Author Topic: NMOS low side switch for battery protection with TI bq77PL157A4225  (Read 1714 times)

0 Members and 1 Guest are viewing this topic.

Offline tmh983Topic starter

  • Newbie
  • Posts: 3
  • Country: au
Hi All,

new joiner here to the EEVblog forum. I'm an EE here in Newcastle, Australia. Proffessionally I work in automation - PLC and SCADA system design and integration. Recently i've started a new hobby of electronics. I've managed to build a couple of things so far including a smd oven made from a toaster oven controlled by a pic18. My current project is a bluetooth speaker, which leads me on to my technical question...

I am designing a 4-cell li-on battery pack, i'm working on the battery protection side of it right now. I am looking at using a TI bq77PL157A4225 for this purpose, but i'm stuck on how to set up the NMOS low side protection switch. The TI datasheet shows it like this:
953898-0

but that would put the diode of the NMOS like this:
953890-1
and this would mean that the protection circuit can't actually properly disconnect the battery from the circuit in the event of a fault...
Has TI got it backwards in their datasheet, or am I missing something here.
 

Offline Peabody

  • Super Contributor
  • ***
  • Posts: 1965
  • Country: us
Re: NMOS low side switch for battery protection with TI bq77PL157A4225
« Reply #1 on: March 23, 2020, 02:55:05 am »
I think you are correct.  It's oriented backwards in the schematic.  A guy posted the same question in the TI Power Management Forum five years ago, but never got an answer.

https://e2e.ti.com/support/power-management/f/196/t/371284
 

Offline Daixiwen

  • Frequent Contributor
  • **
  • Posts: 351
  • Country: no
Re: NMOS low side switch for battery protection with TI bq77PL157A4225
« Reply #2 on: March 23, 2020, 09:02:30 am »
This component is for overcharge protection only. And I think the MOSFet orientation on there schematic is correct, it will interrupt the charge current when opening the MOSFet.
If you want undercharge/over current protection while discharging, you will need another circuit. You can find some that provide protection both during charging and discharging, but then you need two MOSFets connected in opposite directions.
 

Offline Peabody

  • Super Contributor
  • ***
  • Posts: 1965
  • Country: us
Re: NMOS low side switch for battery protection with TI bq77PL157A4225
« Reply #3 on: March 23, 2020, 01:42:38 pm »
Well, I guess it depends on the direction of the current flow you're trying to interrupt.  It seems to me that when you are charging, the flow of conventional current is from the positive charging source into the positive 2S terminal, out through the negative terminal to ground (which is the negative return point of the charging source).  Ground is the lowest voltage point in the circuit.  In that case the drain should be at 2S- and the source at ground, and the body diode would block that flow when the mosfet is off.  But that is not what's shown in the drawing.

When discharging, the flow is from the 2S+ battery terminal out through the circuit to ground and then back into the negative 2S terminal.  2S- becomes the lowest voltage point in the circuit.  In that case the drain should be at ground and the source at 2S-, and the body diode would block that flow when the mosfet is off.

The DW01 datasheet shows the double-mosfet circuit Daixiwen described, but the DW01 only works for 1S.

I guess I still think the mosfet is oriented backwards in the drawing.

 

Offline Daixiwen

  • Frequent Contributor
  • **
  • Posts: 351
  • Country: no
Re: NMOS low side switch for battery protection with TI bq77PL157A4225
« Reply #4 on: March 24, 2020, 08:19:00 am »
When charging the battery, the current from the charger enters through the PACK+ terminal, into the cells, then through the Mosfet and out through the PACK- terminal. The potential on PACK- is lower than the "ground" on the schematic, so I still think the orientation of the MOSfet is correct on the drawing.

And of course when discharging it's the other way round, and then the "ground" on the schematic is the point at the lowest potential.
 

Offline Daixiwen

  • Frequent Contributor
  • **
  • Posts: 351
  • Country: no
Re: NMOS low side switch for battery protection with TI bq77PL157A4225
« Reply #5 on: March 24, 2020, 08:25:55 am »
You can use a BQ7790400 for example for a full pack protection, that will protect your cells against overcharge, over discharge, short circuit and over temperature.
 

Offline tmh983Topic starter

  • Newbie
  • Posts: 3
  • Country: au
Re: NMOS low side switch for battery protection with TI bq77PL157A4225
« Reply #6 on: March 24, 2020, 10:49:41 am »
Yeah I see your point - the NMOS will only be effective in either the charge or discharge direction depending on which way its configured -  but can't protect for both cases.
I'll check out that other chip you suggested - cheers.
 

Offline Peabody

  • Super Contributor
  • ***
  • Posts: 1965
  • Country: us
Re: NMOS low side switch for battery protection with TI bq77PL157A4225
« Reply #7 on: March 24, 2020, 02:49:49 pm »
When charging the battery, the current from the charger enters through the PACK+ terminal, into the cells, then through the Mosfet and out through the PACK- terminal. The potential on PACK- is lower than the "ground" on the schematic, so I still think the orientation of the MOSfet is correct on the drawing.


The charger must have a positive terminal and a negative terminal, and conventional charging current must flow out of its positive terminal, through the battery, back to its negative terminal.  The current doesn't just disappear into the battery.  It has to go through the battery and return to the charger's negative terminal. 

Where is the charger connected in the circuit?  I assumed it was connected to PACK+ and to Ground.  If that is  correct, then Ground must be at the lowest potential in the circuit, and charge current must be flowing out of PACK- through the mosfet to Ground.  The potential at PACK- must then be higher than at Ground, otherwise charging current wouldn't return back to the charger.

Where do you assume the charger is connected?

 

Offline Peabody

  • Super Contributor
  • ***
  • Posts: 1965
  • Country: us
Re: NMOS low side switch for battery protection with TI bq77PL157A4225
« Reply #8 on: March 24, 2020, 04:39:13 pm »
I've modified the original diode drawing to add the charger.  To me, the body diode would continue to pass current when the mosfet is switched off.

Edit.  OK, I'm wrong about this.  I was looking right at the original drawing and just not seeing what was show there.  The battery stack is actually on the left, with the bottom cell connected to ground.  The charger would be connected at PACK+ and PACK-.  The current flow during charging is counterclockwise, so the mosfet orientation is correct.  I don't see that well, and thought the little cell symbols were capacitors, and the batteries would be connected between PACK+ and PACK-, but of course that's wrong.



955532-0
« Last Edit: March 24, 2020, 05:32:14 pm by Peabody »
 

Offline Daixiwen

  • Frequent Contributor
  • **
  • Posts: 351
  • Country: no
Re: NMOS low side switch for battery protection with TI bq77PL157A4225
« Reply #9 on: March 25, 2020, 08:40:22 am »
Yes it was a misunderstanding, I understand your point of view now.
When designing a battery pack we usually call PACK+ and PACK- the protected battery terminals, and the rest (especially the cells) is not accessible to the user. I shouldn't assume that everyone knows that or follow this convention ;)

I think the main problem with the schematic on that application note is the usage of the ground symbol as it confuse people on where to connect the negative terminal of the battery. If you don't have the ground symbol at all then it is easier to see that you connect your charger at the PACK+/PACK- terminals.
 

Offline Peabody

  • Super Contributor
  • ***
  • Posts: 1965
  • Country: us
Re: NMOS low side switch for battery protection with TI bq77PL157A4225
« Reply #10 on: March 25, 2020, 01:55:38 pm »

I think the main problem with the schematic on that application note is the usage of the ground symbol as it confuse people on where to connect the negative terminal of the battery. If you don't have the ground symbol at all then it is easier to see that you connect your charger at the PACK+/PACK- terminals.

Well then would PACK- be connected to ground?  If the battery is going to power the circuit when the charger is not connected, there has to be a connection to circuit ground somewhere.  If that's at PACK-, then the entire load current has to flow through the mosfet.  But I guess that's ok.  After all, in a DW01 circuit, the entire load current flows through *two* mosfets.
 

Offline Daixiwen

  • Frequent Contributor
  • **
  • Posts: 351
  • Country: no
Re: NMOS low side switch for battery protection with TI bq77PL157A4225
« Reply #11 on: March 26, 2020, 08:42:24 am »
Yes the actual ground for the application connected to the battery is PACK-. The "ground" symbol on the schematic is just an internal ground, not accessible outside the battery pack. That's why I don't like to use that symbol for battery schematics.
The current always go through the MOSFet, but when it is conducting its rds is very low so it's not really a problem.
 

Offline tmh983Topic starter

  • Newbie
  • Posts: 3
  • Country: au
Re: NMOS low side switch for battery protection with TI bq77PL157A4225
« Reply #12 on: March 26, 2020, 12:10:21 pm »
thanks for all your input on this, i've got a good handle on how to set it up to protect the battery during the charge cycle. During discharge it is essentially unprotected by this circuit.  Im using a system power controller that monitors the batt. current and can disconnect the batt. in case of overcurrent, so i think thats ok.

Is there any tips on selecting a mosfet for the batt. protection purpose? I am looking at a NTTFS4824N device, but im not sure how best to choose it. This device can easily handle the current rating, and has an pretty low rds at the gate drive voltage of the protection circuit. So i'm pretty sure it will basically work. Is there any other criteria i should look at? I'm guessing you would choose a different device for a high speed modulation vs a simple low side switch like this, but i'm not really sure what to look for...?
 

Offline Peabody

  • Super Contributor
  • ***
  • Posts: 1965
  • Country: us
Re: NMOS low side switch for battery protection with TI bq77PL157A4225
« Reply #13 on: March 26, 2020, 04:30:28 pm »
The only other thing I would look at is power dissipation.  I don't know how much current you're looking at, and with low RDSon it may not matter, but you just want to be sure it doesn't get hot.

Remember that discharge protection is not only protecting against shorts, but also shutting things down when the battery gets too low.  You can damage a LIPO cell by over-discharging it.

I would be interested in seeing your charging circuit.  I do ok with 1S circuits, but I would like to see how you balance the four cells during charging.

 

Offline Daixiwen

  • Frequent Contributor
  • **
  • Posts: 351
  • Country: no
Re: NMOS low side switch for battery protection with TI bq77PL157A4225
« Reply #14 on: March 27, 2020, 08:04:51 am »
For a charge protection MOSFet you don't need to look into anything other than the max current, rdson, power dissipation and price.
Don't forget that current will also flow through the mosfet during discharge, so it is usually the maximum discharge current that you need to take into account. It also needs to survive a short circuit current during the time it takes for your protection to kick in.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf