Phev Insight

[Mike Dabrowski 2000]

While not as good as a direct measurement, I just pulled out a spare BCM, and measured 1 meg between each of the taps(VHB6 to VHB5)( VHB7 to VHB.
This is likely the load on the ~ 15V -12 cell subpacks between the taps. That is about 15 uA, so a 10K divider (100K total) should work just fine.
The 180 ohm series resistors would be virtually invisible as far as any effect on the measured voltage.

[retepsnikrep]

Mike

I thought the load would be very low, so for this mobile test I have gone for the 20x9v batteries block made up so it plugs straight into my battery block bms connector points. It will pass through the standard 180 ohm resistors on the way to the bcm. I'll try and make it up this morning, might get to test it tomorrow and I don't even need to open the IPU compartment as the Battery Block BMS connectors are outside ipu comp

Yes 180v is a bit high but it's easy to make up the pack, and the 20 batteries I ordered come in a nice little plastic box!

Peter

[retepsnikrep]

Well I tried my 20 block of series 9v cells they were a bit over voltage (brand new) so it was about 189.5v all told

I tried with this first as it gives easy voltage taps for the bcm.

Anyway that gave an immediate P1576 MDM Over voltage error. No surprise really but it might have worked.

I shall try the potential divider tomorrow and I'll add an extra variable 10k resistor at the positive end so I can lop off the top few volts. Or I'll remove a 9v cell or two Until it works. I might try it with 100k resistors first as that would reduce by a factor of 10 the current drain from my little test cells. Down to 180ua from 1.8ma if I use 10 x 10k.

Peter

[Mike Dabrowski 2000]

Sounds like a good plan. Waiting for the results.

[retepsnikrep]

BCM Findings

Well I made up my little potential divider BCM fooler. And add a 50k pot to the + end and a 2ma led for a visual indication.

I adjusted output to 176v and Car took that without errors, even doing a positive recal on the drive with engine ticking over when the actual Li-Ion battery was less than the recal voltage level so that was good.

However when driving it after a couple of seconds under 50% assist threw P1576, there must be a double check on voltage perhaps in Inverter stage and if BCM and Inverter don't agree then I will test this a bit further and report back, it's freezing again here today. I suppose this is sensible function because the differing voltages would indicate a seriously bad connection/bad component between pack and inverter on a standard car.

Interestingly the led I had put into the pd circuit flashed brighter about once a second indicating BCM was perhaps polling the taps in order to assess pack voltage.

Cleared codes and back to normal li-fepo4 bcm taps for now.

[uhtrinity]

Well, you could use your divider to perform quick reset of the SOC gauge if anything, then switch back to pack voltage. That could be useful to get autostop back or avoid any 3 bar bug issues.

[retepsnikrep]

Originally Posted by uhtrinity

Well, you could use your divider to perform quick reset of the SOC gauge if anything, then switch back to pack voltage. That could be useful to get autostop back or avoid any 3 bar bug issues.

I could try just actually switching the supply to the pd either from my battery or the li-fepo4 packs.

That just needs one double pole change over relay. So when not moving the supply is from my little battery at 177v hopefully triggering a positive recal at regular intervals, when on move supply is switched to the li-fepo4 pack.

That avoids having all the taps as well on a HV replacement pack you just use the end voltage to supply the pd. They cells appear to the bcm to be as perfectly in balance as the 1% tolerance pd resistors allow. Small current drain penalty 1.5ma using 10 x 10k resistors and a 50k pot but that's not a lot.

Lot's to try.

[retepsnikrep]

Tweaked up charger voltage to 179v to aid cell balancing. Not that I have a problem, just decided to try it.

I've also wired back in my Master current sensor to the Master board so I shall see if that works.
I should be able to test this now, calibrate it and get it working on monitoring SOC!

I'll add battery temp function back in next if above works ok without interference.

Edit

Master BMS Current sensor works ok no interference Using MIMA lights as a rough guide I see I have it a factor of 2 out. It's showing 25A under 50A (two leds) drain. I'll chase that bug in my code down tonight. Master SOC determination and accumulation was working but due to above was off of course.

Finally concluded Autostop requires a minimum of 4 bars on soc gauge to work. When at 3 bars no autostop regen a bit and as soon as it hits 4 bars then works again.

MIght relocate charger to ouside the ipu compartment down the side and bolted to the side near the inverter under that styrofoam rear shelf support.

Edit

Edited soc code and added 10x over sampling. Seems to work well on short test run. Oem soc fell from 19 to 3 bars. My soc gauge said I had used just over 4ah. May be reading slightly high we shall see. Now on charge.

[retepsnikrep]

Now having had a few weeks with Lithium I can say I don't think I'll ever go back to the old Nimh system. It works far to well

With that in mind I keep looking at the Inside of the IPU compartment and how i can make better use of the available space.

I don't like the way my present installation takes up the wheel well, and the cells are a bit vulnerable to a rear impact.

I think I'm going to move the bcm and mcm out of there and bolt them to the reverse side of the back of the ipu compartment behind the seats if there's room. With the charger also moved outside the ipu comp and the old battery case removed i think i can squeeze the cells in there. I'll need a little bit of the battery case just to retain the battery power control board in place.

These further mods will impact on the layout and be of a more permanent nature but i think worth it considering i am very unlikely to sell the car. These changes will also shorten the power cables presently running from my battery blocks.

I don't think i will be able to simply bolt my battery trays into the ipu compartment (that would be too easy) and it will require a redesign of that layout, but I can take into account what I have learned about the rf noise etc and use layout/shielded cable accordingly.

Moving the cells to a slightly more central location will help reduce the tail end weight as well.

Just never happy with our projects are we!

[retepsnikrep]

Just Back from my trip 110mpg (UK) for the 270 miles. Not as good as I hoped (Reasons below) but not bad considering and a lot learnt.

1) Weather apalling snow/rain and wind. Temp 5C max
2) Hilly terain with lots of ups and downs.
3) BCM 3 bar bug/feature disabling assist at critical times.

The good news was no ima errors or codes and assist worked fine and I just about managed to flatten batteries after about 130 miles. Assist power went from 50% to 100% as cells warmed during journey, temps started at about 5c and finished at 22C and went to 25C after being charged.

The problem was the BCM bug which with a much larger pack was much more obvious. Yes after the initial pos recal you could drain down to 3 bars and a total of about 8-10ah out, but then it would drop off and you had to regen to get back to 4 bars or you lost assist. Sadly this kept happening at bad points on the journey and when faced with a 5 mile gradient to climb I was a bit unlucky but you were then forced to climb with no assist or even while charging destroying you previous figures I almost stopped at the side of the road and thought allowing it to regen there to 4 bars might have been better than continuing with no assist.

So basically it works but I need to fool the BCM with the current in and out hack. I'll implement that this week and report back on MIke's resistor tweak.

The 3 bar drop did not seem to be voltage related, but my cells never went near the bottom cut off voltage.

The cell remained well in balance and worked fine I was very pleased with them.
My BMS Soc counting worked well, still over reading amps a bit, but very interesting to watch.

Had a long and interesting chat with Greg Fordyce a UK EV enthusiast and one of my BMS supporters. Went out in his two EV's one running Lithium and our BMS and one running pb gel. Both very good.