Author Topic: Reverse polarity / low batery protection issue  (Read 1221 times)

0 Members and 1 Guest are viewing this topic.

Offline iggarpeTopic starter

  • Contributor
  • Posts: 13
Reverse polarity / low batery protection issue
« on: April 23, 2021, 06:13:12 pm »
Hi all,

I'm experiencing a fault in one in my designs and just cannot figure out what the problem is.

The system is powered by two 12V lead-acid bateries in series and specified for an input voltage from 20V to 30V.

I have to PMOS back-to-back at the input rated for that voltage and a 10V zener to limit Vgs well below the maximum. See attached schematic.

There is a low power high input voltage linear regulator (not shown)  that powers a microcontroller (now shown) that implements the low battery detection, hence the shown op amp to sense input voltage.

The fault I'm experiencing is Q2 dies and becomes shorted. I'm positive this happens without any overvoltages at the input (plus that should kill Q1 too, right?).

Any ideas?

Thanks.
 

Online Ian.M

  • Super Contributor
  • ***
  • Posts: 12807
Re: Reverse polarity / low batery protection issue
« Reply #1 on: April 23, 2021, 06:21:15 pm »
If Voen is oscillating while current is being drawn from the battery, Q1 switches between channel conduction and body diode conduction, so its dissipation is very unlikely to exceed 0.7V*Iload, but Q2 will be repeatedly swept through its linear region with massive dissipation.
 

Offline iggarpeTopic starter

  • Contributor
  • Posts: 13
Re: Reverse polarity / low batery protection issue
« Reply #2 on: April 24, 2021, 09:18:07 am »
No, it's not that. VOEN is driven by the microcontrolller and is never oscillating. Plus death by overheating would be apparent.

VOEN is high all the time except when battery is low, and hysteresis plus delay is implemented:

1- Disable mosfets if voltage is below 22.5V for more than 10s.
2- Enable mosfets if voltage is above 25.5V for at least 60s.
 

Offline fcb

  • Super Contributor
  • ***
  • Posts: 2117
  • Country: gb
  • Test instrument designer/G1YWC
    • Electron Plus
Re: Reverse polarity / low batery protection issue
« Reply #3 on: April 24, 2021, 10:05:01 am »
Things that will kill your MOSFET's:

1. Over current
2a. Over voltage (VDS) or
2b. Over voltage (VGS)
3. Over temperature

You're protecting against 2b only.

You could have an inductive source (2a) - try a capacitor to ground at the front end of your switch?
You could have an inductive load (2a) - back EMF? less likely but the body diodes of the FETS won't save you in this config, perhaps a diode across the whole switch?
You could have a heavily capacitive load (1) - less likely to be a problem as you have a reasonably slow turn-on with that 50K pull-down. Have you calculate turn-on time?
You could be taking to long to switch on (3) - perhaps your turn-on is too slow leading to too long in the linear region and overheating? Reduce 50K, or try a different driver topology.
https://electron.plus Power Analysers, VI Signature Testers, Voltage References, Picoammeters, Curve Tracers.
 
The following users thanked this post: iggarpe

Online Someone

  • Super Contributor
  • ***
  • Posts: 4510
  • Country: au
    • send complaints here
Re: Reverse polarity / low batery protection issue
« Reply #4 on: April 24, 2021, 02:29:33 pm »
you have a reasonably slow turn-on with that 50K pull-down. Have you calculate turn-on time?
You could be taking to long to switch on (3) - perhaps your turn-on is too slow leading to too long in the linear region and overheating? Reduce 50K, or try a different driver topology.
So another possible failure mode: outside the SOA curves. The particular MOSFET looks like a poor choice for a load switch on that basis (100x worse than typical parts).

Should be pretty obvious when recorded with a scope capturing the Q2 waveforms and drain current.
 
