How to turn ON a Mosfet about half way or less?

[gary350]

How do I turn on a mosfet just enough so the output is 13 amps for continious operation?

Datasheet says, Mosfet STW13NB60 is rate 500 volts, 13 continious amps. I want the mosfet to be able to run for 8 hours and not over heat. 13 amps may need to be reduced to a current that allows the mosfet to run for hours and not over heat.

My idea is to use 120 vac directly from the wall outlet. After the bridge rectifier and the filters capacitor 120 x 1.414 = about 170 VDC power supply voltage. 2 mosfets are connected to make an oscillator so each mosfet is ON 50% of the time and OFF 50% of the time. Operating frequency about 60KHz.

I experemented with current limiting resistors, 20w, 50w and 70 watt resistors all get hot enough to smoke. Resistors are not practical, too massive, too hot, too much wasted power. It appears to me to solution is to limit the current another way by not turning the mosfet on all the way.

 

[tunedwolf]

That would actually make things worse. By limiting how hard you turn them on, you are in effect, creating a higher Rds, therefore dropping more voltage across the devices, which will be dissipated as heat. You might be better looking into controlling the actual on time, thereby reducing the average current through the devices. However, even at half your required 13A, you are still looking at some serious heat being generated.

 

[Ian Rogers]

I use 12 amp mosfets with a 4 amp load..... They don't get hot!!! Maybe that is what you need to do... If I run my motors a half speed ( pwm 25% on ) they still don't get hot....

 

[ronsimpson]

I don't know what your output voltage is.
Example 170 volts form rectifiers. Output is 70V so the resistors or MOSFET must drop 100 volts.
Power=100V X 13A = 1300watts.
Very hot, burn out, smoke.
It does not matter if the 1300watts is lost in resistors or in MOSFETs, it must be pushed into the air.

This in not the way to do this job. Please describe what you want to build.
110VAC in
Battery charger or power supply.
_____VDC out at ____current

 

[gary350]

I want to build a portable hand held induction heater for my Son to use at work. It may only need to be used for 5 minutes at a time but there is a chance he might need to use it 15 to 20 times in a row. My idea is, plug it in, hold it where it is needed, push the ON button on the body. Heat a bolt to only 600 degree F, dont want to remove the temper in the bolt and have the head twist off. Rusty bolts can be heated hotter.

These tools are already available $450 from the factory. I can build one, save him some $$$ and have FUN doing it.

**broken link removed**

 

[alec_t]

My idea is to use 120 vac directly from the wall outlet.

Bad idea! You risk electrocuting your son or any other user of your gadget . Use an isolating transformer.
With the power necessary for an induction heater you will have to use seriously large heat-sinks on the FETs. Alas, your hope to fit this in a D-size battery volume is a tad optimistic .

 

[crutschow]

I agree, that circuit is dangerous and needs to be isolated.
You could use an isolated line operated switching regulator to generate the DC.

 

[gary350]

I have an idea to use a SMALL isolation transformer that is small enough to saturate the core at 5 amps 25 vdc. This should limit the current to the mosfets to 5 amps with no other current limiting. If I wind the secondary coil to have 25 volts with 5 amp wire and totally saturate the core wonder if that will pull the voltage down. I am sure I will need a second tiny 12 watt transformer for 12 vdc at 1 amp for the gates. I would like to keep the hand held part small and light weight, transformers will make it heaver. The factory made unit is no larger than a D size flashlight body and every thing is inside the handle. I am trying to copy the factory model.

 

[alec_t]

I have an idea to use a SMALL isolation transformer that is small enough to saturate the core at 5 amps 25 vdc. This should limit the current to the mosfets to 5 amps with no other current limiting.

'Fraid not. If the core saturates, the transformer inductance and hence its impedance drops dramatically; so it won't limit current much.

 

[shortbus=]

Agree with Tunedwolf and Ian, this is what PWM is for. You can still use the 60kHz and dial back the off time. How about the power unit in a portable battery charger size box and the work coil that has a cord that plugs into the power supply.

 

[gary350]

What is PWM?

 

[ChrisP58]

What is PWM?

Pulse Width Modultion.

Instead of each of your mosfets being on 50% of the time, you control the power by reducing the on time.

 

[gary350]

Pulse Width Modultion.

Instead of each of your mosfets being on 50% of the time, you control the power by reducing the on time.

I had and idea this could be done but dont know what is it called. I was thinking to use a 555 timer to be ON 5% and OFF 95% and using it to drive a mosfet ON 5%. There is probably a better way than a 555 timer?

 

[ChrisP58]

While you could probably do it with a 555, you'll need some extra logic to give you two outputs with exactly the same timing.

It would be a lot better to use a dedicated PWM control IC. There are many parts from many suppliers. Here are a few to look at:
https://www.ti.com/lsds/ti/power-ma...troller-products.page#p1129=Push-Pull&p32=2;2

 

[gary350]

I know nothing about PWM. Is there a circuit drawing that shows how to make it work?

 

[Ian Rogers]

I know nothing about PWM. Is there a circuit drawing that shows how to make it work?

Alec has done a simple PWM circuit and kindly posted it on this forum..
https://www.electro-tech-online.com/articles/pwm-pulse-generator.635/

Just to give you something to get started...

 

[Tony Stewart]

maybe https://www.ti.com/lit/an/slua125/slua125.pdf might interfere with AM radio though.

 

[gary350]

I have several questions. The circuit drawing for CD4011A only shows 1 output how will that turn on/off 2 mosfets?

I can find no Datasheet for CD4049B online but I did find CD4011B but not much information other than it is a CMOS devise. Circuit shows +14v and Vcon, what is Vcon? After a 20 minute search I can not find my CMOS book so I ordered another one it should be here tomorrow. I want to learn how these devices work.

Datasheet for UC3825A says device turns on at +9v and off at +8.4v with 2 outputs A and B. It appears the PWM will work fine on 12vdc from the 7812 but I dont know how much current it needs. The PWM might need its own 7812 voltage regulator. No circuit information so maybe I can figure this out when the CMOS book arrives.

I am not sure how the PWM attaches to the circuit I already have? The original circuit both mosfet gates connect to +12v through a 240 ohm resistor, it appears gates are triggered by the zener diodes. I am using a 7812 voltage regulator to get 12vdc from 20vdc. The new circuit I have PWM outputs A and B connected to the mosfets gates, I am not sure this is correct. How does the PWM know to run at 60KHz? The LC circuit determines the Hz and the zeners switch the mosfets on/off. If the PWM takes the place of the zeners then the diodes are no longer needed. Something is wrong in this circuit?

Do the zeners in the original circuit turn the mosfets ON or do they turn the mosfets OFF?

**broken link removed**

 

[jjw]

They are not zeners, but schottky diodes and they turn the fets off.
I think the pwm frequency should be synchronized to 60kHz resonance frequency with a pll.

 

[jpanhalt]

They are not zeners, but schottky diodes and they turn the fets off.
I think the pwm frequency should be synchronized to 60kHz resonance frequency with a pll.

Last I saw, he was using N-channel mosfets. How do the diodes tied to the upper rail (20 VDC) help turn them off?

The usual design has them draining toward the driver, which hastens turn-off.
John