Here is the mosfet driver stage I'm about to implement to drive an inductive load at 24V and a maximum of 10 amps. The pulse input has bee simplified, but the final device will contain a TTL totem device driving the mosfet. Operating frequency will be a maximum of 500kHz.

$ 1 4.9999999999999996E-6 10.20027730826997 43 5.0 50
f 256 256 304 256 0 1.5
d 336 272 336 240 1 0.805904783
d 304 112 304 160 1 0.805904783
w 336 240 336 192 0
w 304 272 304 304 0
w 336 304 336 272 0
g 320 320 320 336 0
w 336 304 320 304 0
w 304 304 320 304 0
w 320 304 320 320 0
w 336 112 320 112 0
w 320 112 304 112 0
R 176 256 144 256 0 2 40.0 5.0 2.5 0.0 0.5
w 176 256 256 256 0
l 352 80 448 80 0 0.01 -2.525026118406124E-7
R 528 64 528 32 0 0 40.0 24.0 0.0 0.0 0.5
w 528 80 528 64 0
w 320 112 320 80 0
w 320 80 352 80 0
w 448 80 528 80 0
r 256 272 256 320 0 500.0
w 320 320 256 320 0
w 256 272 256 256 0
x 372 51 425 57 0 24 Load
x 351 266 381 272 0 24 D2
x 238 224 268 230 0 24 U1
x 234 147 264 153 0 24 D1
w 336 192 336 112 0
w 304 160 304 240 0
x 181 305 211 311 0 24 R1
o 14 128 0 43 2394.524282602951 0.37414441915671115 0 -1
o 1 128 0 33 1280.0 0.4 1 -1


D1 is a SBL1030 schottky diode for reverse current protection.
D2 is a UF4006 fast recovery rectifier to allow the inductor to freewheel. It is included to prevent the body diode in the mosfet from activating.
U1 is a FQP9N50C mosfet.
R1 is used as a basic pull-down resistor and to provide a minimum load to the pulse generator.
There will also be a 10 amp fast blow circuit breaker included in the built circuit.

Any ideas or suggestions? I'm getting tired of frying output mosfets.

Views: 86

Replies to This Discussion

Hi James
We use mosfets to drive motors ect on a regular basis, usually only 20 khz.
I see a list of numbers, can you post dwg?
We usually incorportate half bridge drivers to get fast switching & high side isolation
also zener on gate to clamp inductive kick back from capacitor coupling in fet.

Regards Peter
The numbers are copy/paste for the circuit simulator applet linked on this group's home page. The simulator has issues with AC simulation, but works fairly well for DC or low frequency AC.

I chose a single mosfet rather than a half or full bridge for simplicity and cost effectiveness. The load is non-tapped, so just letting it freewheel felt like the right way to go.

Thank you for the zener idea. I'll have to see if it would be useful in this application.
Thanks for showing me the cute simulator. I would not pwr up this circuit as the fet would likely blow up.
When the fet turns Off the anode of D1 will go very high, likely several hundred volts & break it down, also the gate cap will cause the fet to turn on unexpectantly. D2 should go directly across the S-D & could be a zener with less volts than fet, a diode across the coil will clamp the inductive kick back, the gate resistor could be much lower like 100 ohm or less, a zener like 1N5350 will save the gate. the gate drive should be capable of several amps or the gate cap will cause slow rise generating heat. the driver chip is aprox $1, problem is SMD part & hard to do point to point wiring. I just finished a PCB for
Solar Car BMS, Google Xof1, I added a 3 phase driver to turn a car alternator into brushless motor. see Discovery link
http://watch.discoverychannel.ca/daily-planet/april-2010/daily-plan...
I'm behind car to keep wind from closing it down on Jay
Have fun Peter
Soldering is my specialty. I solder high density, fine pitch boards all day long with 0402 components. Everything I do must meet IPC class 3 standards.

The device driving the gate of the mosfet supplies up to an amp of current, so I don't think the 1030pF of capacitance the gate junction has will give too much of a slew rate. D1 will not get any appreciable voltage since D2 is in forward biased when D1 is reverse biased. D1 is there to keep the body diode built into the mosfet from going into conduction. The recovery rate would increase the time needed between pulses for the mosfet so that heat generation would not be an issue.

