Go Down

### Topic: Battery Pack, individual cell voltage measurement (Read 14142 times)previous topic - next topic

#### PeterH

#30
##### Dec 03, 2012, 01:24 am

Nope, leastways I can't think of a way to use zeners.

If you just want to drop the voltage down the easiest way is to use 45 voltage dividers, all set to provide the best range possible for the voltage at that point in the battery string.

The voltage divider approach obviously loses resolution the further you need to drop the voltage - I don't know what resolution is needed here, but there must come a point where the voltage is so high that it isn't possible to achieve enough resolution.

I was thinking with the zener something like connecting each battery junction to a zener and then through a voltage divider to ground, the idea being that the zener would drop the voltage by a known constant amount to bring the voltage down to a range that can be measured, with no loss of resolution.

By the time you have several cells in series all with varying voltages the range of values that needs to be measured could easily exceed 5V hence the voltage dividers to reduce the range - still some loss of resolution due to the voltage dividers used for that, but nothing like as much as there would have been if dividers were used to drop the full voltage down to 5V.

#### sbright33

#31
##### Dec 05, 2012, 08:29 pm
Works for me!  This is the easiest solution for hardware.  Calibrating it using a DMM in the code will take some time and effort.

#32
##### Dec 06, 2012, 12:44 am
I get the zener idea now (told you I was slow), but if each voltage divider is tuned to the voltage surely there's no loss of resolution anyway.

We haven't heard from the OP since post #7 so as is often the case we're talking to ourselves, still it's good to bounce some ideas around.

______
Rob
Rob Gray aka the GRAYnomad www.robgray.com

#### jolphil

#33
##### Dec 06, 2012, 02:15 pm
Hello
Would the temperature coeff. of zener/avalanche diodes skew the accuracy ?
http://users.tpg.com.au/users/ldbutler/ZenerDiodes.htm
jolphil

#### kk4zmg

#34
##### Aug 16, 2015, 03:38 am
I realize this is old. But I'm working on a vary similar project using LifePo4 batteries.  My batteries modules are from the Fisker cars. There 24v each. I am trying to brake them up to 12v. I am planning on splitting the packs to get 2 12v batteries out of each module. I had thought about using voltage dividers but that is assuming that the other cells before the cell you would like to measures stay at the same voltage. so if a cell that is before the one you would like to measure is low that will throw off the sell that you are measuring. Switching between cells sounds like a good idea. instead of using relays could you use transistors? How would you write the code for that?? Also I would like to have a serial print so it would need to tell me what the cell voltages are. I guess you can do that with a delay of some kind.   I have a basic sketch made up for 4 cells using the analog inputs that will trigger a digital output if any of the 4 cells goes below a certain threshold voltage. That would trigger an external alarm/relay to pull that module out of the group.    I have 4 of the 24v modules that i was going to do this mod to and parallel them in my battery bank. Each module will be getting a monitor.   I was also planning on having individual cell balancing driven by a comparator to trigger a mosfet to a resistor bank. That would keep the cells from over charging in the event one or two cells have not fully charged.

There are a bunch of good ideas here.

Hope this can revive some of them.

Thanks Rich.

#### jack wp

#35
##### Aug 16, 2015, 04:05 am
"Switching between cells sounds like a good idea. instead of using relays could you use transistors?"

I lean toward mechanical relays for this ( double pole). Transistors have voltage drop, relays do not.

Each relay can switch in a cell ( positive, and ground), to be tested, using only one analog pin.

#### kk4zmg

#36
##### Aug 16, 2015, 04:36 am

I thought it would be something like that. ( voltage drop across transistors) I have started writing a new sketch. I'm getting a little lost on how to have it print the reading of individual cells and hold it till it reads that cell again. I will post a copy of what I have so far. Maby you can tell me if I'm on the right track.

Thanks again.
Rich.

