Author Topic: Power switch using LTC4412; P ch. MOSFET does not open  (Read 3352 times)

0 Members and 1 Guest are viewing this topic.

Offline zzattackTopic starter

  • Regular Contributor
  • *
  • Posts: 126
  • Country: nl
Power switch using LTC4412; P ch. MOSFET does not open
« on: December 01, 2015, 03:22:34 pm »
I built a power switch circuit based on a LTC4412 that will disconnect my battery when an auxilary power source is available. Schematic is built exactly as presented in the data sheet:


The wall adapter input, when available, is a regulated 6V DC supply. The battery is a single-cell lipo that should be between 4.2-3.7V. The load goes toward a 3.3V low-dropout voltage regulator.

Testing the operation of this circuit, I found that with just the battery hooked up, the LTC4412 pulls its gate low and Vbat is fed into the regulator. With the auxilary source enabled I verified that the gate pin measures a high. So far so good. Now for the problem: the MOSFET does not seem to open completely. I'm measuring 6V at both the drain and gate pins, but the PSU also hits the current limiter and I read about 5.5V at the source pin (thus going into the battery!). I've replaced the MOSFET but the problem remains.

So, my best guess is that I did not choose a correct MOSFET for the job. The part I'm using is the NDS332p, with important parameters
VDS(MAX): -20V
VGS(VT): -0.6V
RDSon: 0.4Ohms typical

The VGS(VT) seems most important and should work with the 6V-4.2V difference between aux and battery supply.
How can I figure out what's happening here?
 

Offline homebrew

  • Frequent Contributor
  • **
  • Posts: 293
  • Country: ch
Re: Power switch using LTC4412; P ch. MOSFET does not open
« Reply #1 on: December 01, 2015, 07:35:53 pm »
I'm measuring 6V at both the drain and gate pins, but the PSU also hits the current limiter and I read about 5.5V at the source pin (thus going into the battery!).

Did you put the FET in reverse? (Accidentally swapped drain and source?)

Would at least explain the behavior as it would operate as a diode (but in reverse) ...

Just an idea ...

 

Offline Radiohead

  • Contributor
  • Posts: 36
  • Country: nl
Re: Power switch using LTC4412; P ch. MOSFET does not open
« Reply #2 on: December 01, 2015, 08:23:59 pm »
The Mosfet does not behave as an ideal diode, as it has an internal parasitic diode between drain-source. As you have noted, it charges your battery through the parasitic diode when the auxilary power source is connected. Use back-to-back mosfet configuration for ideal mosfet operation and connect CTL to GND, as described in the datasheet. See figure 4 and read "Ideal Diode Control with a microcontroller" above figure 4.
 

Offline homebrew

  • Frequent Contributor
  • **
  • Posts: 293
  • Country: ch
Re: Power switch using LTC4412; P ch. MOSFET does not open
« Reply #3 on: December 01, 2015, 08:48:34 pm »
Sorry, äh, no!

That is the REASON that this part exists. It should PREVENT reverse current flow! However FET does behave as a diode in one direction. Hence you could not turn the load off in that configuration and hence you need the back-to-back FETs to have that feature if desired. Not for the normal behavior of switching between two supplies...
 

Offline zzattackTopic starter

  • Regular Contributor
  • *
  • Posts: 126
  • Country: nl
Re: Power switch using LTC4412; P ch. MOSFET does not open
« Reply #4 on: December 02, 2015, 03:03:30 pm »
I'm measuring 6V at both the drain and gate pins, but the PSU also hits the current limiter and I read about 5.5V at the source pin (thus going into the battery!).

Did you put the FET in reverse? (Accidentally swapped drain and source?)
I truly thought I had moved past this level of stupidity but must shamefully admit this was indeed the cause.. Thanks!
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf