What can cause MOSFET explode? / General Science and Electronics / Forums

Forums

4hv.org :: Forums :: General Science and Electronics

« Previous topic | Next topic »

What can cause MOSFET explode?

Move Thread

LAN_403

hsieh

Fri May 27 2011, 04:42PM

Registered Member #1412 Joined: Thu Mar 27 2008, 04:07PM
Location: Taipei Taiwan
Posts: 278

I see my classmates blow up MOSFET very often.Sometimes the fragments can fly very far.Although my circuits never blow up,I am always worrying that my MOSFET might explode if something goes wrong while testing my SSTC.

Of course,I know that many factor can kill MOSFETs.But sometimes dead MOSFET just short-circuit between drain and source and didn't explode,Sometimes it explode and even produce fragments.

Which is more likely to explode a MOSFET?overheat or over current?I have ever seen MOSFETs explode after a short run while my classmates testing a big SMPS,What might be the cause of the problem?

Thanks

radiotech

Fri May 27 2011, 05:28PM

Registered Member #2463 Joined: Wed Nov 11 2009, 03:49AM
Location:
Posts: 1546

When the MOSFET explodes, the heat rise causes parts to
enter vapor state and expands against the case wall. The optimum
electrical case is when the impedance of the voltage source
equals the conjugate of the impedance of the current path inside
the device.

If someone could arrange high speed X-Ray photography,
that would be an interesting class project.

Forty

Fri May 27 2011, 05:49PM

Registered Member #3888 Joined: Sun May 15 2011, 09:50PM
Location: Erie, PA
Posts: 649

I'd say over current or improper connection would cause more of an explosion. whenever i have one overheat it just tends to sizzle and emit magic smoke. A big current surge is more likely to cause the internal vaporization that leads to the bang. Do not stare directly into the exploding mosfets.

IamSmooth

Fri May 27 2011, 06:11PM

Registered Member #190 Joined: Fri Feb 17 2006, 12:00AM
Location:
Posts: 1567

Overvoltage will cause the case to explode. Overcurrent will blow a hole out the side. This is what I have seen. As the mosfet heats, it takes less current to cause it to fail as a short.

Sulaiman

Fri May 27 2011, 07:20PM

Registered Member #162 Joined: Mon Feb 13 2006, 10:25AM
Location: United Kingdom
Posts: 3140

generally speaking, some thing causes a semiconductor to go low impedance
(over-voltage, overheating, S.O.A. stuff etc.)
what happens next depends mainly upon the power/energy that is available to the now low-resistance semiconductor.
e.g. 'explosive' energy sources include capacitors, batteries and mains.
e.g. resistance in series will limit the current to a level where something other than the failed semiconductor will fail, leaving the semiconductor internally short but externally not damaged.
Sometimes the bonding wires between the semiconductor and lead-out connections blows open with no external signs.
So semiconductors can die by going short-circuit, then external circuitry decides whether you find an intact but short- or open- circuit, package, or holes, or shrapnel.
If your dc bus is good (low impedance) then the expected result for transistors in an sstc bridge is not quiet !

Design Tip .
If the transistors in your SSTC bridge fail quietly then your DC bus needs improvement.

hsieh

Sat May 28 2011, 12:22AM

Registered Member #1412 Joined: Thu Mar 27 2008, 04:07PM
Location: Taipei Taiwan
Posts: 278

Is there any possibility that exploded MOSFET can injure human body?

One of my classmates lied to me that his arm was injured by the fragment of a exploded electronic component.But later I know his arm was in fact injured in a car accident.

Weston

Sat May 28 2011, 03:54AM

Registered Member #1316 Joined: Thu Feb 14 2008, 03:35AM
Location: Cambridge, MA
Posts: 365

Not all drastic MOSFEt/IGBT deaths are violent. I had a series stack of 6 IGBT's all fail short circuit. The foil on the PCB was completely vaporized in many places, but there was no actual physical damage on the case of any of the transistors.

Dr. ISOTOP

Sat May 28 2011, 07:13AM

Registered Member #2919 Joined: Fri Jun 11 2010, 06:30PM
Location: Cambridge, MA
Posts: 652

I love the warning in EVR's manuals:
"This is a solid state power device. Components may fail explosively at any time and eject high velocity projectiles.
EYE PROTECTION IS REQUIRED AT ALL TIMES"
If the piece of shrapnel wanders in your eye, your eye will probably suffer drastic performance hits.
If you are afraid of exploding FETs (and you should be) shield you bridge.
As Sulaiman said, large bus caps are a sure way to have spectacular semiconductor failures.
Of course, this doesn't mean you should remove your bus caps, as they are important to the function of the bridge.

Steve Conner

Sat May 28 2011, 01:23PM

Registered Member #30 Joined: Fri Feb 03 2006, 10:52AM
Location: Glasgow, Scotland
Posts: 6706

As Sulaiman says it is the energy available from the power source that determines whether you get an explosion. Wear safety goggles when working on high energy circuits. When testing really big stuff like locomotive traction inverters, it is standard practice to hide behind a Lexan blast shield. On a more modest scale, I have been hit in the forehead by TO-220 transistor shrapnel, although not hard enough to hurt.

A really good power electronic design will include fast blow, DC rated, high breaking capacity fuses in the DC bus. They may not save the devices from burning out, but they will stop further damage like explosions and vaporised boards. It's not always done because the fuses add stray inductance and the good semiconductor rated ones cost several dollars each.

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob