62 kWh Battery Heater - Quest to Find It!

[knightmb]

A long time since this topic was updated, but I do have an update.

Dala made a great video about a 62 kWh repair where the video shows the inside of the 62 kWh in all it's glory. To no one's surprise, no physical battery heaters. But also to my surprise, nothing else that would facilitate any type of "external" heating system. Dala is confident and I have no reason to doubt his experience and expertise that just operating the Leaf is enough to produce the necessary waste heat to prevent cold battery damage.

That makes it easy for me to perform experiments for this. Simply cooling a Leaf battery down to ambient air temperature, just turn on the Leaf with nothing else running (climate control, lights, heaters, radio, etc.) and just have LeafSpy plot the battery temperature for hours and come back to check the results every so often to see if the battery temperature is rising or remaining the same.

Since it's still slightly above freezing temperatures where I am for the rest of the week, the battery is already cold-ish, but if I just drive my wife's Leaf in place of mine for a day or two, should be plenty of time for it to cool off for experiments!

 

[mux]

I'll just drop in my two cents since we've investigated this a long time ago and came to unsurprising conclusions.

First, there simply isn't room anywhere or any mention in any variant of the service manuals of heating devices in the gen5 packs. It just doesn't exist.

Li-ion cells hover at around 1kJ/KgK specific heat, so say you want to get to a safe-ish 0C (at that point, the pack can safely accept and deliver more than 1C, i.e. motor power isn't limited anymore), with a 400kg pack and starting from -15C would take 6MJ or about 1.7kWh.

At 0F, the allowed charge rate is about 0.1C at about 10mohm per cell in series (i.e. about 1 ohm total pack resistance). 0.1C is about 17A, so that only produces ~300W of heating power in the pack. That's when the pack is empty, at full the internal resistance is about half that, so say the average heating power during charging is ~225W over a charging session. From empty to full, you're putting 2.25kWh of heat in, so that's about on the edge of enough.

Discharging at these temperatures is probably more effective, because you can consistently draw more current than charge. In this weather, average power draw is probably 0.2C or more, so that would give you almost 1kW of heating power. That goes down considerably as the pack heats up, but even at 0C it should still be enough to keep the pack at least from degrading like crazy.

 

[knightmb]

Thanks for doing the research and math. I dug out my old 2020 manual and read the entire 62 kWh battery heater section.

This section got my eye:

The automatic climate control auto-
matically turns on when the Li-ion bat-
tery warmer uses electrical power
from the Li-ion battery. This is not a
malfunction. When the Li-ion battery
warmer operates, the temperature in-
side the vehicle may be warmed up.

As you mentioned, more heat will be produced extracting power from the cells. That leads me to believe that the "software" battery warmer mode is simply running the Leaf with a power load. I did some experiments last night to test this and I'll post the results in the next posting.

 

[cornbinder89]

That is what I guessed, but without a way to confirm, I didn't post. Sounds like the two post above have done some reading of the documentation on the subject, thanks for sharing.

 

[knightmb]

After gaining some more knowledge, an experiment to use this method to warm up the battery should be pretty easy to test. Pick a cold night, give the battery some time to cool down, then try out these "software" heater modes manually to see if they have any effect on the battery temperature. My first test was going to be, "just turn on the Leaf and leave it" which is simple enough. I have my Leaf parked outside in the cold, I'll just turn it on, turn off all accessories and climate control and let it sit like that for hours to see if that is a big enough power draw to either stop the battery from cooling down or even start reversing the battery temperature.

Experiment #1:
First I start with a baseline temperature reading. In this test, the (3) battery sensors were showing 57.0 F (13.9 C), 54.5 F (12.5 C), and 52.0 F (11.1 C) respectively. The reading takes place at 12:19AM. I turn the Leaf on to "ready" mode at 12:20AM. The outside temperature at this time was 28 F (-2.2 C).

For the first few minutes, the power load is around 559 watts before it settles down. I suspect this is because some power is being used to charge the 12V battery.

I come back to get another reading at 12:27AM, the power load has stabilized to around 311 watts and it stays this way during the hours I let it run.

I come visit 2 hours later at 2:23AM to get a temperature reading, the temperature has continued to fall, now the battery sensors are reading 54.1 F (12.3 C), 52.0 F (11.1 C), and 49.4 F (9.7 C) respectively. As I expected, this isn't enough power to have any meaningful effect on the battery temperature. Outside temperature is now 27 F (-2.8 C)

More experimental data to come!

 

[knightmb]

After seeing the results of the experiment and not being surprised by it , I figured it is time to crank up the power load. So I only turned on the cabin heat, set it for auto on 72 F (22.2 C) with air set to re-circulate. I figured this should easily put a 2 to 3 kW load on the battery now, certainly enough to do something with the battery temperature.

I started at 2:27AM, the battery temperature sensors showed 53.8 F (12.1 C), 51.7 F (10.9 C), and 49.2 F (9.6 C). The battery load started out high it around 4.2 kW.

I returned 12 minutes later to get a quick peak, the power draw had stabilized around 3.0 kW, but the battery was still losing temperature? The sensors showed 53.6 F (12 C), 51.5 F (10.8 C), 48.9 F (9.4 C). Odd, but I did think of something just then. The radiator fans are blowing because of the heat pump, maybe that freezing air is blowing around the pack and negating the heat being generated? To be fair, when using remote climate control on the Leaf, it leaves the heat pump switched off and just uses the resistive heating instead.

I switched off the cabin heat manually, make sure all accessories were off. Then I powered the Leaf off.

 

[knightmb]

For my final extension of this experiment, I then used my Nissan EV app to remotely activate the climate control. Sure, it will only run for about 15 minutes, but the battery temperature continues to fall as though everything I'm doing has no meaningful effect. At least this way, I can rule out the radiator fans cooling the pack more than what the ambient air temperature is doing. Once I could see that the remote climate control was active, I got a quick LeafSpy reading to see how much power was being used, get battery temperatures, etc. As you can see form the screen-shot, the heat pump is completely off (and I could tell from how quiet it was outside the Leaf as well) with no power draw. This final part started at 3:12AM. Battery sensors were showing 52.9 F (11.6 C), 50.7 F (10.4 C), and 48.0 F (8.9 C), the outside temperature was now 27 F (-2.8 C) at this time with a 2.4 kW power load.

I waited about another 12 minutes to check if anything was happening with the battery temperature, I only had a few minutes left before the remote climate control shut off automatically, so I wanted to get a quick LeafSpy reading. To my surprise, after all of that, the battery pack was still losing temperature. Granted, the power load was less too at 2.0 kW, so that was probably not helping, but at least the radiator fans were not blowing freezing air over the battery. So the final temperature after all of that was 52.7 F (11.5 C), 50.6 F (10.3 C) and 48.0 F (8.9 C)

When the next opportunity presents itself, I'll just block the radiator fans and let it run a lot longer (like an hour) with climate control running on a higher setting like 75 F (23.9 C) cabin; to see if has any positive effect on heating the battery. This concludes my first experiment with results I did not expect. So while only 24 minutes of running a 2 to 3 kW load isn't a long time perhaps? I was hoping to see at least some turn-around on battery temperature. I can only imagine how much heat loss exist when the Leaf is in true brutally cold weather conditions.

 

[BillAinCT]

Is it possible to use a can bridge of some kind to hijack the battery temperature readings and pass on much lower temperatures? If so, as one dropped the temp the BCM sees slowly, it could be demonstrated exactly what the Leaf does at 0F, -4F and below.

If you wanted to solve the problem, you would ask, "What could you do in software to draw the most power without driving?". By far the PTC heat at 6+kW would be about the best you could do (turn the high beams on too perhaps). If you could get the resistive heat on full for say an hour, then you might find that this is sufficient to sustain or even raise the temp. But wouldn't this also deplete the battery too quickly to be a solution? So is there some PTC "battery warming" setting that is the happy medium?

 

[knightmb]

Is it possible to use a can bridge of some kind to hijack the battery temperature readings and pass on much lower temperatures? If so, as one dropped the temp the BCM sees slowly, it could be demonstrated exactly what the Leaf does at 0F, -4F and below.

If you wanted to solve the problem, you would ask, "What could you do in software to draw the most power without driving?". By far the PTC heat at 6+kW would be about the best you could do (turn the high beams on too perhaps). If you could get the resistive heat on full for say an hour, then you might find that this is sufficient to sustain or even raise the temp. But wouldn't this also deplete the battery too quickly to be a solution? So is there some PTC "battery warming" setting that is the happy medium?

If I remember, the BCM takes care of this internally, we couldn't find a way to hijack it via the CAN bus. It would have been perfect to test with. Even better to manually switch it on before doing a Quick Charge to maximum power and speed.

That would be my concern too. Maybe I didn't wait long enough, but if I have to lower all the windows in the Leaf, turn the heat to 90F on max cabin fan speed and just burn 6 kW to start moving the temperature, either I'm doing it wrong or Nissan has another way that won't waste a ton of energy versus the simple solution of just running battery heaters like the smaller battery models do.

 

[mux]

I'm pretty suspicious of that passage in the owner's manual. That reads a lot like the e-NV200 manual, which *does* have a (heat pump operated) battery heating function that does exactly this. Because the refrigerant moderator is a fixed mechanical one, they can't fully shut off cabin heating when the battery is being heated, so the cabin will slightly warm up during a battery conditioning event. Did they just copy this to the gen5 Leaf without checking? I really think this is an error.

That all being said: the AC Auto Amp and VCM are doing the heating, not the battery. It's totally possible to send these messages with a CAN bridge. Maybe it's still using the same messages as the old TCU/LBC/VCM loop using 0x5C0. As long as LB_HEATEXIST is true (byte 4 bit 0), you can start battery heating using LB_HEATSTARTMAIL_REQ (byte 0 bit 5) and stop with LB_HEATSTOPMAIL_REQ (byte 0 bit 4).

I'm very interested to know if the 62kWh reports having battery heating. I only have one gen5 Leaf with traces and it says:

5C0|00 FE FF FF F8 FF FF 00

with byte4 = 0xF8 = 0b11111000, that last bit is LB_HEATEXIST and it's 0. So it's never going to be able to request heating. By contrast, this is a 24kWh with cold weather package:

5C0|80 64 64 01 6A B8 00 00

byte4 = 6A = 0b01101001<-- LB_HEATEXIST=true

 

[MikeinPA]

But the Plus will still need a way to warm up the battery. Just a thought, but if the car is plugged in, in theory it could initiate shallow discharges through the resistance heater, with intervals of charging, and keep repeating discharge/charging as needed. Would this warm the battery enough to make a measurable change?

 

[knightmb]

Ok, first time I've seen the info about starting the battery heater manually, nice!
As you suggested, probably won't work on the 62 kWh if they have changed this since it physically has no battery heaters. I was going to try another night of running a load on the battery, but I believe I'll just get the same results that an insane amount of power is needed to do it that way and would not be practical in a real-world use condition.

My next route would be to locate where the 3 temperature sensors are in the pack and just trying freezing those spots with dry ice, but I suspect that the entire pack will end up spreading the heat loss around and I won't be about to focus these cold spots on the sensors in a way that would trick the Leaf BMS into triggering this "software" heat mode, but I could try on one sensor spot and see if I can affect it in any meaningful way.

 

[MikeinPA]

I am seeing values of 3 kw and 5 kw for the Ptc heater assy. Cycling this on and off, charging at 6 kw in between, would create 200 to 300 watts of heating if I am reading MUX's technical discussion properly. This is similar to the values I am seeing for the non-60kw Leaf battery heater of 300 watts.

MUX: Li-ion cells hover at around 1kJ/KgK specific heat, so say you want to get to a safe-ish 0C (at that point, the pack can safely accept and deliver more than 1C, i.e. motor power isn't limited anymore), with a 400kg pack and starting from -15C would take 6MJ or about 1.7kWh.

At 0F, the allowed charge rate is about 0.1C at about 10mohm per cell in series (i.e. about 1 ohm total pack resistance). 0.1C is about 17A, so that only produces ~300W of heating power in the pack. That's when the pack is empty, at full the internal resistance is about half that, so say the average heating power during charging is ~225W over a charging session. From empty to full, you're putting 2.25kWh of heat in, so that's about on the edge of enough.

 

[knightmb]

So in theory, I could get a baseline temperature reading via LeafSpy. Lower all the windows in the Leaf. Turn everything off. Then use the Nissan Connect App to just keep activating remote climate control over and over. Then use the same App to start Charging for a specific amount of time, then get a temperature reading to compare. All from the comfort of my home, I like those kind of experiments.

I guess now I need to figure out how much time to spend doing this. Maybe 2 remote climate controls back to back for 30 minutes of waste heat and then 30 minutes of charging, so an hour to compare before and after temperatures.
It's going to be below freezing again tonight; my wife is going to think I'm crazy, lol.

 

[MikeinPA]

So in theory, I could get a baseline temperature reading via LeafSpy. Lower all the windows in the Leaf. Turn everything off. Then use the Nissan Connect App to just keep activating remote climate control over and over. Then use the same App to start Charging for a specific amount of time, then get a temperature reading to compare. All from the comfort of my home, I like those kind of experiments.

I guess now I need to figure out how much time to spend doing this. Maybe 2 remote climate controls back to back for 30 minutes of waste heat and then 30 minutes of charging, so an hour to compare before and after temperatures.
It's going to be below freezing again tonight; my wife is going to think I'm crazy, lol.

