In rec.radio.amateur.homebrew Ken Finney wrote:


Silver cased wet slug tantalums DO explode, most contracts that
allow the use of wet slugs require the use of tantalum cased parts.

Some years ago I wandered into one of the design labs to ask some questions
about a new thingy we were doing.

On a bench was a metal trash can with wires leading to a power supply,
thermometer, and a chart recorder.

"What's all this", I ask.

"Getting some real data on stressing tantalums. The trash can is a blast
shield just in case" was the answer.

Just after the guy running the test uttered the words "looks like nothing
bad is going to happen" the cap exploded with the trash can acting as
a megaphone for the bang and director for the shrapnel; everyone around
hit the deck.

We got off the floor and looked up to the acoustical ceiling where there
was capacitor pieces and an alligator clip embedded therein.

When the guy running the test said "Maybe we had better rethink this
design", I decided to go back to my lab and come back on a better day.

We continued to use tantalums (aerospace), but there were a lot more
explosions in the lab to make sure they didn't happen on the shipped
product.

--
Jim Pennino

Remove -spam-sux to reply.

wrote:

In rec.radio.amateur.homebrew Ken Finney wrote:




Silver cased wet slug tantalums DO explode, most contracts that
allow the use of wet slugs require the use of tantalum cased parts.



Some years ago I wandered into one of the design labs to ask some questions
about a new thingy we were doing.

On a bench was a metal trash can with wires leading to a power supply,
thermometer, and a chart recorder.

"What's all this", I ask.

"Getting some real data on stressing tantalums. The trash can is a blast
shield just in case" was the answer.

Just after the guy running the test uttered the words "looks like nothing
bad is going to happen" the cap exploded with the trash can acting as
a megaphone for the bang and director for the shrapnel; everyone around
hit the deck.

We got off the floor and looked up to the acoustical ceiling where there
was capacitor pieces and an alligator clip embedded therein.

When the guy running the test said "Maybe we had better rethink this
design", I decided to go back to my lab and come back on a better day.

We continued to use tantalums (aerospace), but there were a lot more
explosions in the lab to make sure they didn't happen on the shipped
product.


Few years ago, my boss came in, laid a printed schematic on my desk, and
said "Put this in PSpice, so we can demo it to a customer. I said OK,
and started entering the design. He comes in a couple of hours later,
and wants to know if it is finished! I go, SaWha? He now tells me that
he is getting on a plane in a few hours, and has a meeting with the
customer in the morning! So I get to it, end up emailing it to him the
next morning, as there were some digital parts not in the standard
libraries, had several components that I had to create, etc. There are
still major bugs in the simulation, etc. That afternoon, he calls, and
says that they were mad because they didn't see what they were looking
for in the simulation. I ask, What were they looking for? And he tells
me, this cap in the power supply keeps exploding! I then realize he is
talking about the tantalum! Since I had just seen a thread like this
one, I knew about tantalums, so I told him that there was no way we were
going to create a special model for PSpice for tantalum caps! Most
designers just knew not to use them as power supply filters!

Got a bad review from him that year. I now know that, when you get an
assignment, ask LOTS of questions, to the actual customer whenever
possible! 8-)

Charlie
Edmondson Engineering
Unique Solutions to Unusual Problems