Code: [Select]
`int safty = 13;int check = 12;int threshold = 2.5;int Cellv = A0;int CellVal = 0;float Cellpin = 0;float CellVoltage = 0;int gnd = 2;int Cell1 = 3;int Cell2 = 4;int Cell3 = 5;int Cell4 = 6;  void setup(){    Serial.begin(9600);    pinMode(safty, OUTPUT);pinMode(check, INPUT);pinMode(Cell1, OUTPUT);pinMode(Cell2, OUTPUT);pinMode(Cell3, OUTPUT);pinMode(Cell4, OUTPUT);}     void loop(){        int checkVal = digitalRead(check);    int sampleCellVal = 0;    int avgCellVal = 0;        for (int x = 0; x< 10; x++){            CellVal = analogRead (Cellv);      sampleCellVal = sampleCellVal + CellVal;            digitalWrite(Cell1, HIGH); delay(0.0);      digitalWrite(Cell1, LOW); delay(0.5);      digitalWrite(Cell2, HIGH); delay(1.0);      digitalWrite(Cell2, LOW); delay(1.5);      digitalWrite(Cell3, HIGH); delay(2.0);      digitalWrite(Cell3, LOW); delay(2.5);      digitalWrite(Cell4, HIGH); delay(3.0);                        if (CellVoltage < threshold) digitalWrite(safty, HIGH);      else{ digitalWrite(safty,LOW);          }delay (10);    }        avgCellVal = sampleCellVal /10;    Cellpin = avgCellVal * 0.00610;        CellVoltage = Cellpin;                delay(10);   }`

I know I said I was going to have 2 groups of 4 cells. But for trial I'm only going to do one group. I can add the other group after proof of concept. I'm also doing .5 seconds of read per cell. I'm not really shore if this is enough time to get an accurate reading?

#### jack wp

#37
##### Aug 16, 2015, 03:57 pm
I believe you meant to say, you decided to use mechanical relays to connect each cell for analog reading on just one analog pin. Right?

I think you will want to use a couple arrays, since you will have several cells to test.

You will need a relay for each cell, and be able to control each relay. You may not have enough output pins to control that many relays directly, and may need some sort of multiplexer chip. How many cells do you think there will be?

#### kk4zmg

#38
##### Aug 16, 2015, 04:40 pm
Yes I am going the rout of the relay. I'm only going to be doing 8 cells per module. So then I would only need 8 relays, 8 outputs for the relays. But I'm also using an output for a safty cutout and and input to check the safty cut out. So that would be a total of 10 digital I/O. I put in some comments I hope this will help. I'm still trying to figure out or to serial print the different voltage reading of the different cells.

Code: [Select]
`int safty = 13; // An output to Cut out the moduel if any cell is under a specified voltage.//int check = 12; // An input to check the output of safy.//int threshold = 2.5; // The spesified voltage of the safty cutout.//int Cellv = A0; //The analog input to meshure voltages of the cells.//int CellVal = 0;float Cellpin = 0;float CellVoltage = 0;int Cell1 = 3; // Output to triger relay 1 for cell one to read analog voltage.//int Cell2 = 4; // Output to triger relay 2 for cell 2 to read analog voltage.//int Cell3 = 5; // Output to triger relay 3 for cell 3 to read analog voltage.//int Cell4 = 6; // Output to triger relay 4 for cell 4 to read analog voltage.//  void setup(){    Serial.begin(9600);    pinMode(safty, OUTPUT);pinMode(check, INPUT);pinMode(Cell1, OUTPUT);pinMode(Cell2, OUTPUT);pinMode(Cell3, OUTPUT);pinMode(Cell4, OUTPUT);}     void loop(){        int checkVal = digitalRead(check);    int sampleCellVal = 0;    int avgCellVal = 0;        for (int x = 0; x< 10; x++){            CellVal = analogRead (Cellv);      sampleCellVal = sampleCellVal + CellVal;            digitalWrite(Cell1, HIGH); delay(0.0); // delays to change between cells of 5 mills.//      digitalWrite(Cell1, LOW); delay(5.0);      digitalWrite(Cell2, HIGH); delay(10.0);      digitalWrite(Cell2, LOW); delay(15.0);      digitalWrite(Cell3, HIGH); delay(20.0);      digitalWrite(Cell3, LOW); delay(25.0);      digitalWrite(Cell4, HIGH); delay(30.0);                        if (CellVoltage < threshold) digitalWrite(safty, LOW);      else{ digitalWrite(safty,HIGH);      // if statment so if the voltage goes below threshold it trigers the output of the safty pin.//    }delay (10);    }        avgCellVal = sampleCellVal /10; // calibration of the anilog voltage pin.//    Cellpin = avgCellVal * 0.00480;        CellVoltage = Cellpin;          Serial.print("  B1C1 = "); delay (0.0); // delay is an attempt at distiguishing between the difrent cells as it parces them.//Serial.print(CellVoltage);Serial.print("  B1C2 = "); delay (10.0);Serial.print(CellVoltage);Serial.print("  B1C3 = "); delay (20.0);Serial.print(CellVoltage);Serial.print("  B1C4 = "); delay (30.0);Serial.print(CellVoltage);Serial.print("  safty = ");Serial.println(checkVal);    delay(1000);   }      `

