MOSFET drain-source voltage drop

I am using a MOSFET driver in a PWM circuit. I am switching 12v (battery) with a Schmitt trigger oscillator on the gate. It's running about 7kHz. The MOSFET is a IRF1404

formatting link

The oscillator works nicely and puts 12v on the MOSFET gate. But the MOSFET drops 2v and I don't get full speed from the motor that I'm driving. The MOSFET was salvaged from the original motor controller which did run the motor full speed (I think).

Am I missing something?

Thanks, Bob

Reply to
Bob Engelhardt
Loading thread data ...

Look at the gate risetime under load; there's a lot of gate capacitance and some Miller effect. DC motors' commutators are a source of much noise and consternation. Could switching transients be depressing your gate voltage at switch edges?

Reply to
whit3rd

What chip for the oscillator, and what's it's supply? That looks like an older chip, and it may not turn on all the way with just 5 volts.

--
www.wescottdesign.com
Reply to
Tim Wescott

Like whit3rd said, I'd like to see a 'scope shot of the gate voltage. How much current is it drawing?

George H.

Reply to
George Herold

He said he's putting 12V on the gate.. so that should be OK. (Assuming it is getting to 12V.)

George H.

Reply to
George Herold

The gate rises to 3 or 4v in about 1us then to 12v too fast for me to see (7kHz is 140us period).

I've tried it unloaded and it still only outputs 10v. (This is a brushless 2-wire motor - the electronic commutation is inside the motor.)

Thanks, Bob

Reply to
Bob Engelhardt

Probably gate capacitance. At high PWM rates, the gate capacitance is significant and slows the rise time until the device is working in the linear region.

Your options are to drive the gate with enough current (a low impendence driver amplifier) to switch it quickly, or lower the frequency you are driving it with.

Here's an explanation for some ready-made chips designed specifically to drive power mosfets and IGBTs:

formatting link
A gate driver is a power amplifier that accepts a low-power input from a controller IC and produces a high-current drive input for the gate of a high-power transistor such as an IGBT or power MOSFET. Gate drivers can be provided either on-chip or as a discrete module. In essence, a gate driver consists of a level shifter in combination with an amplifier.

Reply to
default

The oscillator is a CD40106

formatting link

It is using a 12v battery as its supply.

Thanks, Bob

Reply to
Bob Engelhardt

I tried a couple of things:

- I was way off on the gate rise time; here's a picture:

formatting link
sweep is 0.1us/ & signal is 5v/

- I paralleled 3 of Schmitt triggers for extra drive & that had no effect (driven by oscillator & driving gate)

- I swapped the MOSFET with another that I had also taken from the original controller. No change.

- I put a 10R on the MOSFET & the voltage dropped to 7-1/2.

Bob

Reply to
Bob Engelhardt

You could always use a Fet Driver

Or build one from discretes. As below.

Mikek

Reply to
amdx

Are you measuring the Vgs at the mosfet, or assuming that the source is the same as some other ground in your circuit?

Are you measuring the Vds drop at the mosfet, or calculating it based on some other measurement?

IMHO you should connect the scope's ground to the mosfet's source pin, and two input channels to gate and drain, and see what's actually happening at the mosfet. You might find, for example, that you have a dead battery ;-)

Reply to
DJ Delorie

This is altogether weird. I was going to ask how much current you were flowing, but if you're using the MOSFET you say you are, if the Vds is 2V with full gate drive because of high current the FET would just be a loud noise and a bad smell.

Does the source really go to ground? Is it really the FET you think it is?

How much current is the motor pulling?

--
Tim Wescott 
Wescott Design Services 
 Click to see the full signature
Reply to
Tim Wescott

Hmmm OK maybe they are bad FET's... or you have source and drain reversed? (I hooked up a IRF520 to a power supply and measured a D-S voltage of

25mV with 200 mA flowing. An easy check of the FET.

George H.

Reply to
George Herold

What's the load current?

Post the circuit!

--
John Larkin         Highland Technology, Inc 
picosecond timing   precision measurement  
 Click to see the full signature
Reply to
John Larkin

please take measurements between Source and Gate of the FET, close to the body.

Jamie

Reply to
M Philbrook

You're trying to PWM an electronically commutated motor. I'm not so sure that will work without problems. What's the circuit? What's the make and model of the motor?

Jeroen Belleman

Reply to
Jeroen Belleman

The motor is that of a cordless jig saw. Its control board died and this project was to make a replacement. The motor only has 2 wires, so I _assumed_ that it was PWM control.

At this point I'm just going to abandon the variable speed and use on-off control.

Bob

Reply to
Bob Engelhardt

More:

To totally remove effects of motor & oscillator, I connected the MOSFET alone: 12v to D & G, S to gnd thru 1K:

V(DS) = 2-4v pp, about 3MHz sine (kinda) (S to gnd "output" 8-10v pp)

The MOSFET is clearly marked IRF1404Z

Double (triple, ...) checked pin order (G-D-S)

battery: 12v drops to 11v with 10R load

At this point I am abandoning the idea of making a PWM controller, but the MOSFET REALLY bugs me.

Thanks to all, Bob

Reply to
Bob Engelhardt

Oh ... the last thought is that the "problem" isn't with the MOSFET, but is something really stupid that I did. A "duh" thing.

Bob

Reply to
Bob Engelhardt

That's what I did, but make the R smaller (50 or 100 ohms) for more current. Then measure Vds at DC.

George H.

Reply to
George Herold

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.