1. Home
  2. Electronics Design
  3. Transistor torture...(or gruesome murder anyway)

Transistor torture...(or gruesome murder anyway)

FGA40N60DTU:

formatting link

Now this is a neat one. I had no idea there was enough tin plating on these things to make a perfect casting of the sil-pad surface.

Suffice it to say, the power dissipation was underestimated, or alternately: sil-pads stink. ;-)

Both of them cracked, and the right hand one was just about falling apart. Amazingly, it failed open circuit: in fact, for all the transistors I've killed, this is the Very First transistor I've actually blown open circuited!

formatting link

Gentle prying removed the case,

formatting link
exposing a fairly intact copak diode die (now fully shattered from seperation, but still shiny anyway), and absolutely no remains of what used to be an IGBT die.

Looks like the die overheated due to switching loss (this circuit was going way too slow for 80kHz!), so when it hit ~200C, it stopped turning off, causing dramatically increased switching loss and shoot through, then becoming an effective short circuit. The 1820uF of supply caps, plus a 20A circuit breaker, discharged through the two of them (they both look similar inside), making for what I'm sure was an exciting display inside the package.

Neat how the emitter bondwire seems to have turn into a fulgerite.

Tim

--
Deep Friar: a very philosophical monk.
Website: http://webpages.charter.net/dawill/tmoranwms
Reply to
Tim Williams
Loading thread data ...

Glue them back together and sell them on E-BAY for half price. Under not tested and as is.;-)

Reply to
Hammy

"Tim Wanker Williams"

** ??

** Utter nonsense.

..... Phil

Reply to
Phil Allison

Sil-pads are indeed garbage. They usually have thetas around twice datasheet specs. You pay a dollar for a premiun sil-pad, and it reduces the value of your mosfet by 10:1.

But Phyllis loves sil-pads, so you will have to endure her wrath.

You can usually squeeze them edgewise in a vise and shatter the plastic off.

ftp://jjlarkin.lmi.net/ExFets.jpg

Speaking of which, yesterday I blew up a Murata sip dc/dc converter, and hacked it open to see what's inside. It's potted with some semi-hard, sort of gritty epoxy, so by the time I x-actoed my way in I'd destroyed enough stuff to make analysis difficult. Anybody know of a way to dissolve epoxy? The substrate is ceramic, so it will survive.

Looks like a classic 2-transistor push-pull oscillator, tiny toroid, and a dual-diode half-bridge rectifier, but I can't make out the details. Runs around 60 KHz. These dc/dc bricks are handy gadgets.

Blowing up/dissecting stuff is fun.

John

Reply to
John Larkin

these

ely:

t.

sed

I use to disolve epoxy seals with methylene chloride.. nasty stuff. Worked in a fume hood and put everything in a sealed container so it wouldn't evaporate. Took a few days and the epoxy would soften up... then added a bit of heat and I could pop the seals open.

"Blowing up/dissecting stuff is fun."

A yup,

George H.

Reply to
George Herold

these

ely:

t.

sed

Warm it to ~200=BAF, and it gets soft. You can pick it away with a needle, heating as needed to keep it soft.

Works great.

James Arthur

Reply to
dagmargoodboat

You might try carburetor cleaner? In which case you could get a wide-mouth can with a basket to dunk parts that you can reseal for less fumes (with the parts dunked in it) at an auto parts store. California version is different than elsewhere, per one set of MSDS I looked at, but either seems worth a try.

I recalled it having the aforementioned methylene chloride, but that was missing from the MSDS's I happened to find for the current stuff - perhaps some brands still use it, but if something else will work, might be just as well. When I had all 4 carbs form a 140 corvair soaking, I had tinfoil over them and used a fume hood and still was not overly happy about it - nasty stuff.

One brand of commercial epoxy solvent ("Attack") seems to be mostly methylene chloride (based on some mentions, have not found an MSDS for it), so there's an indicator that that may be the stuff you want. In which case, all due caution, it is unpleasant stuff.

--
Cats, coffee, chocolate...vices to live by
Reply to
Ecnerwal

That's the hard way. Hot sulfuric acid is easier. Or, bake in a kiln and dust after cooling with a camel hair brush...

Reply to
whit3rd

You should have local suppliers who can remove the epoxy for a reasonable fee (like $50 or something). They use nasty stuff like hot spray nitric acid, so it's worth outsourcing. The key search word is "decapsulate".

Best regards, Spehro Pefhany

--
"it's the network..."                          "The Journey is the reward"
speff@interlog.com             Info for manufacturers: http://www.trexon.com
Embedded software/hardware/analog  Info for designers:  http://www.speff.com
Reply to
Spehro Pefhany

Is this a new Impulse LED design?

Reply to
Robert Baer

At Signetics, to get inside standard epoxy package ICs, we used IRFNA and a hot plate (don't ask temperature, do not remember); use DRY acetone as a wash to get the bubbling glop off. Milled into the package to near die first to decrease time; also the acid eventually dissolves the metals.

Reply to
Robert Baer

When I needed a die photo in a hurry (back when I ran a thick film hybrid line), I'd simply run the plastic package thru our firing furnace (~1100°C, IIRC). It'd emerge from the other end, looking intact, though now white in color. Turned out the plastic became ash, which was easily brushed away ;-) ...Jim Thompson

--
| James E.Thompson, CTO                            |    mens     |
| Analog Innovations, Inc.                         |     et      |
| Analog/Mixed-Signal ASIC's and Discrete Systems  |    manus    |
| Phoenix, Arizona  85048    Skype: Contacts Only  |             |
| Voice:(480)460-2350  Fax: Available upon request |  Brass Rat  |
| E-mail Icon at http://www.analog-innovations.com |    1962     |
             
      The only thing bipartisan in this country is hypocrisy
Reply to
Jim Thompson

When I was destroying 1206 resistors recently, with 1 ms 60-watt pulses, we looked at one with our IR viewer. It looked just like a pulsing LED, with a big blob of illumination between the end caps. I think that a properly designed thinfilm on the right substrate can make microsecond thermal flashes.

John

Reply to
John Larkin

I've used lacquer thinner to dissolve epoxy, but it works rather slowly, and I'm not sure about different formulas of epoxy. It does work slowly on the stuff from the handyman store.

Jim

Reply to
RST Engineering

.

Depends on your definition of hard, doesn't it? :-) Hot acid scares me.

OTOH, dipping a part in boiling water is pretty benign.

Epoxy gets crumbly (like a gum eraser) when so warmed. That, then picking away the crumbs with a dental pick had it clean in no time.

(This was for epoxy used to encapsulate / pot. Dunno if, fer instance, IC packages would respond as well.)

-- Cheers, James Arthur

Reply to
dagmargoodboat

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.