mosfet bridge rectifier not working!

[bluebloodtea]

I have been trying off and on for weeks to get the circuit posted on here to work. I hope some of you have tried this one and got it working because I need help!!!

For the MOSFETs, I am using 2 x IRF540n and 2 x IRF9540. As far as the chip goes, I found an NTE859 because there were no TL084s available to me. Although not exactly the same, these chips are definitely comparable.

The problem I am having is that as soon as I apply power to the circuit, the n-channel (and ONLY the n-channel) FETs start to heat up to an absolutely ridiculous degree (to about 150 deg C in about 2-3 sec.). Now this says to me that there is a dead short somewhere in the n-channel part of the circuit, but I cannot see one. I have also wired and rewired the breadboard, and meticulously checked it over, all to no avail.

Help, please!!

[hairylee]

can you put a link or attach the circuit, so I can look at it?

[bluebloodtea]

Here is a link to the circuit on this site: http://www.elektor.com/magazines/2006/july/power-mosfet-bridge-rectifier.58316.lynkx

I can attach print screens of the article as well, if need be.

[Elektor Editor]

your construction may self-oscillate as a result of the component changes with respect to the original as published.
The oscillation may cause thermal runaway and destruction of the parts.
Use an oscilloscope to investigate.

Jan

[bluebloodtea]

Although I don't believe self-oscillation to be the case, I am going to investigate.

The article states that the choice of MOSFET is not critical to the circuit. As far as the change of the TL084, the NTE859 that I am using is very similar to the TL084.

Any ideas on how to investigate this when I can't power up the circuit?

[hairylee]

Hi,
Unfortunately this is not a good circuit and you will have problems with heat in the mosfets.
The mosfets specified in the schematic have about 10 times lower on resistance than the parts you substituted. There is nothing wrong with you using the NTE859 or the MOSFET's you have. The circuit has a few flaws that will cause the mosfets to heat up.
Supplying the op amp from the output with a load of around 3 Amps results in bad mosfet gate drive. The mosfets are driven nowhere near their fully on state and will dissipate a few watts. If you have no heatsink on these parts they will quickly become very hot.
Supplying the op amp from a separate 12V supply (Maybe via a normal diode rectifier made up of 4 1N4148's and a 220uF cap) provides the MOSFET's with a decent gate drive. Substituting the op amp with a rail-rail output op amp will also improve things.
Also, the lower the MOSFET on resistance, the better.

There is probably a better way of doing this.....

[bluebloodtea]

Thanks for your help. The reason I was interested in this particular circuit was the output being able to supply a capacitor load, unlike other MOSFET bridge rectifiers. Also, I thought the output supplying the input to the chip was a great idea. I guess I was wrong!

Well, back to the drawing board....

[thanx]

Although I don't believe self-oscillation to be the case, I am going to investigate.