Electronics BUZ11 MOSFET help

I have never used a MOSFET in my life!
I want to build a UPS.
What will the Gate voltage be if the Drain-Source voltage is 12-14v?
If I understand the datasheet, it should be the same as the Drain-Source voltage, right?

Sputnik

 

In your switching application the gate voltage used will set the maximum current that can flow. With the low voltages you're using you can safely use high & low (supply & ground) to switch.

Intro to FETS/MOSFETS by Ray Marston

 

I have read somewhere that MOSFETs current capbility is limited by thermal characteristics or something like that, unlike normal transistors (or was I dreaming). Is this correct? If so, could I draw over 27A by the BUZ11 so long as they remain sufficiently cool?

Sputnik

 

You don't get thermal runaway with MOSFETs, but it doesn't mean you can over-rate them in safety. In an amplifier the wattage VI (heat) is the limiting factor, but in a switch it's the current I.

You can parallel them for more current or less heat (provided you give each a gate resistor and a small source resistor to help balance out the shares). Or there are plenty over 36A rating for switching applications.

 

This is how I am trying to use this MOSFET, but it still gets hot with just 1A current draw. I'm not sure what value resistor I'm supposed to be using. Is there anything I'm doing wrong?

Sputnik

 

Two problems,

1. The load is on the wrong side of the n-channel MOSFET for switching, the MOSFET will be dropping about 4V so producing a few watts waste heat.
2. The 2N3906 PNP transistor is upside-down so you won't get the fast turn-off it's there for.

Try 470R-1k for the resistor. It's not critical.

 

I wasn't quite sure how to draw the 2N3906, but it is correct in the circuit. I'll try swapping the MOSFET and load in the morning.

Here is the updated circuit diagram that I hope is correct and will work.
(nevermind)

Thanks,
Sputnik

 

Arrow is emitter, and arrow always points from positive to negative (unless the transistor is deliberately used reverse-biased).

 

I can't take the confusion anymore! I'm now using a BUZ11 MOSFET to make a PWM motor controller. My problem now is that I keep destroying them. They work for a while, but then it just conducts current straight through it. The resistance between the Drain-Source is something like 200ohms when off. I have now killed two MOSFETs today and I'm getting irritated. What the heck must the gate voltage be if I'm now using 24v from Drain to Source?

I can't stand MOSFETs! Is it 8v, perhaps?

Sputnik

 

Your circuit's vanished.

 

I hope this will work:


Should have actually drawn R2 after R1.

Sputnik

 

Two probable reasons why you're killing them. First is you're destroying them with back emf from the motor. You need a reverse protection diode (and capacitor to be sure). Second reason is MOSFETs are extremely sensitive to ESD. I've destroyed two in the last week just by picking them up after just taking off my wrist strap. You need to wear a strap all the time to be sure you don't destroy them.

The first section of your circuit doesn't look right either. Q1 and Q2 are the wrong way round really. NPN should be switching the low side, and in that case, you can get rid of the PNP and tie the gate to the motor supply with a 10k.

 

Agree on your first points, but the NPN/PNP layout is similar to the driver used by Oatley in a 10A/20A high-frequency PWM controller, though such a system should only be needed when fast switching is essential (HF PWM).

However,

• they're using a 12V supply - the BUZ11 isn't designated a 'logic-level' MOSFET so may not be full-on with a gate voltage a tad under 5V, depends on the current the motor demands. Datasheet suggests it's typically limited to under 8A but YMMV. Does it get hot when it fails?
  
  
• they also protect the MOSFET gate against spikes with a 16V zener.

 

I'd suggest using only a NPN to pull the gate low, and a pull up resistor to the motor supply ensuring the FET is fully on.

 

remembering the NPN will invert the signal; in the original a 'high' turns Q1 so Q3 on.

 

I can't tie the gate to the motor as the BUZ11 has a maximum gate threshold of 20v and the motor is being run off 24v. According to the datasheet, a voltage of 8v or 9v will be required for the BUZ11 to carry it's full load of 27A. During no-load on the motor, current draw is typically 2.3A. The diode and capacitor will be added. I'm still not sure of the switching side of things.

I'm still working on the PWM side of things. The PIC coding is coming along well.

Sputnik

 

Drop the voltage with a zener or just create a voltage divider.


edit: Your pictures on this server http://klepc.homeip.net/sputnik/... are causing a password prompt when you visit this page.

 

I have decided to revive this project after it lost steam a while back. I suppose I gave up after destroying 4 MOSFETS and numerous transistors with one exploding with a briliant shower of sparks.

Now my question is: Should I pull the gate high and drive it low with a transistor or pull the gate low and drive it high? I know earlier in this thread I was told to pull it high and that the NPN transistor should be on the low-side. But I'm not sure of the MOSFET being default on (pulled high). I'd prefer it to be default off, unless I merely invert the signal and use the pull-high-drive-low configuration.

I'm not very good at this,
Sputnik

 

Have you thought about driving it high AND low? or even using a MOSFET driver to make things easy?

 

Yes, I have thought of it and tried it. Only to have destroyed many transistors. I can't seem to get it to work. If I use a push-pull configuration, do I need a -ve supply? How does the PNP transistor turn on then if I don't use a -ve base current? Perhaps I should bury myself in some electronic books...

MOSFET drivers are not sold locally and I can't order over the internet.

Sputnik