Hello, I bought these MOSFETS

http://www.fairchildsemi.com/ds/FQ/FQP13N06L.pdf

I have a question regarding why the heatsink has 5V when I apply 5V to the drain?

I bought a lot of these and a lot of them are faulty, like some will have 2-5V @ source when 5V is at drain (gate is grounded/ high impedance)

I'm thinking that the heatsink is whats causing the short across the terminals of the mosfet. Does this happen often?

Any clarification will be greatly appreciated

 

There is nothing wrong with them. Look at the data sheet.

The tab on these cases is also center pin, they are electrically the same. This is durn near universal for a 220 style case.

To turn a MOSFET on fully you need at least 10 volts between Gate and Source. There are types of MOSFETs called logic MOSFETs that can use 5V, but the norm is 10, less than that and you get what you're getting.

MOSFETs

 

The datasheet shows that the Vgs threshold voltage for 0.25mA conduction max between drain and source is 1.0V to 2.5V so of course it conducts a little if its gate is 0V and its source is a high impedance to ground. Usually the source is grounded and its gate voltage controls its conduction.

 

These do appear to be logic level MOSFETs. From the data sheet...

FQP13N06L​

60V LOGIC N-Channel MOSFET

I am not sure I understand how you are connecting them for your test. Can you provide a picture?

 

There is nothing wrong with them. Look at the data sheet.

The tab on these cases is also center pin, they are electrically the same. This is durn near universal for a 220 style case.

cool i never knew that, thanks.

The datasheet shows that the Vgs threshold voltage for 0.25mA conduction max between drain and source is 1.0V to 2.5V so of course it conducts a little if its gate is 0V and its source is a high impedance to ground. Usually the source is grounded and its gate voltage controls its conduction.

I'm using these to control stepper motors, I didn't want to pass current through the motors if there was no signal at the gate.

 

Motors are supposed to be connected to the drains of Mosfets, not the sources. The sources are grounded for N-channel mosfets and are at the positive supply voltage for P-channel Mosfets. When the gates are at 0V for N-channel and at the positive supply voltage for P-channel then the Mosfets do not conduct. But then the Mosfets are inverters.

 

I spent a while trying to figure out how to make a H-bridge with the source's grounded. I can't seem to find a solution with just these NMOS. How do you suggest i make an h-bridge if all the sources must be grounded?

I couldn't find any power mosfets with the source & body separated. If i only knew what they were called...

Thanks

 

www.jameco.com/Jameco/catalogs/c253/P56.pdf - see the part at the top right, #259160CJ for an insulator. The small flanged washer insulates the 4-40 mounting screw.

 

I bought a bunch of this kind of shoulder washers from my local source. I use kapton tape for my insulation between the tab and metal, which usually works. You can find this tape from a variety of sources, but it tends to be a bit expensive (per roll).

The old standard was mica, but it is a bit rarer nowdays. Kaptons main use is to cover parts of a PCB being wave soldered, it is a handy high temperature tape that is mechanically very strong (at least to crushing forces). I've been known to use kapton to make the shoulder washer.

Use thermal compound, similar to what modern CPUs use.

For what it's worth, I like the MOSFETs, its something I would keep in my POP (pile of parts).

 

BTW, I've been kinda working on understanding how you can drive with the same MOSFETs on both sides of the H Bridge.

High Side MOSFET Drivers

 

As was mentioned the drain to tab connection is much more common.
International Rectifier made an ISO FET TO-220 with an isolated
tab. Haven't looked for that in a while. IIRC the thermal resistance was
high and it was more expensive than the common non-isolated package.

A common way to permanently mount TO-220 FETs and rectifiers to
a heatsink is to use a ceramic insulator and epoxy.

(* jcl *)

 

Thanks for the responses!

But... ermm maybe my question wasn't clear enough.

I was asking if anyone knew what the 4-pin mosfets are called (where the source and body is not connected); So I can treat it like a pass-transistor without worrying about the body effect affecting the Vth parameter. (which is the reason why I am assuming people suggest the source to be grounded?)

I remember coming across this in my studies, but never been able to find this component.

Thanks for the heat sinks, it seems much better than my home-made solution .

 

The original post referred to the package body having a voltage of +5V.

Is your question about the package body or the body region of the semiconductor?
I have never seen a FET where the body region of the die is not connected
to the source. If these aren't connected you get a parasitic bipolar transistor.
This connection is what gives you the intrinsic body diode.

ISO-FETs isolate the drain from the package body (see my original answer).

The only other FETs I have seen are SenseFETs. These FETs have four leads --
gate, drain, source and sense. When the source and sense leads are kept
at the same voltage the sense lead current will mirror the source current
at a specified ratio. This enables the current to be measured without a current
sensing element.

The source does not have to be grounded to turn the FET on. The voltage between
the gate and the source just has to be above the threshold voltage for conduction
to begin. It is usually easier to do when the source (for an N channel device)
is grounded.

A battery will work. An isolated DC-DC converter works. IR makes some photo-voltaic
gate drive circuits (very slow turn-on).

(* jcl *)

 

Wikipedia mentions them, but I've never seen them either.

http://en.wikipedia.org/wiki/MOSFET