The following users thanked this post: iggarpe

Offline fcb

  • Super Contributor
  • ***
  • Posts: 2117
  • Country: gb
  • Test instrument designer/G1YWC
    • Electron Plus
Re: Reverse polarity / low batery protection issue
« Reply #5 on: April 24, 2021, 03:23:29 pm »
Total gate charge on IPD50P04P4L is 45nc (10V)
Zener limits to approx 10V, although you might not be getting that if you are pulling a small amount of current.

Q=CV therefore you have the equivalent of 4.5nF capacitor on each gate to charge. Two gates gives you 9nF. This is quite a bit to discharge.

Plot the gate voltage when you switch-off the 2N700x, you'll be spending along time discharging that gate (Vgs) below the 3.6V plateau.

Switch off responsible for the majority of the heating, more so than switch on.

https://electron.plus Power Analysers, VI Signature Testers, Voltage References, Picoammeters, Curve Tracers.
 

Offline iggarpeTopic starter

  • Contributor
  • Posts: 13
Re: Reverse polarity / low batery protection issue
« Reply #6 on: April 24, 2021, 05:31:30 pm »
Thank you very much for your analysis.

You could have an inductive source (2a) - try a capacitor to ground at the front end of your switch?

Source is usually two lead-acid 12V batteries in series with their charger. It's a quality charger but I've experienced the same failure in systems powered by a mean well 27V regulated power supply.

You could have an inductive load (2a) - back EMF? less likely but the body diodes of the FETS won't save you in this config, perhaps a diode across the whole switch?

The load is a TAS5622 class D audio amplifier. Definitely not inductive.

You could have a heavily capacitive load (1) - less likely to be a problem as you have a reasonably slow turn-on with that 50K pull-down. Have you calculate turn-on time?

That is the case. The load has about 3000uF of decoupling.

I've gone as far as adding a 100nF capacitor in series with R35A for a super-slow turn on and it doesn't seem to fix the issue.

In any case, those PMOS are good up to 200A peak current, I don't think I'm getting even closer with a 3000uF load.

You could be taking to long to switch on (3) - perhaps your turn-on is too slow leading to too long in the linear region and overheating? Reduce 50K, or try a different driver topology.

As I said, I tried way way way slower turn on and it doesn't fix the issue. Plus I would expect some physical damage in the parts if they have been killed by overheating.

 

Offline iggarpeTopic starter

  • Contributor
  • Posts: 13
Re: Reverse polarity / low batery protection issue
« Reply #7 on: April 24, 2021, 05:39:20 pm »
Total gate charge on IPD50P04P4L is 45nc (10V)
Zener limits to approx 10V, although you might not be getting that if you are pulling a small amount of current.

Q=CV therefore you have the equivalent of 4.5nF capacitor on each gate to charge. Two gates gives you 9nF. This is quite a bit to discharge.

Plot the gate voltage when you switch-off the 2N700x, you'll be spending along time discharging that gate (Vgs) below the 3.6V plateau.

Switch off responsible for the majority of the heating, more so than switch on.

Turn off time is not an issue, even if it's a few milliseconds. The load is a class D power amplifier and in the worst case scenario (i.e. playing when battery low is detected) it would be drawing ~5A
 

Offline iggarpeTopic starter

  • Contributor
  • Posts: 13
Re: Reverse polarity / low batery protection issue
« Reply #8 on: April 24, 2021, 05:42:52 pm »
you have a reasonably slow turn-on with that 50K pull-down. Have you calculate turn-on time?
You could be taking to long to switch on (3) - perhaps your turn-on is too slow leading to too long in the linear region and overheating? Reduce 50K, or try a different driver topology.
So another possible failure mode: outside the SOA curves. The particular MOSFET looks like a poor choice for a load switch on that basis (100x worse than typical parts).

Should be pretty obvious when recorded with a scope capturing the Q2 waveforms and drain current.

Power consumption when the system starts up is low (~0.2A), so slow turn on should not be a problem.
Power consumption when the low battery condition is detected and thus the mosfets turned off could be, in worst case scenario, ~5A, so linear region doesn't seem a problem here either.
I've not observed any kind of physical damage in the dead parts, not even the slightest, which I'm guessing should be evident if killed by overheating.

