Old BMS General Thread

[retepsnikrep]

Finally installed lithium batteries in Insight rear spare wheel well. Fit's like the proverbial glove!

http://www.solarvan.co.uk/bms/BatteryInstall.jpg

Also installed display unit in dash cubby hole and connected it to a nearby + feed.

Here's an old pic of it running earlier Master BMS software

http://www.solarvan.co.uk/bms/CubbyHoleDisplay5.jpg

Not installed cabling to rear yet or removed old nimh pack and replaced it with empty case and new BMS. But making some progress.

Interesting to see how extra weight in back effects car. Still got to take out 22kgs of old nimh.

I intend install/testing over next week or so in stages.

1) basic setup with onboard NG3 charger/bms and standard Insight assist/regen etc.

2) as above + test with MIMA which is the manually activated assist/regen system. (MIMA is already installed but can be deactivated for initial testing in stage 1)

3) as above + Speed sensor input (I need to opto isolate this) Software already has (mph/odometer) built in but I then should be able to work on the wh per mile stuff estimated range etc etc!!!

4) as above + Security feature (Insight fuel pump cut out possibly) 5 digit Security code system is already in software and drives a relay, just need to find the best circuit to interrupt in Insight to disable vehicle.

5) I may need to hack the Insight oem current sensor and divide output by a factor of 10 to compensate for the 10 fold increase in available battery capacity. We shall see

[GregsGarage]

Peter, the pics look good.

Regarding a fuel pump cutout, probably the best option is a relay in line with the feed to the fuel pump, usually located in the fuel tank (the pump that is, not the relay ). Should be easy to hide the relay. Another option is to disable the original fuel pump relay, problem with this method is that the engine ECU may pick this up as a fault, the fuel pump relay is monitored by most car's ECUs.

[retepsnikrep]

My car is now off road with it's innards scattered round my garage

Old Nimh pack out.

New Batteries are in. I'm finishing off charger install into the old empty battery case now and mains/bms wiring, hoping to get bms in tomorrow or thursday.

Perhaps have it all together working or not as the case may be by friday!!

Greg.

I had already looked at the fuel pump as that seems a good option. I'll just interrupt the feed to the pump with a relay as it's nice and handy near the rear compartment. I won't be tackling/connecting the security/speed sensor/whmile etc until basic bms is in and running correctly.

[retepsnikrep]

Well it's in and connected, no flashes or bangs and the car started. No IMA codes yet

Sadly I let the 12v aux battery go flat during the upgrade so that's now on charge. More serious testing tomorrow

Mains charger worked on command but charger control not working correctly yet

Slaves are working OK and seem to be responding correctly.

The initial oem forced charge may cause a problem we shall see

Knackered now and working nights

[ChrisB]

Yay, bet your looking forward to your first few drives, its always nerve racking connecting that last cable, you knw that it wont go bang but it always scares the willies out of me

The display looks the mutts, as does the whole install really, a credit to your hard work.

ChrisB

[retepsnikrep]

Eta for test engine start and forced recharge bcm recalibration is 10:00 GMT

I will have drained 10ah out by then down to 30ah, this should give sufficient headroom for the initial Oem forced charge which you can't avoid. The Oem battery control module uses this forced charge to give itself an initial figure for soc etc. Takes a few minutes and as cells were fully charged when I installed them they could not accept this extra charge so I had to burn off 10ah overnight to make room

Corrected a stupid error in Master Software which was source of the failure to control charger correctly. Debugging don't you love it

Might video it for posterity

[retepsnikrep]

Started OK and took the initial forced charge fine, oem Batt gauge is now showing about 1/3charge, no untoward faults and I actually drove it down the road for a couple of 100yards, standard assist and regen seem to be working fine.

My BMS is all over place at moment though as strangely one of the most inaccessible slave boards is false triggering, dumping data on the Master bus, setting off the following 5 cells as well and upseting the Master. I now have the batteries back out on bench and am making another Slave board as I suspect a dry joint or bad pic, interference, oddly it has been working fine on bench and in car so long as IMA is disabled/off. Might add some local screening as well.

Back into garage now!


Well back indoors again now, bloody cold outside in garage. I'm quite happy at the minute with the way the OEM BCM has accepted the new cells. I've deliberately triggered a few faults by disconnecting oem temp sensors etc and they flag up with IMA/check engine lights as expected. Glad I've got a Scangauge!

