DIY Electric Car Forums banner

1 - 20 of 41 Posts

·
Registered
Joined
·
164 Posts
Discussion Starter #1
Hello all,
My Boxster conversion recently became drivable. Siemens 1PV5135, DMOC645, and Leaf pack with Leaf BMS.

When fully charged (4.10V) the cell difference sits around 30mV. But during discharge, after about 25 miles at highway speeds, the difference reached nearly 600mV, with the lowest cells under 3 volts as the higher cells stayed above 3.5 volts.
When the pack was charged after the 25-miles drive it accepted 12 kWh.
The cells are from a 2015 Leaf, gen 2 from a 24kWh pack. They were purchased from ebay as a lot about 6 month ago and the seller is refusing to replace any of them.
The majority of the cells are unusable. I need at least 50 miles of highway range.
What should I do?
I have a few ideas.
-Replace the worst 20 or so cells
-Replace the whole battery using a salvage pack
-Replace with Kia Soul batteries? There is a pack available nearby for $3500. I need information on the cells though.
Thanks!
-Isaac
 

Attachments

·
Registered
Joined
·
1,478 Posts
Have you checked the cell voltages with a voltmeter directly to compare with the bms values? your bms screen has several data fields that seem to indicate that it may be somewhat defective.

Did you make any capacity measurements on the cells when you made the purchase? Obviously you would be more likely to get an exchange or refund if it were requested soon after the transaction.

Did you experience any over-discharge events, such as driving around until the pack was out of energy and all the cells were pulled down low? We all know it can happen when first enjoying the full torque of an EV...

The kia soul cells might be a good investment if the capacity checks okay.
 

·
Registered
Joined
·
164 Posts
Discussion Starter #3
The BMS screen is an Android tablet running Leaf Spy Pro in a special BMS-only mode. I have verified that the voltages match.
When I got the cells all the voltages were exactly the same. I did not run capacity checks as the seller claimed that all the cells were at least over 80% capacity. In hindsight that was a bad idea.

No over-discharge except on the range test drive, where the pack went from full to LVC in 25 miles.

The real issue with batteries in this car is the space. The Leaf pack barely fit and a Volt pack won't work at all, else that would have been my choice.
 

·
Registered
Joined
·
975 Posts
Leaf cells are an incredibly bad investment, the sheer speed of degradation makes them hard to use
a volt pack from a junkyard you can check with a voltmeter is a much more reliable option
 

·
Registered
Joined
·
779 Posts
Only a CC-load discharge test gives useful health data other than "dead or not"

Unlike lead, peukert is so close to 1.0 you don't need to use the 20-hour rate, 3 hours is fine if the seller lets you.

Teach them how to do it, and offer to pay double for packs above 95% SoH, true low-mile wrecks.

Industry standard is EoL at 80% for perspective and once the decline's started it only accelerates.

Those bargain days are over I think.
 

·
Administrator
Joined
·
6,157 Posts
Isaac

Have you tried actually balancing your cells? - you can't balance them at an operating voltage
To top balance them you need to charge them so that they are on the rapidly rising part of the voltage curve

One of the reasons that I suggest this is that "BMS" - Battery management systems used to be notorious for actually UNBALANCING peoples cells

Try manually top balancing your pack - THEN see what capacity you have
 

·
Registered
Joined
·
164 Posts
Discussion Starter #9
Duncan - The Leaf BMS only discharges at about 10mA so it shouldn't be an issue. What voltage would you recommend for top balancing?
I have a Powerlab 8x2 which I could use with a computer to graph discharge tests.
 

·
Registered
Joined
·
164 Posts
Discussion Starter #10
What is the easiest balancing method? Should I parallel all the cells and connect a charger or charge each one separately?
 

·
Registered
Joined
·
975 Posts
What is the easiest balancing method? Should I parallel all the cells and connect a charger or charge each one separately?
If cell pairs are internally imbalanced it becomes a mess


Ideally charging each 3.7 volt cell in parallel with the rest is what you want taking them up via a regulated fixed voltage supply to near the top 90% SOC of the charge range for your chemistry.
 

·
Registered
Joined
·
779 Posts
The problem with "buy new" is that "new" is well over twice the price and much much lower quality
I was not talking about any particular brand / model of cell or pack.

Unless just learning playing around, second-hand sucks and getting poorer value higher risk every month.

So choose the best pack / cell type that you can find, new and from trustworthy sources.
 

·
Registered
Joined
·
164 Posts
Discussion Starter #13
If cell pairs are internally imbalanced it becomes a mess.
For this reason I'll charge the low cells of the modules first. Or use two separate chargers (without common ground) and bring both cells within the modules to the same voltage.

So choose the best pack / cell type that you can find, new and from trustworthy sources.
Volume and cost were driving forces in the selection of the Leaf pack. I have the entire pack in the engine compartment and there is no other space for it. Even a single Volt pack would not fit. Do you know of a cell type with better volume density and similar price?
 

·
Registered
Joined
·
779 Posts
Of course if new cells were offered at a price competitive with secondhand, their quality would be just as suspect.

I'm not addressing your case specifically, do as you please,

just expressing that over say 5+ years of cycling, normal mileages, I doubt the savings from buying secondhand packs is worth all the extra non-monetary costs and unreliability.

Of course going cheap is perfect for those just learning experimenting etc
 

·
Registered
Joined
·
779 Posts
If you can get to the individual cell terminals, I believe you can precisely balance them against each other to an average voltage point using

capacitive cell balancing devices, aka "charge shuttling", "transfer balancing"

while still leaving the pairs hooked up in series.

Check out ZHC "battery equalizer balancer"
Shenzhen Huaxiao Technology Co., Ltd

HA02 looks like good value, will do two pairs at a time.
 

·
Registered
Joined
·
779 Posts
Just realize if the cell **capacities** vary too much, packs are limited by the weakest link, balancing at the top SoC vs the bottom just shifts the same restricted range up or down.

There is still a protective BMS in place with per-cell detection that shuts the whole pack down, right?

Fun is fun until it isn't, boom bad
 

·
Registered
Joined
·
164 Posts
Discussion Starter #18
Just realize if the cell **capacities** vary too much, packs are limited by the weakest link, balancing at the top SoC vs the bottom just shifts the same restricted range up or down.

There is still a protective BMS in place with per-cell detection that shuts the whole pack down, right?

Fun is fun until it isn't, boom bad
Yes. I have an Arduino communicating with the BMS and monitoring all the cells.
Now to take out the pack and disassemble it.
 

·
Registered
Joined
·
28 Posts
Hello all,


When fully charged (4.10V) .....
Could you tell me exactly how your charging system determines that your pack is "fully charged" please ? -

..... remembering that a cells terminal voltage is not an accurate guage of being "fully charged"
 
1 - 20 of 41 Posts
Top