[sciencegeek]

I would be curious how the Balancing works. If the Module has 12 cells, you cannot charge/balance the Module as one unit..you need to do the 12 cells. Handling the 12 as 1 would be a disaster.

I think you described it rather well.

I very much doubt that each actual cell is addressed separately. Here's why: The SoH report by JLR's Pathfinder addresses each module, not each cell. And they call each module "Cell," officially; even though a module is not a cell. Why would they not call the module "Module" when they could actually address each Cell? Their inaccurate semantics mean that there is a reason, and the reason is that the Module is the unit of control.

 

[bitzerj]

I very much doubt that each actual cell is addressed separately. Here's why: The SoH report by JLR's Pathfinder addresses each module, not each cell. And they call each module "Cell," officially; even though a module is not a cell. Why would they not call the module "Module" when they could actually address each Cell? Their inaccurate semantics mean that there is a reason, and the reason is that the Module is the unit of control.

Each Module could have it's own integrated balancing system. From the main control unit, it only looks at the total voltage of the Module. Most multi cell batteries have a balance port and a power port. I've seen 3s (3 cells) and 4s packs with an integrated balance board charge them through the power leads, or the typical setup that has two connections, the power cable and the balance cable. If the balance logic is per module (internal to the module), it would explain why the SoH report can only see the module.

I'm just guessing here... I would like to peek under the covers and see how this works.

 

[Astronomical_Phenomena]

I very much doubt that each actual cell is addressed separately. Here's why: The SoH report by JLR's Pathfinder addresses each module, not each cell. And they call each module "Cell," officially; even though a module is not a cell. Why would they not call the module "Module" when they could actually address each Cell? Their inaccurate semantics mean that there is a reason, and the reason is that the Module is the unit of control.

Nonsense. I've met maybe 5 people that use the terms "cell" and "battery" in their proper technical sense, and that includes a plethora of EEs. I think it far more likely that the terms have been mangled by JLR marketing, and that JRL battery engineers would use the terms "cell", "battery" and "module" correctly.

 

[sciencegeek]

Hmm. Here's the manual for a diagnostic / conditioning unit that second life EV batteries uses to diagnose and condition used I-Pace modules that they resell. It doesn't seem plausible that the I-Pace has that sort of system on board to do "cell" balancing. Maybe the per-cell diagnostic, but I find that unlikely too given what the pathfinder report does, and that pathfinder is purely diagnostic and can't fix a bad module. I really think that they treat the module as one unit while it's in the car. Once they take it out they can do the things that this unit does.

 

[bitzerj]

Hmm. Here's the manual for a diagnostic / conditioning unit that second life EV batteries uses to diagnose and condition used I-Pace modules that they resell. It doesn't seem plausible that the I-Pace has that sort of system on board to do "cell" balancing. Maybe the per-cell diagnostic, but I find that unlikely too given what the pathfinder report does, and that pathfinder is purely diagnostic and can't fix a bad module. I really think that they treat the module as one unit while it's in the car. Once they take it out they can do the things that this unit does.

The nature of LiPO batteries are such that if you use multiple cells you must balance them or they will become unbalanced and the risk of a charging/discharging fault increases exponentially. Remember Discharge is also an issue...Pull the cell down too low and you trash the cell, thus the need for a good balance. Again, I'm just talking about general LiPO battery pack feeding and care. The more I think about this the more I'm interested in learning more! Thanks for the food for thought!!!

 

[dernotte]

Interesting to see the "cell tap connector" pinout on a module (last page). They mention Cell 0-1-2-3. Because of 3s4p configuration ?

 

[bitzerj]

This is a pretty good description of Cell Balancing (Active vs Passive).

Be prepared to enter the Rabbit hole

 

[bitzerj]

Interesting to see the "cell tap connector" pinout on a module (last page). They mention Cell 0-1-2-3. Because of 3s4p configuration ?

3S4P means 3 in Series and 4 in Parallel, Series gives you the multiplication of Voltage, Parallel increases the Capacity (and raises the amps you can draw because it is spread across multiple cells). So (just for example), if the cell is 4volts and 1000mah, the 3s4P pack would be 12volts and 4000mah.

 

[Astronomical_Phenomena]

Interesting to see the "cell tap connector" pinout on a module (last page). They mention Cell 0-1-2-3. Because of 3s4p configuration ?

I would expect so, yes.

Just like an RC battery pack, 0 would be negative. (Probably.)

Of course, RCs are usually 3S1P - in our case, there are 4 strings in parallel.

At first glance the DLB1200 just looks like a jumped-up IMAX B6. Same idea, I would think.

This issue is very like what I was worried about when I looked at a FE that had 1 module significantly lower than the others. Here, one "bad" cell holds down the charge for the others in a module, but I don't think it's a stretch to think that the one "bad" module will hold down the charge level for the other modules too.

 

[Maxwell_400]

I would expect so, yes.

Just like an RC battery pack, 0 would be negative. (Probably.)

Of course, RCs are usually 3S1P - in our case, there are 4 strings in parallel.

