpy battery bar readings). At present L1 charging (for the next 12 hours of so!).

The battery values at shutdown:

I am going to re-iterate that it looks like you might have a couple slightly weak cells, but nothing I would call defective. This test really shows the car is running fairly well and the BMS is balancing out the differences. My 7-bar battery had significantly worse values than this, although it was a canary pack and heavily degraded.

The first LeafSpy you linked showed an average cell voltage of 3.45V, whereas the new graph shows a voltage of 3.1V. While this sounds like a huge difference, it's actually very small for the 24 kWh cells. See the SECOND discharge curve on this page: (the top discharge curve is the new 40 kWh cells):
https://pushevs.com/2018/01/29/2018-nissan-leaf-battery-real-specs/

The second curve shows the voltage falloff occurs at around 3.6V (use the top light blue line, as you were presumably not driving while getting the LeafSpy screenshot, which means you should be using the 1/3C line). If you look at this graph, you can see the difference between 3.45V and 3.1V is less than 1 Ah. Each cell holds 30 Ah, so a 1Ah difference is very small.

We know that the central battery cells in the Leaf pack heat up faster and stay hot longer, so they will wear out faster. This is amplified in hot environments, like yours. It is normal as the pack ages to have the center cells get slightly weaker than the outside ones and show small differences like this.

Honestly, I don't think you have a defect.

Marktm said:
Yes, now the decisions about how to make that happen quicker vrs extending battery life - within the confines of acceptable usage and care. I believe the "defective" is 10 years. However, a replacement "Lizard" start date might be in question??

Your car looks fine to me. I would not try to stress it to bring out the "defect", as I don't think you have one. You're likely to just wear out your battery more quickly.

Are you getting the range you would expect out of this pack for your driving conditions? Are there any other symptoms, or is the only concern that the car is showing voltage differences and turtled 2-3% sooner than it should?
 
Losthan;

I appreciate your factual, technical insights. I helps us that are not so familiar with Li cell characteristics (used lead/acid most of my life!), to understand the significant voltage/amperage parameters that can be used to analyse cell performance/aging/failure. I don't believe I have a range issue, nor a power issue. Yes, my main concern is degradation that seems to be quite rapid by any realistic expectations of a modern EV - especially when compared to (let's say) a Tesla.
 
Lothsahn said:
The first LeafSpy you linked showed an average cell voltage of 3.45V, whereas the new graph shows a voltage of 3.1V. While this sounds like a huge difference, it's actually very small for the 24 kWh cells. See the SECOND discharge curve on this page: (the top discharge curve is the new 40 kWh cells):
https://pushevs.com/2018/01/29/2018-nissan-leaf-battery-real-specs/

The second curve shows the voltage falloff occurs at around 3.6V (use the top light blue line, as you were presumably not driving while getting the LeafSpy screenshot, which means you should be using the 1/3C line). If you look at this graph, you can see the difference between 3.45V and 3.1V is less than 1 Ah. Each cell holds 30 Ah, so a 1Ah difference is very small.
This weak cell test *should* be performed dynamically, while the car is putting out maximum power. That is where the weak cells will matter, because they will cause the car to limit power or go into turtle despite a SoC that is otherwise high enough to not trigger those conditions.

We have been told in multiple threads that Nissan will not perform, or will ignore cvli results unless a specific DTC is logged. It is tempting to guess that DTC occurs when the system hobbles power output, but that would occur only if the power demand exceeds the system allowance.

Put another way, I suspect that sustained maximum power is the way to expose weak cells and log the needed DTC.
What is sustained ? I'm not sure. 30 -60 seconds sounds reasonable to me since it could be seen in mountain driving.
If I hard Mark's LEAF I would stress test it every 6 months or so, and a week before the warranty lapses.

@lorenfb, do you have an opinion or advice ?
 
SageBrush said:
This weak cell test *should* be performed dynamically, while the car is putting out maximum power. That is where the weak cells will matter, because they will cause the car to limit power or go into turtle despite a SoC that is otherwise high enough to not trigger those conditions.

We have been told in multiple threads that Nissan will not perform, or will ignore cvli results unless a specific DTC is logged. It is tempting to guess that DTC occurs when the system hobbles power output, but that would occur only if the power demand exceeds the system allowance.

Put another way, I suspect that sustained maximum power is the way to expose weak cells and log the needed DTC.
What is sustained ? I'm not sure. 30 -60 seconds sounds reasonable to me since it could be seen in mountain driving.

