driver for HiPerFET module

If I wanted to drive 2 or 3 of these modules in parallel, what kind of driver would I need?
thanks

How fast do you want to drive? 60hz or 100khz? It looks like you need 10 volts for the gate.

If you check out the spec for Qg you can get an idea of the time to switch (in ns) with 1 amp of gate current. More gate current faster switch time. So say a 9 amp driver and 3 in parallel. This would get you a switch time of around 1 us.
There is a lot of power dissipated in the module when it is half on and half off. That's why it is important to know how often it will be switching.

look up data on any of these;
FAN3227,
IXD604, IXDI604, IXDD604,
LM5111, LM5112
Look for application notes on "gate drive".

ronv said:

If you check out the spec for Qg you can get an idea of the time to switch (in ns) with 1 amp of gate current. More gate current faster switch time. So say a 9 amp driver and 3 in parallel. This would get you a switch time of around 1 us.
There is a lot of power dissipated in the module when it is half on and half off. That's why it is important to know how often it will be switching.

Click to expand...

Ok, so Qg is 2750nC. with 1A and one module, it would take 2750nS to switch? And with 9A driver, 2750nC, it would be 305nS? and then with 3 in parallel, 9A it would be 916nS (1uS)?

ronsimpson said:

How fast do you want to drive? 60hz or 100khz? It looks like you need 10 volts for the gate.

Click to expand...

I was thinking 20khz, just to be outside the audible frequency range

You can use 3 gate drive ICs.
There are some application notes on (not connecting gates together). The oscillation that may happen is very fast and you might not see it. and Three grate drivers will increase your turn on/off speed.

strantor said:

Ok, so Qg is 2750nC. with 1A and one module, it would take 2750nS to switch? And with 9A driver, 2750nC, it would be 305nS? and then with 3 in parallel, 9A it would be 916nS (1uS)?

Click to expand...

Right you are.
Are you sure you need so much power? Maybe so if you are going to do a 400HP racer?

haha, Yes, that's what my mind fancied the other day when I posted this. I have since come down to earth and realized that without any practical experience, all I am going to do is blow expensive stuff up. I'm now trying to learn just enough theory to feel comfortable in the purchase of some (*much) lower rated MOSFETs, drivers, capacitors, uController, et. al. to make a hobbyist scale controller. The math between driver and MOSFET is the same I assume for a smaller scale. In my small scale controller I plan to parallel maybe 5 or 6 MOSFETs, hopefully off the same driver and start trying to make some PWM come out of them. I have read that they will need to have gate resistors, and that will effect the switching time. I don't know the logic behind selecting a gate resistor value though

ronsimpson said:

You can use 3 gate drive ICs.
There are some application notes on (not connecting gates together). The oscillation that may happen is very fast and you might not see it. and Three grate drivers will increase your turn on/off speed.

Click to expand...

Will dedicated gate drivers ensure that all my MOSFETs switch on and off together? I am changing my plan to use 5 or 6 smaller MOSFETs and I have read that trying to get them all to turn on/off at the same time is tricky business.

Probably a good idea to do a lower power prototype. The smoke from a $2 FET smells about the same as a $1000 one. It's a complex project and some things are bound to go wrong along the way. Your right, most of the principals are the same.
Find a driver that is big enough to drive all your parallel FETS at a good rate (say less than a µs). The gate resistor helps reduce "ringing" as the FET switches. The down side is that it reduces the available current to turn the FET on and off. There should not be a problem driving several FETs from one good driver but you do need to be careful to keep the traces short from driver to gate and source back to driver. One other thing to consider when looking for a driver is "shoot thru". FETs always seem to turn on faster than they turn off. So if you have a H-Bridge configuration where the FETs clamp the inductive kick it is easy to have both FETs on the same side of the bridge on at the same time for a short time. Some drivers have a delay that can be set to prevent this. You can also "play" with the gate drive design to reduce this but it's nice to know you can't have both of them on at the same time. It's nice to have a scope to look at things like this since in real life they never seem to be as fast as the data-sheet.

ronv said:

Right you are.
Are you sure you need so much power? Maybe so if you are going to do a 400HP racer?

Click to expand...

What about the turn on delay time and rise time? Are those in addition to this thumb rule?
For example, a IRFP3206PbF has a Qg of 170nC. If I used 10A of current, according to the thumb rule, turn on time would be 17nS
The turn on delay time is 19nS and the rise time is 82ns, for a total of 101nS. is this 101nS in addition to the 17nS?

EDIT: and whatever the reply, how do I also factor in the delay caused by the gate resistor?

ronv said:

Probably a good idea to do a lower power prototype. The smoke from a $2 FET smells about the same as a $1000 one. It's a complex project and some things are bound to go wrong along the way. Your right, most of the principals are the same.
Find a driver that is big enough to drive all your parallel FETS at a good rate (say less than a µs). The gate resistor helps reduce "ringing" as the FET switches. The down side is that it reduces the available current to turn the FET on and off. There should not be a problem driving several FETs from one good driver but you do need to be careful to keep the traces short from driver to gate and source back to driver. One other thing to consider when looking for a driver is "shoot thru". FETs always seem to turn on faster than they turn off. So if you have a H-Bridge configuration where the FETs clamp the inductive kick it is easy to have both FETs on the same side of the bridge on at the same time for a short time. Some drivers have a delay that can be set to prevent this. You can also "play" with the gate drive design to reduce this but it's nice to know you can't have both of them on at the same time. It's nice to have a scope to look at things like this since in real life they never seem to be as fast as the data-sheet.

Click to expand...

Good thing I have a scope, but not too experienced at using it. It's a tektronix 2440(?) I believe, from the early 90's. I have a 10X probe; is there any voltage above which I shouldn't use it? I've read connecting it to mains 120V is a no-no. I'm not using a H-bridge; I have seen in other schematics a 10ohm resistor used as a gate resistor; is this a good value? how do you decide a gate resistor value?

matching drivers to MOSFETs

I found this application Note which is finally shedding some light on how to select the right driver for my MOSFET.
My Mosfet has a Ciss of 6540pF and I want to run PWM @ 20KHz.
I want to drive 6 of them in parallel, and still not sure if I will need 6 drivers or only one.
Anyway, the Application note has a nice chart for selecting one of their Drivers for a given Ciss, but I was thinking of going with a driver of a different brand, which has a max output of 11.4A.
The application note says at the bottom:

Quote:
To match any MOSFET to its proper driver, use the chart above
(which will take care of the largest number of applications), or use
the simple formula: rise time (dt) = driver supply voltage (dV),
times capacitance (C), all divided by driver peak current (I);
restated:
dt = [(dV)xC]/I

I assume that dV in this formula is Vdd in the driver?
So, using the max (for the FAN3121 driver) of 18V, that would be my dv?
dt = (18V * .000,000,006,540)/11.4 = 10.3nS?
seems awefully low. too good to be true? I'm not sure I'm doing this right... plus I need to add a gate resistor. Lets say I add a 10Ω resistor (do I really need to go all the way to 10Ω?), which is going to limit current to 1.8A
then
(18V * .000,000,006,540)/1.8A = 1.1772μS?
That made a big difference. Not sure what size gate resistor to use. the application note shows in figure 4 2.2-22Ω; I would assume you would want to go with the smallest possible, but not sure.

Also I know I read somewhere that frequency plays a part in this, but the equation did not include frequency. AAAHH I don't know what's going on! (pulling my hair out)

Should I just go with the TC4422 that the application note suggests (10,000pF)?

It's really not so important. They never behave like the spec. sheet when you get them in the circuit anyway. The data sheet will usually show the switch times with a gate resistor and 10 volts on the gate. (There is little advantage to a gate voltage over 10 volts.) Since Qg is switch time with a constant 1 amp I usually just use half the peak gate current. Ton=1/2 I gate X Qg. So in your case they spec a 2.7 ohm gate resistor. So 3.7 amps peak/2= 1.85. Qg is 170/1.85=92ns. Compare this with there switch times with the 2.7 ohm resistor of 92 ns. Pretty close I guess. When you get it on a board with traces and stuff it may be twice as slow - enter the scope. I would use a single driver and say 7.5 ohm gate resistors for each FET.
Speaking of gate drivers. I'm guessing you are building a H-Bridge or a Half-Bridge. This will require a high side driver as well as a low side driver. So you might want to look for at least 1/2 bridge drivers.
PS. your scope with 10X probes should work at your voltages.

ronv said:

It's really not so important. They never behave like the spec. sheet when you get them in the circuit anyway. The data sheet will usually show the switch times with a gate resistor and 10 volts on the gate. (There is little advantage to a gate voltage over 10 volts.) Since Qg is switch time with a constant 1 amp I usually just use half the peak gate current. Ton=1/2 I gate X Qg. So in your case they spec a 2.7 ohm gate resistor. So 3.7 amps peak/2= 1.85. Qg is 170/1.85=92ns. Compare this with there switch times with the 2.7 ohm resistor of 92 ns. Pretty close I guess. When you get it on a board with traces and stuff it may be twice as slow - enter the scope. I would use a single driver and say 7.5 ohm gate resistors for each FET.
Speaking of gate drivers. I'm guessing you are building a H-Bridge or a Half-Bridge. This will require a high side driver as well as a low side driver. So you might want to look for at least 1/2 bridge drivers.
PS. your scope with 10X probes should work at your voltages.

Click to expand...

In all 10's of times I've looked over the datasheet, I missed that (gate resistor) every time! Thanks! Do you think that means they determined that 2.7Ω is the minimum required to prevent ringing?
Why do you recommend 7.5Ω? I'm sure it's personal experience, but can you explain? why would the 2.7Ω from the datasheet be insufficient? (I think) I want to make it switch as fast as possible to minimize loss.
Regarding

I usually just use half the peak gate current. Ton=1/2 I gate X Qg.

Click to expand...

is this (the 1/2 gate current) your own thumb rule for accounting for real world losses and the circuit board, or is there a mathematical reason for this?

Thank you very much, you have been very helpful!