2 MOSFET's in series for PFC circuit.

Thread Starter

TAKYMOUNIR

Joined Jun 23, 2008
352
I need to use 2 MOSFET(600 V each) in series , instead of using higher voltage one(1200V), so how I make sure that will turn off and on in the same time.
 

ScottWang

Joined Aug 23, 2012
7,397
You can refer to the circuit as below, the +12V2 and Mid is a another isolation power supply for the upper mosfet, the time delay could be happened in the Vgs, you need to through the testing and measuring to know how the affecting situation, maybe the delay time is too short and doesn't affecting anything.

If you need a 5A output then you should find a mosfet has 20A output is better, at least has 15A, the another issues are the Vgs and Id, when the Vgs=18V then Id =3A, that was shown in the SCT2450KE datasheet of your original mosfet.


2 MOSFETs in series_TAKYMOUNIR_ScottWang.gif
 

Bordodynov

Joined May 20, 2015
3,177
I have the same idea with you, but I'm not sure what kind of the transformer can buy, are you should the transformer can buy online or somewhere or you just provided an idea?
The transformer can be made on the ferrite ring. The parameters for the calculation, I took from my own experience. Optimization to reduce the size I did not do. In Scheme Hc=5 Bs=430m Br=100m and A=28mm^2 Lm=62m. Ferrite can take other parameters. I have approximate dimensions.
In this way, and can include three transistors.
 

Thread Starter

TAKYMOUNIR

Joined Jun 23, 2008
352
You can refer to the circuit as below, the +12V2 and Mid is a another isolation power supply for the upper mosfet, the time delay could be happened in the Vgs, you need to through the testing and measuring to know how the affecting situation, maybe the delay time is too short and doesn't affecting anything.

If you need a 5A output then you should find a mosfet has 20A output is better, at least has 15A, the another issues are the Vgs and Id, when the Vgs=18V then Id =3A, that is shown in the SCT2450KE datasheet of your original mosfet.

View attachment 92992
Thanks so much you mean by the problem the RDS.
 

Thread Starter

TAKYMOUNIR

Joined Jun 23, 2008
352
You can refer to the circuit as below, the +12V2 and Mid is a another isolation power supply for the upper mosfet, the time delay could be happened in the Vgs, you need to through the testing and measuring to know how the affecting situation, maybe the delay time is too short and doesn't affecting anything.

If you need a 5A output then you should find a mosfet has 20A output is better, at least has 15A, the another issues are the Vgs and Id, when the Vgs=18V then Id =3A, that was shown in the SCT2450KE datasheet of your original mosfet.


View attachment 92992
what will happen if the upper FET (FET1) turn first, in this case Vgs wiil be( 12-1200) so this will blow up the mosfet or will turn it off?
 

Thread Starter

TAKYMOUNIR

Joined Jun 23, 2008
352
The transformer can be made on the ferrite ring. The parameters for the calculation, I took from my own experience. Optimization to reduce the size I did not do. In Scheme Hc=5 Bs=430m Br=100m and A=28mm^2 Lm=62m. Ferrite can take other parameters. I have approximate dimensions.
In this way, and can include three transistors.
Thanks so much , can you please explain how you get this number
 

ScottWang

Joined Aug 23, 2012
7,397
what will happen if the upper FET (FET1) turn first, in this case Vgs wiil be( 12-1200) so this will blow up the mosfet or will turn it off?
The Vgs of upper mosfet only have 0V and 12V, that's why it needs a isolation power supply, if only anyone of the mosfet turn on then it could be blow up, so that's why it needs measuring and testing before connecting the load and HV power, I was thought about to using an AND gate logic conception to avoid this situation, but not have result yet.
 

ScottWang

Joined Aug 23, 2012
7,397
I'm not sure why you added that three parts at the Vgs, if you adding the optocoupler at the Vgs with both upper and lower mosfets then the Vgs will get the signal at the same time, so two mosfets should be turn on at the same time too.
 

ian field

Joined Oct 27, 2012
6,536
I need to use 2 MOSFET(600 V each) in series , instead of using higher voltage one(1200V), so how I make sure that will turn off and on in the same time.
In the early days of solid state colour TV, the transistors available (BU205) wasn't quite up to the needed voltage rating - so most manufacturers used 2 in series. Each was driven by separate secondaries on a common driver transformer. Each base had a tuneable series inductor so the drive could be balanced.

Something along the same lines is probably possible with MOSFETs.
 

GS3

Joined Sep 21, 2007
408
When switching two elements in series like that it is common to put a resistor and capacitor in parallel with each so they form a voltage divider and protect the switches because if not the least difference in switching time means one switch will be destroyed by overvoltage (and the other one will probably follow).
 

ian field

Joined Oct 27, 2012
6,536
When switching two elements in series like that it is common to put a resistor and capacitor in parallel with each so they form a voltage divider and protect the switches because if not the least difference in switching time means one switch will be destroyed by overvoltage (and the other one will probably follow).
A crude way of preventing that is a zener clamp from drain to gate, as soon as the drain voltage reaches the zener voltage; it pulls the gate up and the drain clamps any further voltage rise.

Of course its not easy to get zeners with that sort of voltage rating - although I believe there's some types specially for car electronic ignition transistors that might be in the right voltage ball-park.

MOSFET gates probably have more parameter spread than bipolar B/E junctions. The usual trick is to swamp any differences with external components - in this case it would be mostly capacitance. Could be a bit of a balancing act if fast switching is required!
 

Bordodynov

Joined May 20, 2015
3,177
I made the optimization of its gate drive circuit with a transformer. I took into account the leakage inductance of the transformer. Changed (reduced 47uF ---> 47nF) capacity. The number of turns reduced to 5.
See:

TransGate.png
 
Top