Old BMS General Thread

Threads relating to the BMS system begun by Peter Perkins

Moderators: GregsGarage, retepsnikrep

User avatar
retepsnikrep
Posts: 1387
Joined: Sat May 26, 2007 4:50 pm
Location: North Yorkshire England
Contact:

Postby retepsnikrep » Mon Nov 17, 2008 12:35 am

Mike

Be my guest and start a FAQ, Greg can you help/contribute as well?

I started a BMS project write up a few weeks ago, but again time pressure and work has put that on hold until it's actually up and running.

Mike I'm a little confused by your post saying you had removed all the opto's :?

Digital Slaves also need the digital software and vice versa.

I assume you are actually going to use digital slaves with the master is that correct? Have you got yourself a little screen etc, any pics you can post of your progress?
Regards Peter

Two MK1 Honda Insight's. One running 20ah A123 Lithium pack. One 8ah BetterBattery Nimh pack.
One HCH1 Civic Hybrid running 60ah A123 Lithium pack.

mikep_95133
Posts: 28
Joined: Thu Sep 25, 2008 6:38 pm

Postby mikep_95133 » Mon Nov 17, 2008 3:41 am

I only removed the opto's so that I could run the slaves as analog until I get the wiring to the cabin done. At that point the slaves get digital software and the master added.

I have the 3.5" lcd from Ebay. I'm doing fine, just a bit distracted with the 2nd EV purchase and it's repairs. Here are some bench pics.

http://rotordesign.com/s10/lithium/tsbms.jpg
http://rotordesign.com/s10/lithium/tsbms2.JPG

I'm sure we can at least start a faq.

Mike

GregsGarage
Posts: 870
Joined: Tue Apr 01, 2008 5:27 pm
Location: Galashiels, Scottish Borders
Contact:

Connecting slaves in analogue mode

Postby GregsGarage » Mon Nov 17, 2008 11:35 am

Hi Mike,

You still need to use 2 optos, U2 & U4, per board when using the slaves in analogue mode, All the U2 optos will be connected in parallel for the Max cell voltage cut-out. Think of this as a logical OR. The optos are required because the cells are at a different potential to each other. U4 provides the Min cell voltage cut-out, also connected the same. There is also a resistor that is moved as well, refer to the circuit diagram for the Analogue slaves, it is different to the digital slaves.

My slaves are connected to a control board that turns on and off the charger and controller as required. The low voltage cutout is set about 400mv above the lowest allowable voltage for the cells, so 2.4 volts for the 2.0 volt limit the manufacture specifies. My EV Managment Module controlls the charger through a mains relay. Whjile driving, when a low voltage signal is detected illuminates the battery light on the dash. If after 5 seconds the warning is still present it shuts off the controller. This allows you to back off the throttle to put out the warning. In practice I have been able to get the warning to come on with the batteries 70% discharged, at full throttle, climbing a long hill. Backed off the throttle and warning went out, without having to lose much speed, I might add. The deepest discharge I have done is 88%. Because the batteries were getting low I took it a little easier and never had a low voltage warning.
Greg Fordyce

Daewoo Matiz
http://www.evalbum.com/4191

GregsGarage
Posts: 870
Joined: Tue Apr 01, 2008 5:27 pm
Location: Galashiels, Scottish Borders
Contact:

V1.14 test

Postby GregsGarage » Mon Nov 17, 2008 1:57 pm

Peter,

I have had a quick test of version 1.14. Unfortunately it doesn't seem to work reliably even at 250khz although at 4mhz it works fine. Not sure why this is but for now 1.13 looks to be the best option at 125khz. The problem seemed to be with the load not always switching on. May be some room for tinkering. Some very quick measurments at around 3 volts showed a load of around 240ua at 125khz and around 250ua at 250khz and 125ua for 31khz. So if we can get it to work, that would save quite a lot of ua, but I may just leave mine set to 250khz.

Regarding a FAQ/Wiki, that is a great idea. Maybe we could get it on Sourceforge, as just one idea. We can discuss this more when you are back from your hols. If anyone else has suggestions on setting up a wiki/faq, please chime in. :wink:

Edit: I think I might try getting rid of the hysterias in the load and min voltage. That would simplify the program even further and may help things. These 2 values could then be controlled with a if-then-else statement instead of 2 if-then statements.
Greg Fordyce

Daewoo Matiz
http://www.evalbum.com/4191

User avatar
retepsnikrep
Posts: 1387
Joined: Sat May 26, 2007 4:50 pm
Location: North Yorkshire England
Contact:

Postby retepsnikrep » Mon Nov 17, 2008 2:17 pm

Greg

The 125ua test, was that with the 100k resistor for the lm385 load?

Try this version, very slightly different but still set to operate at 31khz.

Peter

www.solarvan.co.uk/bms/SlaveAnalogueOnO ... _v1_15.txt

Not sure about getting rid of the hysterisis :? That will cause the load to switch on and off rapidly at the change over point.