At first glance the DLB1200 just looks like a jumped-up IMAX B6. Same idea, I would think.

This issue is very like what I was worried about when I looked at a FE that had 1 module significantly lower than the others. Here, one "bad" cell holds down the charge for the others in a module, but I don't think it's a stretch to think that the one "bad" module will hold down the charge level for the other modules too.

And there are two temp sensors, this was disclosed by JLR a while ago.

We can read out the min, max and average cell voltage. I would assume the weakest cell (4 in parallel that is) has out of all the (108) 4p cells the minimum voltage discharged and the maximum voltage fully charged. If so we should be able to calculate the capacity of the weakest cell relative to the average and best cell.

 

[dernotte]

I will definitely rescan all PID on the BECM after the update to see if more data are exposed on CanBus.

 

[pa28.rob]

In a 4P configuration, charging to 75% avoids overcharging the 3 good cells in parallel if the 4th cell goes open circuit. I wonder if this is just a numerical coincidence or actual insight?

Full charge on a lipo is 4.2V, 75% is about 3.9V so holding back the last .3V isn't enough to prevent a 3S4P exploding in the event of one cell going short circuit (charging to 3.9 x 3 ie 11.7V would put 5.85V across each good cell in that case, that's well into the pyrotechnic zone.) Cell derating as is common on EVs changes the numbers a little bit, but not enough to change the conclusion.

 

[Maxwell_400]

In a 4P configuration, charging to 75% avoids overcharging the 3 good cells in parallel if the 4th cell goes open circuit. I wonder if this is just a numerical coincidence or actual insight?

Full charge on a lipo is 4.2V, 75% is about 3.9V so holding back the last .3V isn't enough to prevent a 3S4P exploding in the event of one cell going short circuit (charging to 3.9 x 3 ie 11.7V would put 5.85V across each good cell in that case, that's well into the pyrotechnic zone.) Cell derating as is common on EVs changes the numbers a little bit, but not enough to change the conclusion.

Jaguar limits the cell voltage to ~4.15V or 96%. The magic top buffer is the last 4% or 50mV

 

[Astronomical_Phenomena]

In a 4P configuration, charging to 75% avoids overcharging the 3 good cells in parallel if the 4th cell goes open circuit. I wonder if this is just a numerical coincidence or actual insight?

Well, that assumes the string goes open-circuit. I would think a cell short-circuit would be the more likely issue.

Although, maybe the module can intentionally take the "bad" string / stack out of circuit. The bad module would then have 3 of the 4 strings available, putting it at 75% of normal capacity, with no risk of overcharging a malfunctioning cell. Keeps ya goin' until the faulty module can be swapped out, doesn't it?

I've not seen anyone say that the modules are capable of temporarily or permanently disconnecting a bad string, but I have seen similar in unrelated designs, so it's not impossible. If I was designing the HV pack, that would be a cool and possibly very useful feature.

If not, my guess is the software will charge to about 75% max, and is actually programmed to hold down the overall pack voltage to a few hundred mV above the highest voltage attainable by the bad module. But that implies a module operating on 2 of 3 cells (66%, not 75%) and that the car could still be forcing current through a bad cell (usually not the best idea).

 

[Pdxmattel]

I had my 2020 in this week and had the recall work done, which appeared to include more than a software update... replacement of some electronics in the charging system. The first software install failed, but the next day with an additional patch it took.

I slow charged to 100% overnight and my range was almost 20 miles higher than I have ever seen it (over 250). I'll keep you posted.

 

[Dr C]

I just had a similar thing as Pdxmattel. Had my “new to me” ‘19 S into a local independent shop for routine stuff and see if they could update the software. They said they tried to update but couldn’t. However, I went from a max charge range estimate of 220 mi prior to their check, to 260 mi now, FWIW. So something happened. Still have the same battery percentage drain per distance traveled commuting. And BTW, first post here. Love the forums and thank you all for the tips and useful info!

 

[BostonJaguar]

Hi all,

i ran across this in one of my news feeds tonight

GM buys Israel-based software startup to detect early battery defects
..

Read more at:

GM buys Israel-based software startup to detect early battery defects - ET Auto

Too bad JLR didn’t buy this company !

i guess that we all have high expectations for JLR as well as the dealership network to have the same admiration for the I-Pace that we share. It is easy to forget that we are in the early stage of transition to EVs. I went to the grocery store tonight and couldn’t help but admire the shape and beauty of the I-Pace viewing from a distance in the parking lot. The ride hone was great, the stereo system was wonderful, the seats were extremely comfortable. I saw a Mustang - E in the parking lot. Not that nice looking.
in fact it was hard to see a better looking vehicle than the I-Pace

Keep the faith. These are special vehicles

 

[andrewweir]

Does anyone know of any successful iPace "Car returns (Refusals)" against JLR as the battery recall issue is a design and engineering fault related to the specific LG batteries already involved in Claims against Hyundai (82000 recalls), Chevrolet Bolt (142000 recalls), Chrysler Pacifica (16741 PHEV)? My iPace has sat in the garage for 16 weeks so far awaiting the module replacement in the UK.