Old BMS General Thread

[retepsnikrep]

Bill

Pause 1000 waits for 1000ms at 4mhz (1 second)

Pause 1000 waits for 500ms at 8mhz (0.5 seconds)

Default Master speed is 8mhz

You are right in saying

Delay = 10 says to delay 10 milliseconds. Does Delay 10 really delay only 5 milliseconds at 8Mhz?

Yes it's 5ms at 8mhz. sorry my error. It's nice to get these little bugs sorted though, and I just found a damn annoying one in the Master.

As an aside I am sitting here worrying in my meal break at work as I left my test set up running unattended for the first time today charging my 5 cells in the care/control of the Master I hope it doesn't let me down

It's got to be done at some point and the psu is independently set to CV at 18v (3.6v) per cell anyway, so that should prevent any major disaster if the Master fails to cut the charge when the first cell reaches the cut off point. I remain reasonably confident although nervous until i can check at 22:00 tonight

[retepsnikrep]

Well it didn't melt down, and had turned itself off by the time I got home at 22:00 so that's good news. I'll check it all over today.

I'm also assembling and testing the two battery blocks at last

Battery trays arrived today, two aluminium trays which fit into wheel well. They have holes along the right and left hand edges with captive nuts and are bolted through the sides of the wheel wheel from the outside. The cells are strapped to the trays 25 per block in lines of five cells with two 300kg nylon banding straps per 5 cells, that's pretty secure

Trays on bench.

http://www.solarvan.co.uk/insight/TraysBlocks01.jpg

Trays in wheel well.

http://www.solarvan.co.uk/insight/TraysBlocks02.jpg

First block of strapped cells.

http://www.solarvan.co.uk/insight/TraysBlocks03.jpg

Roger Crier another UK Insight owner and friend of mine has made some neat vacum formed clear lexane covers for the cell blocks so everything on top of the cells is well protected and safe

http://www.solarvan.co.uk/insight/TraysBlocks03.jpg

Got to install all the connections and BMS boards now, bench test it all and then drop it into Insight.

I want a plug and socket install so I'm using high current dc connectors for the battery to ima compartment connections. I'm using pc type multi pin d connectors for all the BCM voltage and temp feeds. I want a neat install with 6 bolts and two connectors per block. I can lift one block on my own but it's easier with two.

http://www.solarvan.co.uk/insight/TraysBlocks04.jpg

[GregsGarage]

Just got around to trying V1.08 and V1.07 but they don't work at all. I suspect a simple programming error as it reports 3 bytes downloaded to the pic!

Anyway I don't see anything obvious, will have a look at this later.

[retepsnikrep]

Houston We have a problem!

The software looks OK to me.

Is the editor set for 08M mode?

Check the program syntax prior to downloading it, does it pass the syntax check and report the program size correctly?

Programming lead on the right way round? If you get it wrong even for a short time I don't know the result but could bugger the pic.

Sometimes you need to initiate the program cycle then connect the power jumper on the slave.

Made any changes to the host PC? Serial ports?

If you still can't do it I'll try to load the prgrams on one of my slaves, If it works for me then it's def at your end.

Try one of the versions that worked before 1.04? , if you can't download that you have a problem with the host pc or the lead etc.

RU using the upgraded editor software? Perhaps it has a bug?

Peter

[GregsGarage]

I can download the earlier version o.k. and it works fine, so no problem with pic or download cable. I am still using the earlier version of the editor software.

I can't see any problem with the software either. I am going to use the earlier version for now, maybe make V1_06 with my suggested changes.

Edit: Might have forgoten to compile it. Will try that first.

[GregsGarage]

Programming lead on the right way round? If you get it wrong even for a short time I don't know the result but could bugger the pic.

I've tested this, and no harm has come of the pic. Also no harm connecting the board reverse polarity to the cell!:shock:

Sometimes you need to initiate the program cycle then connect the power jumper on the slave.

This step is required, I think because the slave is not running at the standard clock speed.

[retepsnikrep]

I've tested this, and no harm has come of the pic. Also no harm connecting the board reverse polarity to the cell!:shock:

Arrghh Was this deliberate testing ??

I'm surprised it stood that I'm not going to test it!!

[GregsGarage]

No it wasn't deliberate testing, but no smoke let out. Won't try it again though.

I have got V1.08 to work by copying and pasting the main program, without preceding remarks, into a new editor window and it works. Reports 67 bytes. It was only reporting 3 bytes before when doing a syntax check, maybe it didn't recognize the #ENDREM statement. I have bumped the speed to 250khz, although I think you could get away with 125khz.

I will be wanting one of your master pcb boards since I didn't get so lucky with my e-meter. In a rush to get things connected I wired the meter in before connecting it at the pack. Unfortunately the power supply lead touched the 72volt terminal and that seems to be the end of the e-meter and prescaler! Hopefully lesson learned, cover your pack and terminals with something like cardboard while working on it. You can move pieces around while making connections, but hopefully give some protection if you drop something.

Two things that I will have to change, one is the current sensor. I will need a 500 amp and adjust the software. The other is that it will have to run on 24 volts, I think you said max 20 volts. Is this just a mater of changing the regulator chip, or will it be more involved?

[dillond666]

Greg,

For the master board to run on 24 volts you can make up a simple regulated power supply using an LM7812, a few capacitors and a diode. This goes between the 24v supply and the 12v connector on the master board. The onboard voltage regulator is for the picaxe and other 5v stuff.
If your slave is still alive you are very lucky

Derek

[retepsnikrep]

Derek/Greg. The L7805 I used on the master is just a bog standard series regulator and can accept any voltage upto about 35V ish at the input. It may need a very small heatsink if you use a higher input voltage as you are burning off a bit more power.

However the power input also feeds the relay driver transistor at the input voltage, so it means the charger cut off relay output J7 would be the same as your input voltage. The BC337 can stand about 45v so that should be OK at 24v.

Why do you need to use 24v, can't the Master be supplied from the normal car 12v aux?

It's very interesting about the need to delete the REM stuff from V1.08, I'll fire it up in my editor tomorrow and try and find the issue.

Bad news about the e-meter

Greg, Derek has jut piped you for my 500A current sensor Sorry.

The Master current sensor input accepts a 0-5v output device.
It must not be allowed to go negative or over +5v.

I used a Hal 400S device in my previous vehicles but there may be something more suitable now. Are you using regen?

I do have a circuit to condition the output from a -5v - 0 - +5v type sensor so that it provides an ouput in line with the current master/software, i.e. 0-2.5-5v, with 2.5v being 0A or zero. 0-2.5v is discharge, 2.5v-5v is charge.

This gives a resolution of about 1A on a 500A charge/discharge setup.

Peter