Help with MOSFET

[oyektn]

Hey,

I recently got a ton of power MOSFETs out of an old UPS. I've been trying to build a circuit to take the audio output from my computer and buffer (is this the right word?) it so it can provide enough current to drive a speaker (once I have that working I also plan on making the circuit be able to amplify the voltage).

Anyway the problem is that the transistors I'm using (IRFZ46n https://www.electro-tech-online.com/custompdfs/2009/11/IRFZ46N.pdf) seem to have a zener diode built in across the drain and source. I measured the reverse voltage drop to be around 9V. Does this basically mean that I can't use the transistors to control anything above this? The datasheet lists the VDS breakdown as 55V, but it seems like this wouldn't really matter, as its never going to get above maybe 11.

So is there any way to get around this limitation? I guess I could put a couple of them in series but that seems wasteful and would make the circuit more complicated. Also their drains are all connected to shared heatsinks, so I assume I'd have to take them off those. Anyway, if anyone can help I'd appreciate it because I'd hate to not be able to use any of these.

 

[indulis]

The IRFZ64N is rated for use to 55V. The body diode is not a zener.

 

[audioguru]

The body diode has avalanche breakdown like a zener diode when the drain-source voltage is 55V or more.

 

[indulis]

The body diode has avalanche breakdown...

You mean like any P-N junction.

 

[audioguru]

Mosfet manufactures draw the body diode to look like a zener diode because it acts like one.

 

[mneary]

You mean like any P-N junction.

No.

Pause long enough to look at the data sheet. The junction is rated for avalanche currents of 28A, limited by the device dissipation specs. If it's not a zener, then it's better than one.

 

[indulis]

MOSFET body diodes are parasitic in nature… they are not put there on purpose, but are a byproduct of the MOSFET’s physical construction. Every diode, or P-N junction for that matter is a “zener”, just not a very good one. The doping levels are not carefully controlled for a reverse breakdown operating point as it is in a zener. Doping levels are completely different for diodes where the forward characteristics are of primary concern. Typically body diodes have the same current rating as the MOSFET. In the “power supply world” (DC-DC), the MOSFET body diode of the primary side switching element(s) is often used for reverse polarity protection and when used for synchronous rectification, depending on the drive circuitry, frequently a schottky diode is put in parallel to improve efficiency if there’s any chance it will conduct. Also, the body diode reverse recovery characteristics usually suck for hi frequency use. Besides IR, Onsemi also shows the body diode as a zener, where as Vishay, ST, Fairchild, Infineon and many other show it as a “regular” diode.

 

[audioguru]

I wonder what is the voltage range of a production run of one zener diode?
11V to 13V? 10V to 14V? 6V to 18V? 4.7V to 22V?

Can the manufacturer control the level of doping well enough to produce a zener diode close to one voltage or does a production run make a wide range of voltages that are tested, selected and marked?

 

[indulis]

I would guess they can control it fairly well give today’s manufacturing methods. One thing is for sure… whatever a particular batch of silicon is, they will all be quite close to one another. The consistency in devices beyond one P-N junction, like op-amps’s, is quite different. For some, active laser trimming is done when probing the die while at wafer level… the “bin 1” stuff goes to the government and “bin 32” devices head for Radio Shack.

 

[audioguru]

Are many zener diodes made on one wafer like ICs or are they made separately?

 

[indulis]

It's not practical to make them one-at-a-time.