6 amp FET bridge

Go To Last Post
14 posts / 0 new
Author
Message
#1
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Operating conditions: 10 to 30 volts DC, either polarity.

I'm needing to rectify DC power because the polarity is undetermined. The voltage is 10 to 30 volts and the current up to 6 amps. Since at 6 amps, the conducting diodes would dissipate like 4 1/2 watts each, I wanted to do better for this SMALL device. (They dissipate more heat than the H bridge motor driver.)

So I went looking for mosfets to make a bridge. Cheap ones have like a +- 20v Gate-Source spec, so I used a resistor divider.

Theory of Op: A resistor divider gives me V/2, or between 5 and 15 volts.

V/2 applied to the gate of the PFet on the + side turns it on, connecting the output to the + side. Since V/2 is "above" the - side, that PFet is turned off.

V/2 applied to the gate of the NFet on the - side turns it on, connecting the output to the - side. Since V/2 is "below" the + side, that NFet is turned off.

Does anybody see sparks, smoke or melted plastic trains from this arrangement?

Oh. Yes. The PFet is FDS4435BZ
The NFet is PMN25EN

Only because they're small, rated and available at reasonable cost.

Attachment(s): 

If you don't know my whole story, keep your mouth shut.

If you know my whole story, you're an accomplice. Keep your mouth shut. 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

(I'm thinking the body diodes are going to blow the thing up.)

If you don't know my whole story, keep your mouth shut.

If you know my whole story, you're an accomplice. Keep your mouth shut. 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Quote:
Does anybody see sparks and smoke from this arrangement?

Torby, look closely at the FET body diodes...

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I see that the dyslexic brat drew them backwards!

Assuming that the mosfets are manufactured properly and the body diodes are right-way-around... :roll:

If you don't know my whole story, keep your mouth shut.

If you know my whole story, you're an accomplice. Keep your mouth shut. 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Torby wrote:
I see that the dyslexic brat drew them backwards!

Assuming that the mosfets are manufactured properly and the body diodes are right-way-around... :roll:

they still will blow as they are reversed to the normal polarity of the FET, they will be conducting at some point.....

you could see if there are any schottky diodes with that high current rating, that would decrease the consumed power some what(or watt???)

Also I did not look at teh spec of those fets but 6amp 1ohm is 36watt per FET you need a FET with a low internal resistance to keep the power waste down... 10mOhm will do 360mW still.

the badest thing is that you have DC so it is not that all diodes/fest consume the same amount of power. 2 out of the 4 items will be conducting all the time and the others not.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Well, that's where I got the basic idea. I'd read about it years ago when I was first introduced to mosfets. Found that page the other day when I realized that would improve the device.

Quote:
Also I did not look at teh spec of those fets but 6amp 1ohm is 36watt per FET you need a FET with a low internal resistance to keep the power waste down... 10mOhm will do 360mW still.

You mean this chart in the PFet data? Oh, my. I was thinking it was 20m ohms, not 1.5 ohms! Well, that goofs up everything. I'll need to look for a better fet.

Attachment(s): 

If you don't know my whole story, keep your mouth shut.

If you know my whole story, you're an accomplice. Keep your mouth shut. 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Pololu has a shield with 2 of those 30amp chips on it.

Imagecraft compiler user

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Check out the LT4320 "Ideal Diode Bridge Controller"

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Normalized Drain to Source On-Resistance is unitless - a multiplier to the spec'd Rds which is dependent on your particular operating point. Don't be jumping to conclusions based on the chart alone.

And, a big thumbs up for the LT4320! :D

 

In consideration of others, please RTFM!

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Was hoping to save board space with small FETs. I went back to DPACK size FETs and got the power dissipation down a lot more. Hmm, with that LT4320 part, I would not need P channel FETs.

If you don't know my whole story, keep your mouth shut.

If you know my whole story, you're an accomplice. Keep your mouth shut. 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Torby,
I think I accidentally confused you with my calculation example.
I used the 1 Ohm as an example. As said I did not look into the datasheet. If it is only 10mOhm typical( what is max?) then with a Vgs of only 5V you would have 15Ohm as per your posted chart.
Then you would also be good on the dissipation side(that was what my example was trying to tell)
I have had a collegue that was very good in finding very cheap fets in the past,but they all had crappy Rds_on values (Ohms not milliohms values) so we always had to double check to see if the FET would not have a problem when we started running current through it.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I'll carefully reread data again to see if the smaller fets will do.

If you don't know my whole story, keep your mouth shut.

If you know my whole story, you're an accomplice. Keep your mouth shut. 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Eek!

I populated one of these boards, and indeed, my "FET rectifier" is a "FET Heater."

I had drawn the body diodes the right way around:

But you have to connect your FET's the other way around:

Notice the difference. Where normally you would connect the source to the supply rail and reverse bias the body diode, here, you have the drain connected to the supply rail and FORWARD bias the body diode. Otherwise, the body diode of the "off" fets is forward biased and your bridge is a short. How handy that my bench supply has a 2 amp current limit and I wasn't using one of my 38 Amp "IED" batteries to run it

If you don't know my whole story, keep your mouth shut.

If you know my whole story, you're an accomplice. Keep your mouth shut.