I would appreciate if you could elaborate a bit on why this particular pmos is a bad choice for load switch.
 

Offline fcb

  • Super Contributor
  • ***
  • Posts: 2117
  • Country: gb
  • Test instrument designer/G1YWC
    • Electron Plus
Re: Reverse polarity / low batery protection issue
« Reply #9 on: April 24, 2021, 05:45:32 pm »
Based on what you're saying: the slow speed of your switch-on is probably what is killing your FETS.  They are spending too long in the linear region, and your load will look like a dead-short initially.

A couple of possible things to try:

1. Speed up your turn-on/turn-off, even if it costs you more idle current, at least you can test if it solves your problem
2. Pre-charge your load. You'll need to keep your load in a minmal current condition, and it does mean using another switch arrangement and another I/O on your micro, but the FET's can be much smaller.  We do this with a lot of USB powered products to avoid current spikes on connection.
https://electron.plus Power Analysers, VI Signature Testers, Voltage References, Picoammeters, Curve Tracers.
 
The following users thanked this post: Someone

Online Someone

  • Super Contributor
  • ***
  • Posts: 4510
  • Country: au
    • send complaints here
Re: Reverse polarity / low batery protection issue
« Reply #10 on: April 25, 2021, 12:02:45 am »
you have a reasonably slow turn-on with that 50K pull-down. Have you calculate turn-on time?
You could be taking to long to switch on (3) - perhaps your turn-on is too slow leading to too long in the linear region and overheating? Reduce 50K, or try a different driver topology.
So another possible failure mode: outside the SOA curves. The particular MOSFET looks like a poor choice for a load switch on that basis (100x worse than typical parts).

Should be pretty obvious when recorded with a scope capturing the Q2 waveforms and drain current.
Power consumption when the system starts up is low (~0.2A), so slow turn on should not be a problem.
Should be pretty obvious when recorded with a scope capturing the Q2 waveforms and drain current.
Talk is cheap, my time isn't.
 

Offline iggarpeTopic starter

  • Contributor
  • Posts: 13
Re: Reverse polarity / low batery protection issue
« Reply #11 on: May 03, 2021, 01:51:28 pm »
Based on what you're saying: the slow speed of your switch-on is probably what is killing your FETS.  They are spending too long in the linear region, and your load will look like a dead-short initially.

But that would lead to an overheat damage, which should show some physical evidence in the PMOS bodies, right?
 

Offline fcb

  • Super Contributor
  • ***
  • Posts: 2117
  • Country: gb
  • Test instrument designer/G1YWC
    • Electron Plus
Re: Reverse polarity / low batery protection issue
« Reply #12 on: May 03, 2021, 01:55:06 pm »
Based on what you're saying: the slow speed of your switch-on is probably what is killing your FETS.  They are spending too long in the linear region, and your load will look like a dead-short initially.

But that would lead to an overheat damage, which should show some physical evidence in the PMOS bodies, right?
Nope.  Rarely (in my experience) does the death of a FET result in readily observable damage. 
https://electron.plus Power Analysers, VI Signature Testers, Voltage References, Picoammeters, Curve Tracers.
 

Offline iggarpeTopic starter

  • Contributor
  • Posts: 13
Re: Reverse polarity / low batery protection issue
« Reply #13 on: May 03, 2021, 07:31:11 pm »
you have a reasonably slow turn-on with that 50K pull-down. Have you calculate turn-on time?
You could be taking to long to switch on (3) - perhaps your turn-on is too slow leading to too long in the linear region and overheating? Reduce 50K, or try a different driver topology.
So another possible failure mode: outside the SOA curves. The particular MOSFET looks like a poor choice for a load switch on that basis (100x worse than typical parts).

Should be pretty obvious when recorded with a scope capturing the Q2 waveforms and drain current.
Power consumption when the system starts up is low (~0.2A), so slow turn on should not be a problem.
Should be pretty obvious when recorded with a scope capturing the Q2 waveforms and drain current.
Talk is cheap, my time isn't.

You're aware talking takes time, right?
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf