BQ78350-R1A: End Of Line Testing for Protection Circuit Fucntion

Please clarify exactly what you are doing and measuring with numbers. I'm not clear what you are doing.

Are you saying that you connect all cells except the once between VC1 and VC0, which you replace with a power supply but when you vary the voltage VC1-VC0, the gauge reports a change in the other cell voltages? If so, please show what the gauge reports together with all voltage measurements VC10-VC0 vs. VSS.

Hi thanks for the reply, I'm trying to test that the protection will actually trip once we have built up the pack, so that we can confirm the safety of it before sending it out. I'm trying to confirm that when the voltage of one cells crosses that CUV or COV threshold that it will actually trigger the protection and the fets will turn off. 

What we had been testing was replacing all cells with a power supply as that lets us manipulate the voltage freely to trigger what we need to, but this doesn't do anything to prove that the board did not get messed up in some way during production because you cannot test in that manner once cells are connected unless you actually discharge or charge the cells past those points, that takes much too long and it is not good for the cells themselves. 

I had hoped that applying voltage on the chip side of the sense resistor would have allowed us to fake a cell voltage from the chips perspective using a power supply, but found that caused the chip to read a change in voltage at the adjacent cells. 

So, all cells are connected and ~3.58v per cell, we connect a multi quadrant PSU to VC1 and VC0 pins and set the PSU to 4.2v. We then read from the chip 4.2v at cell 1, but cell 2 then drops to ~3v. If we then disable the power supply both voltages return to ~3.58v. If we then move that PSU connection to VC2 and VC1 and apply 4.2v then we see cell 2 read 4.2, but cell 1 and 3 read ~3v. Then if we switch to testing undervoltage and now apply 2.3v (our cell undervoltage threshold) the see cell 2 read 2.3v, but cells 1 and 3 read to ~4.3v which causes the over voltage to trip making it unable to test functionality of the undervoltage. 

Why would the adjacent cells move opposite the test cell? How would I go about testing this to confirm function on a completed pack instead of a bare board. 

Hello Tim,

Tim Hennesy said:

I'm trying to confirm that when the voltage of one cells crosses that CUV or COV threshold that it will actually trigger the protection and the fets will turn off. 

Yes, this should happen.

For evaluation of the device's performance, I would suggest to use a resistor ladder to simulate cell voltages, similar to what we have in the EVM for this device. Have you attempted this before?

The part will draw a little bit of current when taking cell measurements, this will be small amounts of current, but if there's a large enough impedance on the cell-side, it may show up as error on the measurement.

Cell 2 voltage may drop if the voltage on Cell 1 drops too much, that is the case I can think of as to why the voltage on Cell 2 may be dropping. This is because there is a minimum voltage requirement for Cell2 to work correctly (With respect to Vss).

Best Regards,

Luis Hernandez Salomon

Hi Luis, so we had been testing the boards prior to assembly into a pack by using a resistor bridge for simulating the 8 cells then 1 psu that we move the connection to each cell position. The nodes of the resistor bridge were connected to the cell taps, so VBAT-, BT1, BT2, BT3... in the pics i sent. It works perfectly fine for a precheck and we can manipulate each fake resistor cell voltage without affecting any other fake resistor cell. But what I had put as the issue with that on the earlier posts, that can only function as a precheck on the board, but does not work once the pack is assembled with cells because unlike resistor actual cells can source and sink current. That is what we are trying to do so that we can actually confirm on the built pack nothing happened during assembly that damaged the pcb in any way that would compromise safety.

In regard to your second and third paragraphs. we are seeing not just drops in voltage but increases in voltage when trying to connect to the pins of the chip VC0, VC1, VC2, VC3... in the schematics I sent previously and we never saw a dip or increase in voltage using the resistor ladder when we had that connected to the cell taps BT1, BT2... those would all remain steady, we used 100ohm resistors for that. 

Are you suggesting that on a pack with the cells connected to connect a resistor ladder to the IC pins VC0, VC1,V2... and use a resistor ladder on those to set their voltages, then similar to how we were doing so at the cell taps raise or lower the voltage of the resistor on the resistor bridge? That way each of the VCx pins still has a set voltage not relying on getting that through the 1k current limit resistors going to each BTx tap and then we can vary just the one cell without affecting the neighbors? I have not yet tried that but that is something I could set up a test for. 

Hello Tim,

I am a bit confused as to what is it that you are trying to achieve? What is the goal of trying to manipulate a single cell after attaching the battery packs?

I am not entirely sure what may be going on in your case. Personally when testing a single cell when compared to other cells, I just use a single PSU on the input I want to change and it all works okay. I have never tested using both real cells and a PSU however.

Best Regards,

Luis Hernandez Salomon