Page 132 of 166

Re: 2016-2017 model year 30 kWh bar losers and capacity losses

Posted: Sat Jun 23, 2018 1:35 pm
by dwl
jbuntz wrote:
dwl wrote:
jbuntz wrote:The lowest voltage was 3.426 and turtle had not showed up yet.

From one time back in October I got down to 4 gids. The spread was 233mV and the low was 2.832. Pack at 282v I sense that once you get below 10 gids it is dropping off really fast. On this one, turtle was showing.
It looks your car has changed estimations significantly since last October. Only 2 GIDs between lowest cell at 3.426V now (6 GIDs) and 4 GIDs at 2.832V back then should have far more kWh (GIDs) difference reported which is probably still sitting in the pack.
I got it to 4 gids. Turtle showing and unable to put it in drive. Low cell at 2.869 high 3.099 pack 286.48 temp 105F. Charging now 234V 29A
That real turtle is the point where a charge test needs to start. I believe the estimations are so far out that people now won't get anywhere near that low as the indications of low GIDs, LBW, VLBW are all coming on at much higher energy still in the pack. I note in your signature your SoH was getting low in October but that may also be under-reporting. This can be highlighted by logging with Leaf Spy as the car continues to be driven at moderate load down to actual turtle and I suspect the GIDs would pause dropping even though you are consuming energy. The good news for you is the firmware update is available in the US (but we can't get it here in New Zealand).

[EDIT] I forgot to ask if the voltages were prior to the start of charge or during the charge. Your Hx will swing voltages quite a bit when full current but still big voltage differences from your Oct 4 GIDs to June 6 GIDs.

Re: 2016-2017 model year 30 kWh bar losers and capacity losses

Posted: Sat Jun 23, 2018 6:31 pm
by SageBrush
The voltage curve is steep towards the bottom of the battery. Those last couple of Gids are barely going to register as a difference in charging kWh.

Re: 2016-2017 model year 30 kWh bar losers and capacity losses

Posted: Sat Jun 23, 2018 7:07 pm
by LTLFTcomposite
After two LEAFs and two Volts our family passed the last early adopter baton in vehicle electrification and dumped the 2016. The next stop if it happens will be in late majority/laggardsville several years from now, probably with a Tesla if it pans out for them as I hope it does.

No regrets, it's been interesting to say the least, both the cars and the conversation here.

Best to all.

Re: 2016-2017 model year 30 kWh bar losers and capacity losses

Posted: Sun Jun 24, 2018 12:43 am
by dwl
SageBrush wrote:The voltage curve is steep towards the bottom of the battery. Those last couple of Gids are barely going to register as a difference in charging kWh.
I agree if the low GIDs are correctly positioned on the voltage curve it could be a steep portion but I believe they are in the wrong place on the badly reporting packs. The question is where is that knee on the 30kWh batteries?

AESC publish the curve for the 56.3Ah module at http://www.eco-aesc-lb.com/product/liion_ev/ and for that pack you will see the knee is under 3.4V. There is a huge amount of Ah between 3.4V and 3.6V and if the cells are at 3.6V then the GIDs ver 100mV is far higher. AESC also state the nominal voltage is 3.65V which is saying a lot of energy is stored at lower voltages (noting the 41Ah modules for the 30kWh may have different curve). One challenge is to measure the voltage at moderate discharge rates as voltage will be inflated during rapid charging.

Re: 2016-2017 model year 30 kWh bar losers and capacity losses

Posted: Sun Jun 24, 2018 6:52 am
by SageBrush
^^ Good Link

Here is the relevant graph:

Image

I read it as meaning that a charge test from 3.4 v per module (326v per pack) is a pretty complete charge. Going lower looks risky, if the BMS even allows it.

Re: 2016-2017 model year 30 kWh bar losers and capacity losses

Posted: Sun Jun 24, 2018 8:45 am
by webeleafowners
Hey all. I posted this on another thread but thought maybe it would be useful to some of the techies here. This is a few days after the software upgrade. I find the sawtooth waveform of the last part of the charge interesting. Never seen that before and its not on any other charge graph of the Juice Box. Having said that I use a 120 volt EVSE for more than half my charging on the leaf. Here is the cut and paste.

----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

Charged it to 100 percent this morning. Road trip today. Figured I would post the output graph of the Juicebox as I noticed a couple things.

It started to taper at exactly 95 percent. Maybe a little earlier than normal? Not sure.

The sawtooth thing towards the end of the taper is not something I have seen on any previous charge session...ever.

Total range slightly higher than normal. Normally we are mid to high 180's vice this 197. When it was new it was mid 190's for our driving patterns. Road trip of about 160 KM today but no fast charge planned. Real range for us is always pretty much the same as the GOM. We have lower speed limits in this valley. Mostly 80 to 100 KMhr.

I have no idea if any of this is useful to anyone but I contribute what I can.

Image

Image

Re: 2016-2017 model year 30 kWh bar losers and capacity losses

Posted: Sun Jun 24, 2018 1:11 pm
by dwl
SageBrush wrote:^^ Good Link

Here is the relevant graph:
Image
I read it as meaning that a charge test from 3.4 v per module (326v per pack) is a pretty complete charge. Going lower looks risky, if the BMS even allows it.
I should clarify that these aren't the cells used on the 30kWh but understood to be for the 40kWh. The curves I have seen for 30kWh under test are a flatter line and drop at a similar voltage, although there is more of a slope rather than cliff below the knee, and had about 1.5kWh (under VLBW amount) remaining below 326V - still usable if pushing it and if BMS reporting was accurate. If anyone has measured data under moderate load (e.g. 1/3C rate) it would be interesting.

Re: 2016-2017 model year 30 kWh bar losers and capacity losses

Posted: Sun Jun 24, 2018 1:42 pm
by SageBrush
If NIssan is bringing into play the capacity below 3.4 V then they are idiots.
Since I don't think they are idiots (and don't care if they are, since I am not), we are left with the same recc for capacity determination:

Measure recharge from 326 V or a smidgeon less in a well balanced pack. If the capacity magically increases as a result of the Nissan software change -- great. If it does not then whatever change Nissan made is unlikely to benefit the car owner.

One caveat though: it is worthwhile to document the pack voltage at VLBW. If it was above 326 V before the reset and at 326 V afterwards then Nissan played fair.

Re: 2016-2017 model year 30 kWh bar losers and capacity losses

Posted: Sat Jun 30, 2018 1:34 pm
by jbuntz
SageBrush wrote:If NIssan is bringing into play the capacity below 3.4 V then they are idiots.
Since I don't think they are idiots (and don't care if they are, since I am not), we are left with the same recc for capacity determination:

Measure recharge from 326 V or a smidgeon less in a well balanced pack. If the capacity magically increases as a result of the Nissan software change -- great. If it does not then whatever change Nissan made is unlikely to benefit the car owner.

One caveat though: it is worthwhile to document the pack voltage at VLBW. If it was above 326 V before the reset and at 326 V afterwards then Nissan played fair.
My last test at turtle (5 GIDs) is 253.42V min/avg/max 2.500 2.640 2.786

Re: 2016-2017 model year 30 kWh bar losers and capacity losses

Posted: Sat Jun 30, 2018 8:33 pm
by SageBrush
@Jbuntz,
Can you estimate how many miles you can drive between VLBW and turtle ?
I presume it is off the highway ;-)