Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Do N channel Mosfets have a current requirement?

Status
Not open for further replies.

gary350

Well-Known Member
Datasheet says maximum gate voltage 16v but I see nothing about gate current on any mosfet that I look up?

Is there a minimum amp requirement?


100_2057.JPG
 
MOSFET Gate current is typically uA. or even less. What part?

What is your circuit built to do? It looks like the MOSFET will get very hot.
 
Gate current depends on Ciss with Ig= Ciss dVg / dt but negligible for DC.
 
Not in static conditions, but they can have very high gate capacitance, several nF is not unusual.

It does not matter for occasional on/off control of light loads, but becomes important for such a SMPSUs and PWM applications - you then need a high-current gate driver, so the capacitance can be charged or discharged quickly to minimise the time the FET is in a high-dissipation state with current through it and significant voltage across it.

Gate drivers rated 1 - 2 A or more are pretty common, just to ensure a "fast" transition on<>off transition.
 
If your MOSFET drain and source are connected to the thus-labelled points in your post #1 circuit the supply will be shorted when the MOSFET turns on. Is that your intention?
 
This is all good information & very interesting. The H circuit is the circuit that I will eventually end up building. Every day experiments I learn new things the old circuit runs better with gate voltage of 15v not 12v. Different value choke coils work better too. Old circuit runs at 80KHz with no load at 2.5a but load make it change to 41KHz with 15a to 17a using different chokes, 1mh, 2mh, 4mh, 6mh, 8mh. Changing PS volts from 15vdc to 20vdc makes current increase to 35a. Then PS volts of 30vdc current is 52a. The new H circuit won't have all these problems it will run on a IRS2453D frequency and should stay at 40KHz all then time with or without a load. I have not decides what KHz is best for the H circuit. PS transformer needs more taps maybe today I find time to change it so it has, 15vdc, 20vdc, 25vdc, 30vdc, 35vdc. I have a 1500w & 3000w variac that would solve transformer taps problem but maybe another day.

100_2058.JPG


100_2059.JPG
 
I'd suggest you reduce the gate resistors on that IRS2453 design; no more that 100 Ohm, to get the FETs to switch faster and dissipate less power.
It's not a very high drive device, but the data sheet says 180mA typical but with 1K it's only allowing 15mA peak.

47 Ohms or lower may well be OK.
 
This circuit is an interesting learning experience. 2mh choke seems to be the best choice only because current is highest. 1mh choke is less current and 0mh the circuit will not run amps are 0 even with a load on the work coil. Current is 1a lower with 4mh and 2a lower with 6mh choke. When power supply voltage in increased from 15v 20v current increases from 10a to 20a with the work coil loaded. 10v heats a .160" still rod red hot in 30 seconds, 20v heats the same .160" steel rod red hot is 10 seconds at 40KHz. At the moment it does not seem safe to test the circuit with 25v current could go too high and blow the mosfets. I should probably rewind the transformer again for secondary taps of 21v, 22v, 23v, 24v, 25v.

Tests appears to show the new H circuit will work good at 40KHz with a 2mh choke, but I have not yet determined what PS voltage should be.

100_2113.JPG
 
Today I learned if PS voltage goes up the choke value needs to go up also to keep maximun current at or below 30 amps. Mosfets explode at 35 amps even though they are rated 55 amps. Changing PS voltage from 15v to 25v is a much bigger improvement than change the circuit KHz. I still need to do more testing before building the H circuit. NOW my circuit needs to be built with 45a wire.

Voltage is now 25v with 6mh choke coil. The larger value choke prevents the operating frequency from dropping when circuit is loaded.
2mh choke = 80KHZ drop to 41KHz.
4mh choke = 80Khz drop to 60KHz
6mh choke = 80KHZ drop to 75KHz

Operation frequency
75KHz = 20a
90KHZ = 24a
115KHz = 30a
 
Last edited:
Your FETs are only rated at 60V.
Not allowing for resonance boosting the voltage, the "opposite end" of the coil will go to double supply when one end is grounded.

A high-Q circuit could well exceed that.
 
Why do you have a 3000uF capacitor on the secondary of the 10 VAC transformer?
It should be on the other side of the rectifier. Or is it just misdrawn?

