Maker Pro
Maker Pro

IRFP4332 250V Mosfet Rise & Fall Times

Fish4Fun

So long, and Thanks for all the Fish!
Aug 27, 2013
481
Joined
Aug 27, 2013
Messages
481
I have some IRFP4332 mosfets, and I think they might be suitable for a one-off 90Vdc motor speed controller project, but none of the datasheets I can find specify td on/off or the tr/toff other than to say,

Short Fall & Rise Times for Fast Switching

The most recent datasheet I can find was published by International Rectifier in 2009 ... and "Short" is not helping me decide if I want to use these in a 20kHz PWM switching application. Max duty cycle is 50%, and while I am confident 25uS on/off is within the device's capabilities, I would like the minimum duty cycle to be 5% to 10% (2.5uS to 5uS on time) and I am NOT confident the IRFP4332 will be able to handle these short turn on times ...

The design current is < 5A @ 170Vdc ... and overall usage should rarely exceed 30 minutes/hour.

Another approach would be to "PWM the PWM" ... That is have a fixed 20kHz 50% PWM that is enabled/disabled @ perhaps a 250hz PWM frequency ... This would keep the "switching frequency" @ 20khz, 50% duty cycle ... but only enable it in 4mS increments ... each increment would enable another 80 x 20Khz pulses to the load ... thus 100% duty cycle of the 250hz PWM would still only be a 50% duty cycle to the load ...

Open to suggestions or thoughts!

Thanks in advance!

Fish
 

Harald Kapp

Moderator
Moderator
Nov 17, 2011
13,720
Joined
Nov 17, 2011
Messages
13,720
The datasheet can't tell you rise and fall times as these depend widely on the circuit used. Switching time is largely influenced by the time it takes to charge and discharge the gate.
The datasheet specifies a max. total gate charge of 150 nC @ Vgs = 10 V and an input capacitance of typ. 5860 pF, let's round that up zo 6 nF.
Let's assume you want the switching time to be max. 10 % of the min. on time, then we have Tr <= 0.1 ×2.5 µs = 0.25 µs.
With these values we have from
[math] C = \frac{I × t}{V}[/math][math] I = \frac{C × V}{t} = \frac{6 nF × 10 V}{0.25 µs} = 0.24 A[/math].
So for fast switching your gate driver needs to supply ~250 mA. Just a rough calculation...
 

Fish4Fun

So long, and Thanks for all the Fish!
Aug 27, 2013
481
Joined
Aug 27, 2013
Messages
481
WoW, Great responses!

I am using a TC4420 Low Side Driver (pdf attached) specs state up to 6A Drive Current, Tr & Tf ~ 35nS ... But in a quick prototype using a current limited 30V Drain to Source Voltage, the TC4420 got quite warm, and a scope trace on the IRFP4332 drain showed garbage while the gate voltage trace was very clean ... I was just assuming the IRFP4332 could not handle the switching, but it sounds like I need to look more carefully at what is going on ... I will try a constant PWM of 50% duty cycle on a resistive load (previous was an inductive load) and see what happens ... will worry about the variable speed control when the mosfet/LS Driver are sorted ...

Again, Thanks!

Fish
 

Attachments

  • TC4420 Mosfet Driver.pdf
    360.7 KB · Views: 1

crutschow

May 7, 2021
848
Joined
May 7, 2021
Messages
848
Do you have a diode across the motor to carry the motor current when the MOSFET is off?
Post a schematic of what you have.
 

Fish4Fun

So long, and Thanks for all the Fish!
Aug 27, 2013
481
Joined
Aug 27, 2013
Messages
481
@crutschow ... here is a current schematic of the project ... honestly I had just been working from a scrap of paper, but I have made enough stupid little mistakes already that I went ahead and did a proper schematic ... The primary difference between the schematic and my current prototype is that I made the prototype PCBs in functional blocks .... that is the 5V regulator and the ATMEGA8 are on a small PCB by itself ... the Line Voltage Rectifier & filter capacitor are on a different small PCB and the TC4420, IRFB4332 and other switch related items are on a third small PCB. (And Just to make things fun the very large DPDT Direction Switch is still mounted to the device the speed control is for.)

As a general rule I try to avoid line voltage projects that do not have galvanic isolation, but in this particular case I don't really have a choice ... the motor is a 1/4HP 90Vdc motor and is part of an important tool used every day at my business ... To replace the tool would be expensive, and finding OEM parts for the tool is problematic ... the tool was only in very limited production for ~ 3 years in the late 1990's. I am aware that a quick Google search will display dozens of 90Vdc motor speed controls , but there are two caveats that have prevented me from just ordering one ...

1) The speed control needs to fit inside the tool housing.

2) The speed control has to be controlled by a foot pedal.

So I am re-inventing the wheel and so far have found that squares and triangles are NOT suitable wheels, LoL.

What I need to do is use a 9Vdc motor and the rectified output from a 12V transformer and get the logic & PWM figured out before I do any more testing using line rectified voltage .... but I was convinced I could get it right on the first try, LoL .... However, that didn't work out.

Thanks!

Fish
 

Attachments

  • LineWinder SC Schematic.jpg
    LineWinder SC Schematic.jpg
    128 KB · Views: 4
Top