#### jack wp

#39
##### Aug 16, 2015, 05:12 pmLast Edit: Aug 16, 2015, 05:50 pm by jack wp
Code: [Select]
`      digitalWrite(Cell1, HIGH); delay(0.0); // delays to change between cells of 5 mills.//      digitalWrite(Cell1, LOW); delay(5.0);      digitalWrite(Cell2, HIGH); delay(10.0);      digitalWrite(Cell2, LOW); delay(15.0);`
May need to look more like:
Code: [Select]
`     digitalWrite(Cell1, HIGH); delay(5); // delays to change between cells of 5 mills.//       cellOneVolts=analogRead(cellV);      digitalWrite(Cell1, LOW); delay(5);      digitalWrite(Cell2, HIGH); delay(5);       cellTwoVolts=analogRead(cellV);      digitalWrite(Cell2, LOW); delay(5);`

Once you have all the cell***Volts, then you can display them to the serial monitor. one line at a time, or maybe all 8 on the same line with time (depending on how long of a line your display will take).

Side note: I am not confident that the relays will react in 5 ms, so I would suggest something like 50 ms. Also, just in case the relays are slow, put a ~50 ohm resister on both legs of all the relays. We don't want to short out cells!

#### sonnyyu

#40
##### Aug 16, 2015, 07:18 pmLast Edit: Aug 16, 2015, 07:20 pm by sonnyyu
Monitor IC for Big Battery Pack, individual cell voltage measurement

https://forum.arduino.cc/index.php?topic=340066.msg2344458#msg2344458

#### kk4zmg

#41
##### Aug 16, 2015, 07:42 pm
Jack wp.

cool thanks for you help. I will try and implement your suggestions as soon as I can fix my computer. I went for breakfast when i came back it was crashed have been unable to recover so far.

Sonnyyu,

Thanks that looks good. I will look into it. I'm still vary new on all this stuff. So trying to get the serial link communication working will take some time, and learning. But I think after the proto of the first module I will be going more toward that ic. or witch ever is more easer to implement.

thanks for all you guys help.

#### MasterCATZ

#42
##### Apr 01, 2016, 01:34 am
good idea's

I am needing to monitor 24 x 2v cells

1.8 - 2.3 V range

now I just have to decide if I break it down into 4x separate banks to monitor

or use an ATtiny processor per cell

#### BigBobby

#43
##### Apr 01, 2016, 03:28 am
Wow, this is an old thread.

To the most recent poster that resurrected it, are you monitoring Lead Acid cells?  If so, that's wicked rare.

#### Wawa

#44
##### Apr 01, 2016, 03:30 am
Not a good idea to revive an old thread.