What's the purpose of the 3600 Ohm resistor?
 
Why do you have a 3000uF capacitor on the secondary of the 10 VAC transformer?
It should be on the other side of the rectifier. Or is it just misdrawn?

What's the purpose of the 3600 Ohm resistor?

LOL Someone is paying attention & its not me. :) I drew it in the wrong place. Also I forgot to draw cap & resistor in the other power supply. I don't know why eraser marks on paper show up as blank place for this new camera. My old camera was better than this new camera. Mosfets explode at 25a if I use mosfet wire #2 so I use heat sink tab as wire #2 to get 34a.

100_2155.JPG
 
Last edited:
Mosfets explode at 35 amps even though they are rated 55 amps.
there's a chart on the data sheet for the MOSFET that's labeled "Safe Operating Area" for DC and pulsed operation, the SOA is roughly triangular. since the duty cycle of the MOSFETs are roughly 50%, the SOA will be somewhat larger, and the SOA curve looks like a box with a corner cut off, but still if the devices go outside the SOA curve, they will likely self destruct.

a good explanation of MOSFET SOA is here: Texas Instruments video explaining MOSFET Safe Operating Area
 
Something is different and I don't know why? I think I learned something but I don't know what?

T3 was #12 solid copper wire, I replaced it with an identical shape coil made with #10 solid copper wire. My LCR meter shows both coils are 3.8uh

Nothing else has changed except wire size is #10 instead of #12 on T3.

PS = 25v, idle current with no load was 4a now it is 12a.

The metal test part use to heat up red hot in 7 seconds, now the same part gets about 200° hotter in 4 seconds

T3 gets so hot in a few seconds it is too hot to touch insulation is about to melt off. The other T3 with #12 wire was getting hot too but not this hot. The only reason I swapped the coils was trying too make T3 run cooler. #12 wire is rated 20a and #10 wire is rated 30a, I thought #10 wire will run cooler but it runs hotter. This is crazy??????

The rest of the circuit is wired with #12 solid copper wire it never gets hot the circuit never runs longer than 20 seconds each time.

Is this a lesson on skin effect?


100_2166.JPG


100_2169.JPG


100_2168.JPG


100_2164.JPG
 
Last edited:
12 wire is rated 20a and #10 wire is rated 30a, I thought #10 wire will run cooler but it runs hotter. This is crazy??????
Um. Induction heating works at high frequency. High frequency conducts through skin-effect. Larger diameter wire, more 'skin', more current, more heat.

You need to check the resistance at the frequency of the drive circuit.
 
Is this a lesson on skin effect?

I'd say it's more a lesson on convection. Why would you even use insulated wire? The induction heater I've seen that did have insulated coils had ceramic insulation, not PVC. And most are made from copper tube with cooling water running through them.
 
I'd say it's more a lesson on convection. Why would you even use insulated wire? The induction heater I've seen that did have insulated coils had ceramic insulation, not PVC. And most are made from copper tube with cooling water running through them.

In the past if metal parts touches the 8 turn copper wire coil mosfets explode that is why I started using insulated wire. Now I need high temperature furnace cement I can get that at Allied Boiler Supply Co. I already have a water pump and water tank for testing. I made a 1/4" copper tubing coil T3 it is physically too large so I need to make another 8T coil with 3/16" copper tubing and test it. I also need dry sand to fill tubing so I can bend it with no kinks. I use to have sand but don't know what happened to it.

Test show 3mh choke coil appears to be the best choice with no load idle current is low at 80KHz. When T3 coil is loaded 6mh chokes prevents frequency from dropper below 75KHz the other chokes don't do that. When PS is 20v current is almost steady 8a to 9a. I still need to test a 8mh choke.

1.7mh choke
15v = 5a
20v = 8a
25v = 12a

3mh choke
15v = 4a
20v = 8a
25v = 12a

4mh choke
15v = 5a
20v = 8a
25v = 12a

6mh choke
15v = 6a
20v = 9a
25v = 13a

New T3 is 1.5" diameter 2.5" long made with 1/4" copper tubing.

100_2180.JPG
 
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top