Lithium ion cell resistance generally increases with cold, so testing at your pack at current temperatures would show less internal heating than would occur at very cold conditions. So, you need to do the exact same test at two different pack temperatures, with one test as cold as possible. With three test you could build a curve of internal ohmic resistance...

 

[knightmb]

Last night was another good, cold night to test with. First I started by lowering all the Leaf windows so that I could waste tons of energy via climate control. Outdoor temperatures were 28 F (-2.2 C)

Experiment #2:
I start by getting some battery temperature measurements.
Battery is still cooling, the sensors read 54.1 F (12.3 C), 52.6 F (11.4 C), and 51.0 F (10.6 C)

I begin by using the remote climate control activation by doing two of them back to back for 30 minutes of power wasting fun. Right before the last activation expires, I get a quick LeafSpy reading to see how much power this is wasting along with current battery temperatures. I can see that my Leaf is wasting about 2.8 kW doing this exercise and that the battery temperatures continue to fall. Sensor readings show 53.6 F (12 C), 52.0 F (11.1 C), and 50.5 F (10.3 C). The battery isn't cold enough to negate the net heat loss, but that's ok.

Next, I plug in my L2 EVSE. I let it charge for 30 minutes and then come back to get another LeafSpy reading. No surprise to me, the battery is starting to warm up again. Sensors now show 53.8 F (12.1 C), 52.3 F (11.3 C) and 50.3 F (10.2 C). I was a little puzzled that the sensor at the front of the Leaf showed a slight decrease in temperature, but I suppose it would be within the margin of error since the previous changes in decreasing temperature were a lot greater.

I wait about 15 minutes because I want to get another reading before I plug in my L1 EVSE. The battery is now starting to cool again after only a short time of disconnecting the L2 charging. The sensors show temperatures of 53.4 F (11.9 C), 51.8 F (11 C), and 50.0 F (10 C). Will my L1 EVSE be about to turn around the temperature falling like my L2 did?

Next, I plug in my L1 EVSE. It will charge at 12A @ 120V instead my 220V L2 charger. I let it charge for another 30 minutes and get a temperature reading. I was not surprised to see the temperature started to fall again, now the sensors are reading 52.9 F (11.6 C), 51.3 F (10.7 C), and 49.6 F (9.8 C). Seems my battery is too warm to gain heat from L1 charging, but that seems to be an expected behavior in this setup.

Conclusion:
I was kind of disappointed in my results for trying to replicate a "software" heating mode. The battery is behaving the way I would expect with how much power I am using versus how much power I am pushing back in. What I noticed is that trying to waste power using the climate control is not really as efficient as I thought it would be.

The first issue is that probably 99.999% of Leaf owners are not going to be leaving all their windows down in the middle of a brutally cold weather event. With all the windows up and the heat being set for re-circulate, the Leaf cabin is going to be warmed up within a few minutes of using only PTC heat. Afterwards, the power consumption is going to drop a lot. The default remote climate control temperate is 75F (23.9 C), which is probably warmer than most of us would normally use on a drive, but still easy to reach in a cabin within minutes. The power consumption would easily drop down to the sub -1 kW range and stay that way unless Nissan is over-riding this and setting some insane cabin temperature like 90F (32.2 C) I doubt Nissan would do this, so probably using the default.

That creates a second issue, this only works if you have a way to put power back into the battery. So the Leaf needs to be able to cycle out some battery power for heat and then push energy back in to help create more heat. This means a 62 kWh battery Leaf just left out in the cold by itself has no way to do this. So, what is it going to do then, just run the climate control forever until the battery is depleted?

Mux has pointed out in their earlier post above, via the math, you need to be putting it about 17A of power into the battery to break even at 0F battery temperature. So a 16A L2 could cover that for charging. Using power from the battery of the same amount would also have a break-even point at 0F, but that means you need to find a way to use nearly 6 kW of energy continuously. Even though in theory, the PTC can use 6 kW (plus the other loads that run the Leaf can further use more power), but trying to keep that PTC operating at 6 kW for cabin heating that long is probably not possible without melting the inside of the Leaf.

To make things more complicated, what if you are using a L1 EVSE? It's not going to have enough power to reverse the battery cooling. What if the battery is at a high SOC where you can't push in more power?

I'm wondering now if Nissan has coded the Leaf to activate when any sensor reaches a low enough temperature instead of all 3. This would mean I have a chance to trigger this by just cooling one of them at the front or back of the pack where it is more easily accessible.
If anyone has a video of some 62 kWh in "battery warming" action, I would be curious to see it.