I'm going to add a 5.5V zener in parallel with R1 to protect the gate.
Congrats on your soldering skills, I don't believe D2 will conduct, ever " unless it is zener "
the voltage will be positive on D1, way way above 24v, The usual gate test impedence is 50 ohm
let us know of your success.
never let the magic smoke out
Peter

James Morehead said:
Soldering is my specialty. I solder high density, fine pitch boards all day long with 0402 components. Everything I do must meet IPC class 3 standards.

The device driving the gate of the mosfet supplies up to an amp of current, so I don't think the 1030pF of capacitance the gate junction has will give too much of a slew rate. D1 will not get any appreciable voltage since D2 is in forward biased when D1 is reverse biased. D1 is there to keep the body diode built into the mosfet from going into conduction. The recovery rate would increase the time needed between pulses for the mosfet so that heat generation would not be an issue.

I'm going to add a 5.5V zener in parallel with R1 to protect the gate.
Please excuse my ignorance, but I don't understand why D2 will never conduct.
Peter you can import his numbers above and tweek his design then export the new numbers and repost them here. Read the how to use the group discussion it gives more detail.
Cathode of D2 will never see a negative voltage with respect to it's anode, = no conduction
In this configuration, when fet is turned Off, the voltage is in series from the positive inductive kick back
through forward bias D1 into fet. D1 is redundant unless across SD
The best thing to do is put diode across coil

James Morehead said:
Please excuse my ignorance, but I don't understand why D2 will never conduct.
I just realized I was looking at the kickback voltage with the wrong polarity. Thank you for your patience Peter. Here is what I think will be the final circuit:

$ 1 4.9999999999999996E-6 10.20027730826997 43 5.0 50
f 256 304 304 304 0 1.5
d 336 320 336 288 1 0.805904783
d 304 160 304 208 1 0.805904783
w 336 288 336 240 0
w 304 320 304 352 0
w 336 352 336 320 0
g 320 368 320 384 0
w 336 352 320 352 0
w 304 352 320 352 0
w 320 352 320 368 0
w 336 160 320 160 0
w 320 160 304 160 0
R 176 304 144 304 0 2 40.0 5.0 2.5 0.0 0.5
w 176 304 256 304 0
l 352 128 448 128 0 0.01 -0.30499605301750193
R 528 112 528 80 0 0 40.0 24.0 0.0 0.0 0.5
w 528 128 528 112 0
w 320 160 320 128 0
w 320 128 352 128 0
w 448 128 528 128 0
r 256 320 256 368 0 500.0
w 320 368 256 368 0
w 256 320 256 304 0
x 372 99 425 105 0 24 Load
x 356 307 372 310 0 12 D2
x 273 271 289 274 0 12 U1
x 268 188 284 191 0 12 D1
w 336 240 336 160 0
w 304 208 304 288 0
x 268 361 284 364 0 12 R1
d 352 144 448 144 1 0.805904783
w 448 144 448 128 0
w 352 144 352 128 0
z 224 368 224 304 1 0.805904783 5.5
w 224 368 256 368 0
x 185 343 208 346 0 12 ZD1
x 392 172 408 175 0 12 D3
o 14 128 0 43 149.65776766268445 0.37414441915671115 0 -1
o 1 64 0 35 40.0 9.765625E-5 1 -1


D2 and D3 will be the same type of ultra fast recovery rectifier.

Thank you again Peter. I will follow-up with my test results from the assembled circuit.
Hey James I think you have a bad connection with your Zener Diode an it could blow so you would have to remove it an the circuit would be fine!!
Glad to see D3, this is most important.
Matthew is correct about bad connection on ZD1
I would still remove D1 to reduce losses & D2, the fet body diode is reversed biased so it's speed is not important as it is not in circuit

The gate driver should be added, it seems like too much voltage for zener, is this logic fet ?

What do you intend to do with coil ?

