Uggh! Too many options = brain fog

Topics concerning OEM and open source BMSes
Post Reply
User avatar
ZooKeeper
Posts: 507
Joined: Mon Mar 30, 2020 10:23 pm
Location: USA

Uggh! Too many options = brain fog

Post by ZooKeeper »

I am in possession of a 2017 Leaf HV pack (out of a 2012), it has all the later parts, BMS etc. My initial plan was to use only 240v (10F, 20R), now I am thinking that going higher @ 320v (since all my relay parts are from the same vehicle and can handle it) would provide a more enjoyable vehicle.

Why this topic? I am completely overwhelmed with the BMS question. If I use the whole pack, split in two, it is academic. However, if the two packs become isolated during charge/discharge, the Leaf BMS is fried; so that path has a lot of risk :(

If I split the 320v into two equal halves (20F, 20R), I *could* charge them separately as well, eliminating some measure of complexity as I could use a standard US outlet of 120v. But that affects the weight balance F-R negatively (designed for about 30kg of "fuel" in the front, not 90kg :O) Best option is like 16F & 24R for weight balance. This makes reusing the Leaf BMS essentially impossible.

If I use a hardware solution that will ONLY protect the modules from over-voltage, I can split the pack any way I want, but.... the upper voltage limit would need to be carefully set. From reading everything I can find (here, DIYEV, P&S, Nissan etc) it seems like 4.5v is the absolute max charge voltage - the point at which damage can occur, so keeping lower than that is critical.

Then there is the question of temperature. Charge between 0C and 40C ONLY, stop discharge above 50C or so to avoid thermal run-away. Yea, that is simple for me to implement with hardware on both ends of the limit.

The LAST thing I want is dangerous, then unreliable, then short-lived, then messy, in order. Since I plan to only S L O W charge, that should be easily controlled with the Openinverter VCU to limit total pack voltage. Given the low current, individual module or cell shunts could easily be fitted and provide more than adequate safety. The idea is a 4.1v Zener and 470R resistor for each cell, or *maybe even one Z4.1 + 1000R and Z4.3 + 1000R in parallel per cell* and let the VCU handle the cut in-out voltage for the pack. I would also use snap-switches to enable the charge above 5C and cut everything off above about 50C.

Now for the question... How crazy am I?
Huebner VCU controlling a Gen2 Prius Inverter powering an MGR
"Talent is equally distributed but opportunity is not." - Leila Janah
arber333
Posts: 3241
Joined: Mon Dec 24, 2018 1:37 pm
Location: Slovenia
Has thanked: 74 times
Been thanked: 223 times
Contact:

Re: Uggh! Too many options = brain fog

Post by arber333 »

Well TL431 is essentialy a zener with refference set to 2.5V. You set your resistor divider to 4.1V and connect PNP transistor to its discharge. Then if you want a strong drain you can use a small resistor with NPN darlington. I have that schematic somewhere here. I will try to find it. This is purely analog circuit without feedback function. So your charger better have a good HV limit.

Lately i am using a small circuit with HV and LV triggers and i use mosfets for drain and opto for feedback disconnect for charger.
I consider 10ohm resistor the largest drain, some 0.4A, that is still acceptable under cover of battery box.
User avatar
ZooKeeper
Posts: 507
Joined: Mon Mar 30, 2020 10:23 pm
Location: USA

Re: Uggh! Too many options = brain fog

Post by ZooKeeper »

Thanks Arber! When my parts get here I will do some testing, I think there is enough difference to stay well clear of 4.5v/cell, but I can always add another zener + resistor and adjust the ratings, if needed.
Huebner VCU controlling a Gen2 Prius Inverter powering an MGR
"Talent is equally distributed but opportunity is not." - Leila Janah
User avatar
johu
Site Admin
Posts: 5681
Joined: Thu Nov 08, 2018 10:52 pm
Location: Kassel/Germany
Has thanked: 153 times
Been thanked: 959 times
Contact:

Re: Uggh! Too many options = brain fog

Post by johu »

Leaf cells can't charge to 4.5V, I don't know any lithium cell that can. 4.2V should be absolute maximum.
Support R/D and forum on Patreon: https://patreon.com/openinverter - Subscribe on odysee: https://odysee.com/@openinverter:9
User avatar
ZooKeeper
Posts: 507
Joined: Mon Mar 30, 2020 10:23 pm
Location: USA

Re: Uggh! Too many options = brain fog

Post by ZooKeeper »

johu wrote: Fri Apr 23, 2021 3:26 pm Leaf cells can't charge to 4.5V, I don't know any lithium cell that can. 4.2V should be absolute maximum.
Good to have some authoritative data on that, I have seen max values everywhere between 4.00 and 4.50v, most in the 4.15v to 4.25v range. I was planning to have a max of 4.20v with "do not exceed" of 4.5, but looks like I should reduce both by a few tenths.
Huebner VCU controlling a Gen2 Prius Inverter powering an MGR
"Talent is equally distributed but opportunity is not." - Leila Janah
User avatar
Bratitude
Posts: 783
Joined: Thu Jan 02, 2020 7:35 pm
Location: Canada
Has thanked: 57 times
Been thanked: 168 times
Contact:

Re: Uggh! Too many options = brain fog

Post by Bratitude »

I’d set your max cell ceiling to 4.18 when you really need the full range. 3.8v majority of the time.
https://bratindustries.net/ leaf motor couplers, adapter plates, custom drive train components
User avatar
celeron55
Posts: 774
Joined: Thu Jul 04, 2019 3:04 pm
Location: Finland
Has thanked: 27 times
Been thanked: 110 times
Contact:

Re: Uggh! Too many options = brain fog

Post by celeron55 »

3.8V sounds awfully low. But as a general rule, for every 100mV of reduced charge voltage (from 4.2V), you double the life of the battery (and remove 10% from your current range).
User avatar
ZooKeeper
Posts: 507
Joined: Mon Mar 30, 2020 10:23 pm
Location: USA

Re: Uggh! Too many options = brain fog

Post by ZooKeeper »

Bratitude wrote: Fri Apr 23, 2021 9:45 pm I’d set your max cell ceiling to 4.18 when you really need the full range. 3.8v majority of the time.
I did some more digging, this time for photos and found the attached. Seems like lower than 3.5 and I'm off a cliff and marginal benefit much over 4.1v. This would put the center point at 3.77v (3.75 nominal is often quoted) and agrees with some LeafSpy data suggesting 4.17v is a 100% SOC number. That means, for 40 modules, the pack voltage should be 332v when "full", or as full as it needs to be.

That should work well with 4.03v zeners, which would result in 322.2v on the pack, suggesting a 320-325 volt upper charge limit will not overcharge anything. Turtle mode should come in somewhere around 3.6v/cell or 288v.
Attachments
Leaf charge-discharge
Leaf charge-discharge
Huebner VCU controlling a Gen2 Prius Inverter powering an MGR
"Talent is equally distributed but opportunity is not." - Leila Janah
Post Reply