I get that the test should be performed at maximum load, but we're talking about the state of the cells and the LeafSpy that was pulled. In that graph, it shows that the weak cells are at most 2-3% lower capacity than the rest (excluding effects from cell balancing). This is not a significant difference and it doesn't indicate a cell defect. Encouraging the owner to try to "stress out" the pack to get those cells to fail quickly likely won't work--these cells were slightly weaker, but only slightly. Furthermore, those cells were not showing low voltage again in the next LeafSpy screenshot.

This is very different than, say, Dala, who had two cells that both self-discharged at a rapid rate AND had significantly less capacity than the rest. His car was going to turtle at high SOC's. That is clearly the case of broken cells.

Given how people with actual bad cells (not just slightly weak) have had trouble getting a DTC to trigger, I highly doubt a very small capacity difference will trigger a DTC, and attempts to get it to do so will likely just degrade the battery faster.

SageBrush said:
If I hard Mark's LEAF I would stress test it every 6 months or so, and a week before the warranty lapses.

I would not. Turtling a battery is hard on it, as it gets the charge level quite low. Unless he starts to see evidence of an actual cell failure, you're not likely to trigger a DTC and you are stressing out the battery pack.
 
No intent to accelerate degradation of the weak cells, and I do not recommend bring the battery SoC to normal Turtle (what is that -- 3.1 cell voltage or so ?)

I would stress the battery at somewhere between LBW and VLBW. It that leads to a DTC due to the weak cells the battery is defective.
 
Marktm's "weak cells" are not weak enough to be of concern at this time. For example, I had 363 mV difference last night on my way home. I also see VLBW come in with pack voltage above 360 volts (indicates a lot of energy is left in stronger cells). My experience is that results of CVLI tests are quite variable. LEAF Spy Pro performs the CLVI test whenever the average cell pair voltage is low enough to meet the guidelines in the service manual and the car is in neutral or park. I sometimes see LEAF Spy list 1, 2, or more weak cells and other times see it show all cells are OK when battery SOC is near VLBW or lower. So far, it has not set DTCs and there has been no power limitation (other than risk of sudden turtle mode or shutdown if high discharge rate is sustained long enough to keep minimum cell voltage low for too long). LEAF Spy did not show weak cells when I took the car in for the annual battery test in January so I was not surprised that Nissan did not do anything under the defect warranty--will see what happens next year. Regardless of whether LEAF Spy shows weak cells by module number or not, I can see from the cell pair voltage graphs that several modules located at the front of the battery are weak (very low voltage under load and higher voltage while charging compared to other cells). I will try to post some screen shots when I get a chance.
 
GerryAZ said:
Marktm's "weak cells" are not weak enough to be of concern at this time. For example, I had 363 mV difference last night on my way home. I also see VLBW come in with pack voltage above 360 volts (indicates a lot of energy is left in stronger cells). My experience is that results of CVLI tests are quite variable. LEAF Spy Pro performs the CLVI test whenever the average cell pair voltage is low enough to meet the guidelines in the service manual and the car is in neutral or park. I sometimes see LEAF Spy list 1, 2, or more weak cells and other times see it show all cells are OK when battery SOC is near VLBW or lower. So far, it has not set DTCs and there has been no power limitation (other than risk of sudden turtle mode or shutdown if high discharge rate is sustained long enough to keep minimum cell voltage low for too long). LEAF Spy did not show weak cells when I took the car in for the annual battery test in January so I was not surprised that Nissan did not do anything under the defect warranty--will see what happens next year. Regardless of whether LEAF Spy shows weak cells by module number or not, I can see from the cell pair voltage graphs that several modules located at the front of the battery are weak (very low voltage under load and higher voltage while charging compared to other cells). I will try to post some screen shots when I get a chance.

Has anyone seen evidence that the severely degraded 2011,12,13's will start setting the DTC and fail the CVLI test as the battery reaches end of life (2,3,4,5 bars)? If so, is there any way Nissan can back away from the warranty based on the "language". I'm simply wondering if the 10 year defect warranty might offer some "opportunities" for another round of replacements?
 
Marktm said:
Has anyone seen evidence that the severely degraded 2011,12,13's will start setting the DTC and fail the CVLI test as the battery reaches end of life (2,3,4,5 bars)? If so, is there any way Nissan can back away from the warranty based on the "language". I'm simply wondering if the 10 year defect warranty might offer some "opportunities" for another round of replacements?
What 10 year defect warranty? It's only 8 years/100K miles, whichever comes first.

Check your warranty booklet if you don't believe me.
 
Back
Top