Is this the correct way to connect Mosfets in Parallel?

[gary350]

I found this circuit online.

First thing I need to know, is this the correct way to parallel mosfets?

Next, is this the correct resistance for mosfets pin 1?

Also I dont know how to use the data sheet to determine the value of R1, R2, R3, R4 they are all the same value? I already tested a 2 mosfet circuit using 240 ohm resistors it works good but that does not mean it could not work better. Mosfet number is on the drawing.

**broken link removed**

 

[crutschow]

MOSFETs naturally tend to balance their load when connected in parallel as a switch since they exhibit a positive temperature coefficient of ON resistance with temperature. Thus the one carrying more current will heat up more, increasing its resistance, which then reduces the percentage of current it is carrying and causing the other transistor to carry more.
So I see no need for the source resistors.

But I don't understand the circuit since all the gates are connected together with their sources connected to +12V and the transformer is connected to +15V. Thus the drain-source voltage is only 3V.
What's the circuit supposed to do?

 

[chemelec]

Your Source Resistors can just be 0.1 Ohm 5 Watt Wire Wound Resistors.
It will hanle the current.

For your Gate resistors, I would use resistors of Either 47, 56 or 82 Ohms.
But your Circuit doesn't seem Correctly wired to supply voltages.

 

[Tony Stewart]

MOSFETS are designed with PTC.

Whereas BJT's have the NTC Shockley Effect where Vbe,Vce drops with rising temp and hFE increases causing thermal runaaway on parallel operation, hence series R or PTC wire helps to equalize the current sharing. Low coupled thermal resistance is important however.

 

[gary350]

Here is my circuit. I hope I did not make any mistakes on the drawing. I have a test circuit built with 2 mosfets using 240 ohm resistors it is pully more than 20 amps so 3 mosfets in parallel will probably be more than 60 amps, probably 70 amps. My power transformer secondary coil is rated 65 amps only because there it not enough room to use larger wire but the transformer is rated 70 amps. I may need to change the secondary coil to a higher voltage, low current to get the full 1050 watts from this transformer if that is needs. Maybe I can hand feed larger amp enamel coated copper wire through the transformer holes without taking the welded EI laminations apart. The 7812 voltage regulator is rated 1 amp I am not sure yet how any amps 6 mosfet gates need. I could use two 7812 voltage regulators or center tap the transformer for 12 vdc. I still have room for 3 more turns of 65 amp wire, each turn is .8 volts. The secondary coil is now 10.8 vac 3 more turns will make it 13.2 vac. VDC will change from 15.3 to 18.6 vdc. All suggestions will be appreciated. Power supply caps are 3300uf each = about 20,000. uf. My power supply rectifiers are rated 75 amps now. The mosfets are rated 60v 55a. each.

I revised the wiring mistake.

**broken link removed**

 

[gary350]

This is the power supply transformer.

 

[tcmtech]

I have doubts given the way it drawn.

As mentioned by others the load balancing resistors are pointless on a MosFet based switching circuit. I've got a few older style power inverters and some massive car audio amplifiers in my shop that have something like 10+ MosFets on each side of their primary switching circuits and they do no have any balancing resistors on them despite switching several hundred amps of current.

Now what I do see in your schematic as you have drawn it is a number of dead shorts which should render the overall circuit either totally non functional or highly smokeable when the power is turned on.

I am guessing this is a continuation on your experiments to build some sort of induction heater without actually using any proven schematics or designs as reference?

 

[chemelec]

Diodes, 1N5819 are Only 1 Amp rating.

Does NOT seem Practical on what you show!

 

[Tony Stewart]

Not sure the point of this design as MOT's are not very efficient cores, but they are low cost.
It seems to be a tad lossy.

However the air coil would work fine for RF in a gas chromatography device with substance flaming in argon.

 

[audioguru]

I do not see anything to kick the circuit into oscillating. Both sets of Mosfets will probably turn on at the same time when the circuit is powered then lots of magic smoke will come.

 

[alec_t]

The bridge rectifier is back-to-front. I smell magic smoke from the FET body diodes .

 

[Tony Stewart]

Body diode polarity looks ok to me with the top being a mirror image of the bottom.

Resonant frequency is unknown.

 

[gary350]

Yes, this is a proven circuit I found it online with videos. I had to add 2 extra mosfets then draw it a little different with the mosfets connected right at the capacitor bank. The original circuit uses 4 mosfets 150v 100a each. 2 mosfets in parallel on each side. Those 150v 100a mosfets are $8 each = $32 total. The mosfets I am using are 60v 55a pack of 20 mosfets 50 cents each with free postage from china. The guy claims his circuit is 2000 watts with 4 mosfets with a 30v power supply.

