Operational Amplifier for BMS

[rascupanamuha]

I have tried measuring 4 cells (0-4.2V-8.4V-12.6V-16.8V) with LM321 plugged to cca 12V and all outputs were exactly correct.
Is that applicable to 10 cell BMS if i put LM321 to 7.cell (29.4Vmax), and i want to know the reason why is that good?

You see the problem, LM321's max voltage is 32V and my 10 cell system would have 42V max (and i need to know each cells voltage)

Thanks in advance

 

[audioguru]

Many opamps have a 44V maximum allowed voltage. I have never seen a tiny LM321. Try a TLE2141 but it is not tiny.

 

[simonbramble]

you need a dedicated chip. Analog Devices make one - LTC6804 or LTC6811. You can then stack as many batteries in series as you want and talk to them all over the SPI bus. It will make your life so so much easier
https://www.analog.com/media/en/technical-documentation/data-sheets/680412fc.pdf
https://www.analog.com/media/en/technical-documentation/data-sheets/LTC6811-1-6811-2.pdf

 

[rascupanamuha]

Thanks but i don't want these LTC's. Also, I want that tiny SOT23-5 amplifiers because it seems to work.
Maybe i didn't explained my "problem" very well: problem is that I have wired that on a breadboard and I can measure voltage from let's say 11-14V even though i connected VCC (LM321) to lower than that 14V, it was 9V.
I want to know is it normal that i could sense higher voltages when VCC is at cca 29.4Vmax?

 

[kubeek]

How accurate do you need this to be? The accuracy of the divider resistors will be an issue, since the difference in their reatios is multiplied by the common mode voltage, so for 1% accuracy you will likely need 0.1% resistors on the inputs.
You have a 1/2 divider on the input, so you should be able to sense up to twice the supply voltage of the opamp.

And I have no clue what the optorelays are meant to do, could you explain?

 

[audioguru]

The datasheet of an LM321 shows that with a 9.0V supply and no load, the output low is about 0.06V and the output high is about 7.9V. How can the output get to 11V-14V??

All the rows of contacts, capacitance between them and wires all over the place on a breadboard are antennas that pickup all kinds of interference that you are measuring.

 

[rascupanamuha]

How accurate do you need this to be? The accuracy of the divider resistors will be an issue, since the difference in their reatios is multiplied by the common mode voltage, so for 1% accuracy you will likely need 0.1% resistors on the inputs.
You have a 1/2 divider on the input, so you should be able to sense up to twice the supply voltage of the opamp.

And I have no clue what the optorelays are meant to do, could you explain?

I have done this before and i corrected it with some correction factors for each cell so that is okay
When you turn on optorelay, you short cell with resistor and you discharge that cell that way.

The datasheet of an LM321 shows that with a 9.0V supply and no load, the output low is about 0.06V and the output high is about 7.9V. How can the output get to 11V-14V??

All the rows of contacts, capacitance between them and wires all over the place on a breadboard are antennas that pickup all kinds of interference that you are measuring.

I am not sure about what i got, thats why i came to here to search for the answer. And yep, i found my latest soldered version of BMS which is like on shematics and got the sam results

 

[rjenkinsgb]

I want to know is it normal that i could sense higher voltages when VCC is at cca 29.4Vmax?

Now you have posted the circuit, it shows you are using a differential amplifier configuration.
Wired like that, the voltage at the opamp inputs is half the cell voltage.

As long as he opamp supply is somewhat above half the maximum cell voltage, it will be fine.

[Edit- correct grammar..]

 

[simonbramble]

As rjenkinsgb says, as long as the voltage at the input pins of the IC is less than the voltage at the supply pins, you should be OK. Look at the non inverting input. The resistors from the battery to the input form a 'divide by 2' resistive divider, so VIN/2 appears as the non inverting input. Therefore if your power supply is greater than 15V, you should (theoretically) be OK. Now, the LM321 is an atrocious op amp so its input can only detect signals up to 1.5V below the supply voltage. Higher than that and the op amp will not see the input signal. Therefore for a 15V supply, the op amp will not see voltages greater than 13.5V, so your supply voltage needs to be greater than (29.4/2) + 1.5 = 16.2V

 

[rascupanamuha]

Thank you guys for your help. I have designed my BMS board with LM321 and send it to JLCPCB.
Lets hope it will work