victor 885 parts

[Al Skierkiewicz]

Cirrus,
I believe there is only one voltage regulator, the other things you see are drive transistors for the FETs. You are correct on the the H bridge design, this is very common in motor controls. If you look closely you will also see the equalizing resistors in series with each FET. These help to equalize the current each shares with the other two in it's string. There are very low resistance so measuring them with a normal DVM won't tell you if they are shorted. The values of these resistors are about the same as the resistance of the leads on the DVM.
Remember that depending on the intial failure of the FET, other devices connected to it may also have been damaged. For instance, a short from gate to drain may have attempted to draw full motor current through the drive transistor until it failed. Power supply traces on the circuit board may also have been damaged by the excessive currents. Use a bright light and a magnifying lens when checking for damage. Post pictures and questions if you have them. I love to see stuff that is blown up.

[Cirrussport]

Actually the one i was looking at last night was quite blown. Everything else on the board seemed fine except one resistor that i couldnt get a reading thru. but for the most part everything but the fets were fine i believe. I have about 4 blown fets on this one. i pulled one fet off the board already just for curiousity. out of the others that are blown i see one with just a small burn mark on the front casing but not actually blown apart. 3 that the casing split. and one more that i didnt count as burnt, altho it has to be because the gate and the drain have continuity.

Originally Posted by Al Skierkiewicz

If you look closely you will also see the equalizing resistors in series with each FET. These help to equalize the current each shares with the other two in it's string.

The resistors that go to the fets are on the supply then? I figured they were on each gate to make sure that the gates were getting the right voltage. Well victors seem very well engineered. looking at the board (this is just theory, correct me if im wrong) you could technically solder on 3 more fets to the bottom of the board for each string for a total of 24 on board fets aslong as you put a separate resistor for each gate.(i know its not very practical but in theory it would work right?)

Thanks again

[Richard Wallace]

Originally Posted by Al Skierkiewicz

... I love to see stuff that is blown up.

I do too. Can't wait for these pictures.

[Al Skierkiewicz]

Originally Posted by Cirrussport

The resistors that go to the fets are on the supply then? I figured they were on each gate to make sure that the gates were getting the right voltage. Well victors seem very well engineered. looking at the board (this is just theory, correct me if im wrong) you could technically solder on 3 more fets to the bottom of the board for each string for a total of 24 on board fets aslong as you put a separate resistor for each gate.(i know its not very practical but in theory it would work right?)

Thanks again

The resistors would normally be in the drain lead. It has been a while since I opened one up, but I seem to remember the gates and sources tied in parallel while the resistors are in series with the drain and then out to the load for the high side FETs and in the source for the low side.
Your idea about added FETs is correct, theoretically adding as many FETs as are needed to make up the current. Some factors might get in the way of success such as turn on time and source to drain "ON" resistance but for the low switching frequency of the Victor the turn on time should not be a problem. Of course the circuit traces are sized for the expected current so you might have to redesign the way power and output leads are treated.

All in all, they are pretty cool little devices for $129.

[Cirrussport]

Sorry guys that i havent got the pictures up yet.Ive been having major problems with my camera. But out of curiousity. I noticed researching h-bridges that they usually use both p and n channel mosfets. but the victors only use n channel.
Does anyone know the advantages/disadvantages of this?

Also can anyone give me a link or explain how to build all n-channel mosfets into a circuit?

Thanks Again

Craig

[Richard Wallace]

Originally Posted by Cirrussport

Sorry guys that i havent got the pictures up yet.Ive been having major problems with my camera. But out of curiousity. I noticed researching h-bridges that they usually use both p and n channel mosfets. but the victors only use n channel.
Does anyone know the advantages/disadvantages of this?

Also can anyone give me a link or explain how to build all n-channel mosfets into a circuit?

Thanks Again

Craig

P-channel devices generally cost more and operate less efficiently than N-channel devices.

To control an all N-channel bridge, the gates of the high-side devices must be driven with respect to the motor output terminals, because that's where their sources are connected. So the gate drive circuit must include a power supply that "floats" on the motor terminal. This is often implemented using a special purpose driver IC such as IR2104.

[Cirrussport]

So basically I need to look at whatever voltage is usually needed on the low side to turn on the mosfet. then take the supply voltage and increase it by this number to run the gate.

For example.
If it takes 8v to turn on the mosfet on the low side.
And my source voltage is 18v

Then i need to run the upper gates with the sum of those voltages.

Required gate voltage(top side) = supply voltage + Required gate voltage(bottom side).

And that ic that was mention is im guessing used to help create this extra voltage?

I think i got it but i may be totally wrong.

Thanks

Craig

[Richard Wallace]

Originally Posted by Cirrussport

So basically I need to look at whatever voltage is usually needed on the low side to turn on the mosfet. then take the supply voltage and increase it by this number to run the gate.

For example.
If it takes 8v to turn on the mosfet on the low side.
And my source voltage is 18v

Then i need to run the upper gates with the sum of those voltages.

Required gate voltage(top side) = supply voltage + Required gate voltage(bottom side).

And that ic that was mention is im guessing used to help create this extra voltage?

I think i got it but i may be totally wrong.

Thanks

Craig

You've misunderstood my use of the word "source". I meant it to indicate one of the terminals of a FET, not a voltage supply. I recall being similarly confused when I first encountered that term, as a sophomore EE student about thirty years ago.

A FET has three terminals called source, gate, and drain. To turn it on [i.e., to get it to conduct with a low resistance called R_DS(on)] you apply about 10V between the gate and source terminals. Please refer to the IRL3103 datasheet, conveniently hosted by FRC 358.

So each of the gate drive circuits for high-side devices in an N-channel FET bridge must be referenced to the respective source terminal. This is accomplished by establishing charge on a capacitor by means of the special purpose IC mentioned in my earlier post.

[Gdeaver]

Many FET driver IC's need a resistor between the driver output and gate to dampen transients and ringing. This is especially important at higher switching frequencies. These maybe the small resistors you see on the board.

[Richard Wallace]

Originally Posted by Gdeaver

Many FET driver IC's need a resistor between the driver output and gate to dampen transients and ringing. This is especially important at higher switching frequencies. These maybe the small resistors you see on the board.

Yes. Resistors for that purpose, and to limit dv/dt during switching, are indicated in the typical connection diagram at the bottom of the first page of the IR2104 datasheet, linked in my earlier post.

[Cirrussport]

http://www.chiefdelphi.com/forums/at...tid=4638&stc=1
I know this isnt a picture of the blown up parts. this picture only shows where I removed one Mosfet.
Im still having camera problems so this is a pic from a few weeks ago.
Sorry for the delay. Im workin on more pics

[Al Skierkiewicz]

Originally Posted by Cirrussport

http://www.chiefdelphi.com/forums/at...tid=4638&stc=1
I know this isnt a picture of the blown up parts. this picture only shows where I removed one Mosfet.
Im still having camera problems so this is a pic from a few weeks ago.
Sorry for the delay. Im workin on more pics

This is actually a good picture showing the debris carried in by the fan. I have often found metal grit on the board at this point when opening suspect controllers. Teams are shocked by the amount of stuff once I open one of their controllers.

[Richard Wallace]

Very good photo. I have suspected this condition on many occasions; as Al says it is very likely the leading cause of Victor failure.

The moral of this story is: don't make chips and metal dust around the electrical system. You may think you've cleaned it up afterwards, but that's actually somewhere between extremely difficult and totally impossible.