My circuit with 2 mosfets tests 15v about 23a = 345 watts so 6 mosfets should be about 1035 watts. Right?

I could go all out with 12 mosfets in parallel and try for 2000 watts but that is a big jump with what I have now.

 

[ChrisP58]

Yes, this is a proven circuit I found it online with videos. I had to add 2 extra mosfets then draw it a little different with the mosfets connected right at the capacitor bank. The original circuit uses 4 mosfets 150v 100a each. 2 mosfets in parallel on each side. Those 150v 100a mosfets are $8 each = $32 total. The mosfets I am using are 60v 55a pack of 20 mosfets 50 cents each with free postage from china. The guy claims his circuit is 2000 watts with 4 mosfets with a 30v power supply.

My circuit with 2 mosfets tests 15v about 23a = 345 watts so 6 mosfets should be about 1035 watts. Right?

I could go all out with 12 mosfets in parallel and try for 2000 watts but that is a big jump with what I have now.

Just adding mosfets alone cannot increase your power. It increases your capacity to switch power, but that power has to come from somewhere else.

And a note on the mosfet manufacturers current rating claims.
It is a calculated number that only applies when you are holding the case at 25C while dissipating power. Not really practical unless you're using refrigerated cooling on the heatsink.

 

[gary350]

When I first built this circuit it produced a worthless amount of power. The circuit sometimes will not osocillate so all the mosfets burn up. After reading and learning that mosfets need a +12v gate and not +15v like the power supply the circuit works every time now but power is still low. Skin effect is a problem at high frequency I experemented and found 60KHz is much better than 221KHz like the circuit was originally designed. More capacitors lowers the frequency and supplies more current to the LC coil. Then after experementing with different mosfets I learn some are much better than others and why. This is a very educational project I am having FUN and learning TOO. My power supply had to be improved several times now it is 75 amps, 15.3v, with good filter capacitors and better choke coils. A circuit that was originally only 4 amps is now in the 23 amp range, power has increased from about 60w to 345 watts with 2 mosfets. The circuit I found online uses 4 mosfets 150v 100a with 30v power supply video looks good and the guy claims it is 2000 watts.

I have learned just because a mosfet is rated 50 amps does not mean it can run at 50 amps all the time. I have changed heat sinks several times, now I have a very large 3 ounce heat sinks my mosfets get hotter than the heat sink, the heat is just not physically able to transfer into the heat sink as fast as it needs to. The mosfet is producing more BTUs than can transfer to the heat sink even with good heat sink compound and a good clamp. With 23 amps the mosfet is heating up fast and is always about 5 degrees F hotter than the heat sink and a cooling fan cools the heat sink but the mosfet is still hot even with the heat transfer compound. When heat sink gets to about 130 degree F I shut it down and it continues to get about 10 degrees F hotter. If the temperature goes over 150 degree F mosfets explodes, sound like a tiny firecracker Boom, flames shoot out of both mosfets and there is a small cloud of smoke. I notice one of the YouTube videos the guy has his mosfets and heat sink under water, only the mosfet wires stick straight up above the water. There is another YouTube video the guy has his mosfets sandwich clamped between 2 large heat sinks with box fans.

With 3 mosfets in parallel I should be able to use current limiting resistors on pin 3 to find a safe current at maybe 18 or 20 amps or so then the circuit will not over heat and need to be shut down in 2 minutes. I am trying to find the spot where the mosfets will run at continious current and not over heat. Even if I have to limit current to 15 amps each per mosfet = 45a x 15v = 675 watts.

This is just another fun educational learning experience. I do have a use for this induction heater if I can get enough power to heat a 2" diameter steel pipe to 1400 degrees F other wise I will continue to use a 100 ton hydraulic press.

 

[chemelec]

It is NOT JUST how much a heat sink WEIGHS or How BIG it is to give good Cooling.
It is Partly: How well the heat sink is constructed and how good is your mounting of the mosfet is to it.
(Thermal Conduction between the Mosfet and the Heatsink.)
Even More Important, It is also about Ambient Temperature and Air Flow to help Cool the Heat Sink.

130 Degrees of which?
"C" or "F"

130C is MUCH TOO HOT!

Just Increasing the Number of Mosfet will not really increase the Power.
You need to Increase the Supply Voltage so to also increase Current Flow.