Once triggered an over/under V condition needs to be visible/available to user/system for a certain amount of time or it might not be noticed. The little bit of hysterisis we have helps with that.
Regards Peter

Two MK1 Honda Insight's. One running 20ah A123 Lithium pack. One 8ah BetterBattery Nimh pack.
One HCH1 Civic Hybrid running 60ah A123 Lithium pack.

mikep_95133
Posts: 28
Joined: Thu Sep 25, 2008 6:38 pm

Re: Connecting slaves in analogue mode

Postby mikep_95133 » Mon Nov 17, 2008 3:06 pm

GregsGarage wrote:Hi Mike,

You still need to use 2 optos, U2 & U4, per board when using the slaves in analogue mode, All the U2 optos will be connected in parallel for the Max cell voltage cut-out. Think of this as a logical OR. The optos are required because the cells are at a different potential to each other. U4 provides the Min cell voltage cut-out, also connected the same. There is also a resistor that is moved as well, refer to the circuit diagram for the Analogue slaves, it is different to the digital slaves.


Understood. There is no charger or controller connections for 2 reasons. No cable to either and neither have the capability for external input. This still has to be hacked onto each. So for now the analog slaves just keep themselves balanced.

Mike

User avatar
retepsnikrep
Posts: 1387
Joined: Sat May 26, 2007 4:50 pm
Location: North Yorkshire England
Contact:

Postby retepsnikrep » Mon Nov 17, 2008 3:22 pm

Mike I would still wire the analogue slaves up as described by Greg and simply implement a warning light/buzzer for an over or under V condition.

At least then you know if a cell in the string is out of bounds and can react accordingly.

If you don't do at least this a cell could go low or high without you knowing :shock:
Regards Peter

Two MK1 Honda Insight's. One running 20ah A123 Lithium pack. One 8ah BetterBattery Nimh pack.
One HCH1 Civic Hybrid running 60ah A123 Lithium pack.

mikep_95133
Posts: 28
Joined: Thu Sep 25, 2008 6:38 pm

Postby mikep_95133 » Mon Nov 17, 2008 3:33 pm

Understood. The cells were installed down about 10ah on a 40ah cell. So I run these cells in a safe range. Every two weeks I take them out, recharge them to 4.0v each to see how many ah's they are below 40ah. Then load test them for capacity, then recharge and discharge them again to -10ah. Short drives never take the cells down below -30ah. So far 400+ miles on these cells.

I understand the risk, but without a data cable of any kind leaving the pack I cannot even use an alarm. This battery box is built like a fortress. It's not going to be easy to even add a cable. Right now I just wanted to see if the cells can handle being on the road at all. So far so good.

The battery box gets road spray and rain water accumulating a bit on the top. So the lid has to remain well sealed or the drive system gives me a ground fault and the vehicle stops running at bad times. The vents are about the only easy route for a cable, but the vent won't pass a connector. In the end I have to find a place to drill a hole in my precious box.

Mike

GregsGarage
Posts: 870
Joined: Tue Apr 01, 2008 5:27 pm
Location: Galashiels, Scottish Borders
Contact:

Postby GregsGarage » Mon Nov 17, 2008 4:09 pm

retepsnikrep wrote:Greg

The 125ua test, was that with the 100k resistor for the lm385 load?


Yes that is with the 100k resistor. My "good" meter only goes down to ma, so I am using a cheap chinesse meter for the ua readings.

Not sure about getting rid of the hysteresis :? That will cause the load to switch on and off rapidly at the change over point.

Once triggered an over/under V condition needs to be visible/available to user/system for a certain amount of time or it might not be noticed. The little bit of hysterisis we have helps with that.


Ah, but we are slowing the loop down to around one loop per second at 31khz, so I think the hysteresis is a bit of overkill. It can only switch between on and off at the rate of the program loop and no faster. So even if it triggers for only 1 loop that is still around a second.
Greg Fordyce

Daewoo Matiz
http://www.evalbum.com/4191

GregsGarage
Posts: 870
Joined: Tue Apr 01, 2008 5:27 pm
Location: Galashiels, Scottish Borders
Contact:

Postby GregsGarage » Mon Nov 17, 2008 4:20 pm

Mike, you only need 4 small wires coming out of the battery box, and you may actually get away with three if you share a common for both sets of optos. Parallel the slave boards inside the battery box. That is what I have done. I would recommend that you connect a relay to the high opto out to switch off your charger. What will happen during balancing is that the charger will cycle on and off while the load bleeds off the high cells. Remember that the loads can only drain 250ma, so unless your final charger current is set this low you will be overcharging some cells (and shortening their life). Put an LED on the dash for the undercharge warning at the very least.

3 wires should be easy to sort in your battery box, a small cordless drill should do it in a minute.:shock:

mikep_95133 wrote:Understood. The cells were installed down about 10ah on a 40ah cell. So I run these cells in a safe range. Every two weeks I take them out, recharge them to 4.0v each to see how many ah's they are below 40ah. Then load test them for capacity, then recharge and discharge them again to -10ah. Short drives never take the cells down below -30ah. So far 400+ miles on these cells.

I understand the risk, but without a data cable of any kind leaving the pack I cannot even use an alarm. This battery box is built like a fortress. It's not going to be easy to even add a cable. Right now I just wanted to see if the cells can handle being on the road at all. So far so good.

The battery box gets road spray and rain water accumulating a bit on the top. So the lid has to remain well sealed or the drive system gives me a ground fault and the vehicle stops running at bad times. The vents are about the only easy route for a cable, but the vent won't pass a connector. In the end I have to find a place to drill a hole in my precious box.

Mike
Greg Fordyce

Daewoo Matiz
http://www.evalbum.com/4191


Return to “BMS thread”

Who is online

Users browsing this forum: No registered users and 25 guests