Peter

James Morehead said:
I just realized I was looking at the kickback voltage with the wrong polarity. Thank you for your patience Peter. Here is what I think will be the final circuit:

$ 1 4.9999999999999996E-6 10.20027730826997 43 5.0 50
f 256 304 304 304 0 1.5
d 336 320 336 288 1 0.805904783
d 304 160 304 208 1 0.805904783
w 336 288 336 240 0
w 304 320 304 352 0
w 336 352 336 320 0
g 320 368 320 384 0
w 336 352 320 352 0
w 304 352 320 352 0
w 320 352 320 368 0
w 336 160 320 160 0
w 320 160 304 160 0
R 176 304 144 304 0 2 40.0 5.0 2.5 0.0 0.5
w 176 304 256 304 0
l 352 128 448 128 0 0.01 -0.30499605301750193
R 528 112 528 80 0 0 40.0 24.0 0.0 0.0 0.5 begin_of_the_skype_highlighting              0 0 40.0 24.0 0.0 0.0 0.5      end_of_the_skype_highlighting
w 528 128 528 112 0
w 320 160 320 128 0
w 320 128 352 128 0
w 448 128 528 128 0
r 256 320 256 368 0 500.0
w 320 368 256 368 0
w 256 320 256 304 0
x 372 99 425 105 0 24 Load
x 356 307 372 310 0 12 D2
x 273 271 289 274 0 12 U1
x 268 188 284 191 0 12 D1
w 336 240 336 160 0
w 304 208 304 288 0
x 268 361 284 364 0 12 R1
d 352 144 448 144 1 0.805904783
w 448 144 448 128 0
w 352 144 352 128 0
z 224 368 224 304 1 0.805904783 5.5
w 224 368 256 368 0
x 185 343 208 346 0 12 ZD1
x 392 172 408 175 0 12 D3
o 14 128 0 43 149 begin_of_the_skype_highlighting              14 128 0 43 149      end_of_the_skype_highlighting.65776766268445 0.37414441915671115 0 -1
o 1 64 0 35 40.0 9.765625E-5 1 -1


D2 and D3 will be the same type of ultra fast recovery rectifier.

Thank you again Peter. I will follow-up with my test results from the assembled circuit.
D1 and D2 are there to isolate the fet's body diode to eliminate any possibility of it going into conduction. I have several projects I'm planning to use this driver with; the first of which will be for driving an automotive ignition coil. After that it will be used to excite a stator, then a few other off the wall things. Here is the corrected circuit:

$ 1 4.9999999999999996E-6 10.20027730826997 43 5.0 50
f 256 304 304 304 0 1.5
d 336 320 336 288 1 0.805904783
d 304 160 304 208 1 0.805904783
w 336 288 336 240 0
w 304 320 304 352 0
w 336 352 336 320 0
g 320 368 320 384 0
w 336 352 320 352 0
w 304 352 320 352 0
w 320 352 320 368 0
w 336 160 320 160 0
w 320 160 304 160 0
R 176 304 144 304 0 2 40.0 2.5 2.5 0.0 0.5
l 352 128 448 128 0 0.01 -0.06742360382936634
R 528 112 528 80 0 0 40.0 24.0 0.0 0.0 0.5
w 528 128 528 112 0
w 320 160 320 128 0
w 320 128 352 128 0
w 448 128 528 128 0
r 256 320 256 368 0 500.0
w 320 368 256 368 0
w 256 320 256 304 0
x 372 99 425 105 0 24 Load
x 356 307 372 310 0 12 D2
x 273 271 289 274 0 12 U1
x 268 188 284 191 0 12 D1
w 336 240 336 160 0
w 304 208 304 288 0
x 268 361 284 364 0 12 R1
d 352 144 448 144 1 0.805904783
w 448 144 448 128 0
w 352 144 352 128 0
z 224 368 224 304 1 0.805904783 5.5
w 224 368 256 368 0
x 185 343 208 346 0 12 ZD1
x 392 172 408 175 0 12 D3
w 176 304 224 304 0
w 224 304 256 304 0
o 13 128 0 43 37.41444191567111 0.37414441915671115 0 -1

RSS

*** Translate HODINFO ***

© 2018   Created by gabet123.   Powered by

Badges  |  Report an Issue  |  Privacy Policy  |  Terms of Service

Live Chat