I have steel plates for end caps… Might still work though.
The other 1/2 of the module got a little stretched but I see no hole and don’t smell anything.
Seem to have ruined a cell in one half and maybe the other…
It came apart easy yesterday but that was before I cinched it tight together to get the min width. That must have strengthen the adhesive bond. Darn it.
In hindsight, I should have just sprayed some WD-40 between the cells and continued slowly working it. Cardboard wasn’t working, glue was really tough. Brilliant idea was a metal spatula along with the cardboard but just as I was splitting it, I saw how the handle mounted to the metal blade and thought, ‘oh no’. 
And since I trashed the 2nd half of the battery so well, I might as well remove the metal cover on that half, chop down the sides and make a single 1/2 module. Both halves have a black plastic sheet separating the bag from the metal outside case.
31.75mm spacers cut to size. Out off daylight today but maybe tomorrow they go in the car.
22S battery pack in location!
Was not easy. Picture is after 2nd installation. I lifted it in with no handles and could not stab the two bottom bolts into their holes. Could not lift back out without the handles. Made second insert much easier.
Strapped down and rear brace in place. Will go with .25" aluminum eventually. Only one BMS module there now. Will 3D print a couple of housings for 3 modules each. Cover the battery and seal the modules(wireless).
I’ve been monitoring the charging with the stock Zivan charger and it’s looking great. It’s been running about 6A all along but now that the cells are in the 3.97V range, the charger has dropped the current down to ~1A and it keeps dropping. I only have one BMS monitoring/balancing module installed so I’ve been checking every 1hr to half hour.
So far, it looks like a 22S lithium configuration with a 4.0V/cell charge might not require any charger updating.
NOT SO FAST, once the charger turned off and the cells were at 4.0V(88V), I got -16 over voltage code. It worked fine earlier at 3.95V/cell(86.9V) so the threshold crossed is somewhere in those 2.1 volts. The 12 converter seems to bleed 4A so this is a big deal unless I want to charge and take off right away. Will likely look into what that resistor voltage drop people have used to fake a lower voltage to the controller.
@Inwo would you know how much current Pin 1 of the controller carries?
I put 5 1A diodes on it and it desoldered a joint. I’m guessing Pin 1 is more than a battery sense line and likely powers the controller thereby requiring some more beefy diodes.
Takes forever to discharge these modules and I’m currently at 3.91V with -16 still. I found a broken PC power supply in the scrap bin and pulled it’s bridge rectifier which dropped 1.5V and did the trick.
Yes, it carries a little current, but should be less than one amp.
I use 10w device to drop 15v and it gets hot. But not that hot.
At 1amp, if it was, each diode would dissipate 600mw. So it depends on the package. Some 1a diodes are pretty small.
Over 5w devices are not common so I have used transistors in the past.
Thanks, They were 1N4007’s and their larger cousin I found on another old board, 1N4004 are both rated at the same 1A but differing voltages. 1000V vs 400V which were both within spec( https://en.wikipedia.org/wiki/1N400x_general-purpose_diodes ).
I may order some 1N540x diodes to have on hand and for this patch. In the mean time, if I can find another old PC power supply to scavenge from I’ll just chain the two together for 3V drop.
My plan is to put 4 in series along with a couple of pieces of heavy gauge wire which will be crimped with the diode lead at each end and placed inline with the white wire near the battery terminal. The extra wire is to give the cimp something to hold and also with heatshrink over it all, the big wire will
keep flexing down to a minimum.
UPDATE: Took it out for another spin and as those who have done the conversion to Lithium batteries know, it holds it speed quite nicely. Seeing 3.80V on cells on level ground only dropped to 3.75V accelerating or going up a slight grade. Started with 3.91V and after maybe 4 miles it’s 3.85V steady state.
Let me know if you have interest in fast charging. I have 86v lithium chargers @29amps.
My customers are limited until (if) I can hack them for 100v.
What kind of bms interface do you use? These are enabled with +12v, disabled when pulled low.
Most of the guys I deal with are HV hotrodders.
I run 22s in my EL, so that one of them.
Sending one to Grant set up for J1772 level2, for testing
Interesting… My BMS is a wireless setup which consists of 6 boards which each handles 4S(one Leaf Gen4 module) except the last one which only half the board is uses since I’m running 22S(5.5 modules).
These boards xmit wirelessly the battery status to the master unit which sums them up and displays voltage(s), balance information, etc. The only other thing the battery module does is do topping balancing by shunting power on battery cells which have already reached the balance threshhold value. I have the firmware source so I can tweak my balance level and assign module numbers.
The master has many features which include battery charger contacter control(on/off) and IIRC there’s CAN capability in there too so it can “talk” to a charger and tell it what to do.
So far, with 22S the Zivan charger tappers off and then stops charging at 88V( 4V/cell ) which is perfect for my setup with the Leaf Gen4 modules. But if you’ve got an offer I can’t refuse, I would be willing to put the charger through it’s paces. And since we have a 2017 Leaf, it would be cool to use the same Level1 or Level2(hacked) J1772 charger. I’m aware of the open source J1772 firmware so it’s been on my mind.
I only had one tested with the j1772 vehicle plug, so it may be a week or two.
I’m still waiting for confirmation from Grant.
These don’t need hacking. They have the j1772 circuit built in. Only needs the vehicle end that fits level2 charger.
They also have can or digital input to connect bms.
I can email the pdf.
I will be glad to loan on for testing. I don’t have time to do it all myself.
Price will depend on demand and features when I learn more. These were $750 chargers, 1/2 price is what I’m aiming for. At this point I don’t know if they are all NIB or even working. The first 2 were out of the box and the 2 I opened were brand new.
Testing charge voltage soon as I drain 22s pack below 86v.
9-ohm load and data logger for volts.
1 Like
I will have to read up on the design of vehicle charge systems. It appears the charger is in the J1772 system while I thought the vehicle had the charger and mearly told the J1772 system what voltage and power it wanted.
Sounds like the incar system is much like what I now have for a BMS and the BMS’s job to communicate with the charger to tell it what voltage and current to provide. I will have to read up on how most EV production systems do it.
Yes, charger is in the car. The level 2 “charger” is 240vac power.
The car charger and level 2 device have simple 2 way comms to negotiate how much power is available and how much the charger needs.
In my case I’m assuming that the 2500w charger will work on a 20a 120v circuit at reduced charge rate. Not tested.
This is my initial data from bms on CV.
Charger does not seem to turn off without signal from bms.
Nothing funny on data logger except it reads a volt low compared to bms and meter.
looks like it worked very well. I will have to ask Michael( creator of the BMS setup I’m using ) what charger he has used and how he had his master unit commanding the charger.
As you know, the Zivan is far from a smart charger. From what I see others doing, it looks like their BMS is basically the charging system and battery monitor all in one.
Regarding the Zivan, it was good to see the Zivan go full 6A all the way up to 3.95V and then start tapering off and stop at 3.98V. This is perfect for the 22S lithium battery but for safety reasons, it would still be good to have a contactor to turn the charger off at any point should a cell go bad and start overheating or for some reason the charger didn’t taper off correctly. I’m expecting to only have to recharge once or twice a week at most with these Gen4 batteries.