BMS slightly more stable now the false triggering slave has been replaced, still not right or useable at the moment. Def interference or power supply fluctuations getting into Master causing it to reset etc. I suspect the Master data bus

A few things I intend trying. So got to take battery case out of car again to do a few tweaks.

1) Add some probe points to Master bus, I made the wiring so tidy I can't get access to the master bus without pulling it out of car!

2) Try running Master on a completely independent exterior power supply to see if that corrects problem, if it does it means problem is coming in over 12v car aux supply, may then add a 12v-12v miniature dc converter to isolate Master supply from car.

3) Drill hole in Zivan charger case to get access to voltage pot, should have done this before!! Again I've got to remove the unit to get access at present.

4) Get my scope on the Master bus and power supply, pic inputs to see what's happening.

5) Reduce Master bus serial data speed to 1200baud to see if that helps to make data more error resistant. I'll try this last.

So not working properly yet but I'm not slashing my wrists either, just got some problems to overcome.

[retepsnikrep]

Well

Isolated supply improved things quite a bit, and incredibly when you see the noise still present on the Master bus it almost runs without errors!!

So problem is as Mike surmised horrible noise on Master Bus. This noise is on the paralleled opto isolated side of the digital slaves Master Bus which weaves it's way round the packs before exiting to the Master board input. I need some sort of filtering now to sort this, any ideas more than welcome.

http://www.solarvan.co.uk/bms/Noise1912.jpg

I thought about starting with a 4.7v zener diode to cut off the high voltage spikes as they could damage the Master pic or at least cause it to lock up. It had been reseting itself at will due to this horrible noise. Over to the experts out there.

Appears the Insight is even more of a noise nightmare than I thought!

The Digital slaves all seem to be working fine though even directly exposed to that noise

Some pics.

http://www.solarvan.co.uk/bms/Install1912.jpg All plugged in but not fixed finally.

http://www.solarvan.co.uk/bms/Dash1912.jpg

Dash show batteries accepted by BCM and forced charge stopped after a couple of minutes, soc now 4 bars, bat voltage 167v, interestingly I measured the voltage of the nimh one I removed which had been charged to within one bar of top and it's voltage was about 158v. No IMA errors.

http://www.solarvan.co.uk/bms/Display1912.jpg

[MindMil2]

Well

So problem is as Mike surmised horrible noise on Master Bus. This noise is on the paralleled opto isolated side of the digital slaves Master Bus which weaves it's way round the packs before exiting to the Master board input. I need some sort of filtering now to sort this, any ideas more than welcome.

Hi Peter, I suggest to use simple RC filter. The noise spikes look very high frequency compared to data signal so it should be enough to use simple 1 pole RC low-pass filter. Regarding spike limiter I suggest to utilize PIC's input clamping diodes as they are designed for this purpose and for PIC16F886 the max current clamping current is 20mA. So if you put 1k resistor on input then it will safely protect from spikes from -20V to +25V.
Then you should just pick the right capacitor to filter out the spikes. As you have oscilloscope it should be fairly easy.
So simple protection circuit would be: input->(R 1k)->(C to ground)->PIC input.

I also suggest to look for other potential noise sources coming to master board. It might be that it was not the only one. Take special care about power buses.

http://hr-ev.blogspot.com

[retepsnikrep]

Thanks for the advice. I am back on nimh power for the time being and the car is back together as I need it for work.

I need to do a fair bit more work on the bench, and I'm fairly convinced a lot of the noise is RF so I need a screened aluminium case for the Master board. I intend to revisit the whole layout and add some screened cable in appropriate places as well as some of the RC filters. I intend using an isolated power supply as well. So lot's to do but I don't want to spend all Christmas getting annoyed in the garage when I could be lounging around eating, drinking and watching TV.

What have I learnt so far?

1) Insight rear compartment is a HF noise nightmare.

2) Insight OEM BCM accepted cells without throwing any errors

3) Digital slaves work OK even in the noisy environment.

4) Master worked for short periods but disprupted/reset by all the noise, when it was working for short periods It appeared OK.

5) No I haven't lost heart or binned the project, just got to sort my noise