Old BMS Hardware Thread

[hohisilver]

Hmm ok. Can you point me at a description of the Meanwell current limit mods?

[GregsGarage]

When the charging stops with the highest cell at 3.65, what is the voltage of the lowest cell while still on charge? If it's got to 3.5 volts, that is probably good enough, 3.45 is marginal and below that you need more balancing. With Peter's pbp code the load cut in voltages can be adjusted while charging. I have been dropping it down to 3.45 to start loading a few of the highest cells. Then I increase either when the highest cell reaches 3.50 or nearly all cells are at 3.45. This requires monitoring the charge process, but this should be done with a new pack anyway.

Those look like headway cells, which ones are they and what is your pack configuration? I have been interested lately by the Headway 40160 16ah cells. Even though they are cylindrical, they take up less volume and are lighter than the same capacity prismatic cells.

[GregsGarage]

Finally decided to bite the bullet and take a test run this evening - first thing I'm relieved about, no noise problems (although I am using shielded microphone cable for all the serial connections)

I went a bought some fancy shielded twin pair cable and the shielding was causing some of my problems. I am getting the best results with the shields unconnected. Also the length of the wires seems to have been another problem, which I don't think you will have in the scooter.

[hohisilver]

I can't remember what the lowest cell was, it was certainly 3.35 or 3.40. I'm going to see if I can get a better balance after a few cycles - using charger cutback could cause me major headaches as I hadn't really catered for it I've got the charger cutoff set at a pack voltage of 87.50 and it was already something over 86V when it cut out, so it wouldn't have got much further anyway.

Yes, they're Headway 38140S configured as 25S 2P to give me nominally 80V/24AH.

I went a bought some fancy shielded twin pair cable and the shielding was causing some of my problems. I am getting the best results with the shields unconnected. Also the length of the wires seems to have been another problem, which I don't think you will have in the scooter.

True - I've got a master-slave cable run of less than half a metre

[hohisilver]

Peter:

Hmm ok. Can you point me at a description of the Meanwell current limit mods?

I think I've figured out a way I can implement charger cutback with very little disruption to my existing setup - am I right in thinking if I can find a suitable fixed-value resistor I can switch that in in parallel with SVR2 in the Meanwells? Or would that be in series?

[retepsnikrep]

The meanwell mods are on the endless spehere forum.

I use two muliturn variable resistors in parallel so I can adjust the bulk and balancing current. I removed the svr2 and brought the two connections to the outsdie of the master charger in my setup.

A micro relay switches the lower value resistor into parallel with the higher value one if the opto is not conducting.
So my default is low curent charging when relay is not active incase BMS fails. So in effect you get a low value resistance by default.

I turn my opto/relay on for bulk charging which selects just the higher value resistor adjusted to give whatever current I want. You can adjust both to set the current you want.

I did add a command to the master to allow you to force the BMS into the balancing phase so you could adjust the low current setting. Not sure if it is in latest releassed version but you can add it to your code i'm sure.

[hohisilver]

Thanks

Thinking about it, it makes a lot more sense to bring the current limit adjustment out of the PSU, at the moment I don't have the lids screwed on for current adjustment purposes. I'll probably do something similar to what you've done, then at least I can screw the lids back on!

Of course this is all on the back-burner at the moment because testing the scooter last week my controller blew a FET

[hohisilver]

Finally got around to doing the cutback mods, tested them this evening. However, I can't get the lower value to limit the current to anything less than 1.3A, and then the voltage on each PSU drops to about 15V. Is there another fixed resistor internally that I needed to change? The instructions I followed said to jumper across R37 so that's what I did.

Anyone any ideas?

[hohisilver]

Just as an update on this, the reason the voltage dropped so low is because I was using dummy loads. The voltage stays acceptable (i.e. just above the pack voltage) when charging as you'd expect, but the current still won't drop below 1.3A. However that seems to be perfectly good for my purposes as after a series of charges over the last week or so, gradually increasing the load cutin voltages I have now achieved a balance to within 20mV

[retepsnikrep]

So long as it works that's fine. The resolution of the BMS at best is only around ~20mv