Chief Delphi

Chief Delphi (http://www.chiefdelphi.com/forums/index.php)
-   Extra Discussion (http://www.chiefdelphi.com/forums/forumdisplay.php?f=68)
-   -   pic: Well there's your problem... (http://www.chiefdelphi.com/forums/showthread.php?t=93765)

J.Warsoff 19-03-2011 13:16

pic: Well there's your problem...
 

Alpha Beta 19-03-2011 13:17

Re: pic: Well there's your problem...
 
That looks familiar. I'm guessing that was proceeded by a flash of bluish white light and a little smoke. We saw that in a brown jag the weekend before ship.

J.Warsoff 19-03-2011 15:22

Re: pic: Well there's your problem...
 
Quote:

Originally Posted by Alpha Beta (Post 1042025)
That looks familiar. I'm guessing that was proceeded by a flash of bluish white light and a little smoke. We saw that in a brown jag the weekend before ship.

We actually have never had a problem with our tan/brownish jags. We lost 4 black ones already this year and I think we lost a few last year as well

Mike Betts 19-03-2011 16:57

Re: pic: Well there's your problem...
 
Jared,

Actually, what you see could very well be the result of the failure and not the root cause...

When a power semiconductor fails, it almost always causes a resultant short circuit. This has to do with the inner structure of the device. The actual cause of the failure could be a voltage spike, over-current, loss of cooling (over-temperature), et cetera.

As an example: An over-current condition can be the result of a logic failure upstream of the device causing a "shoot-through" condition (the upper and lower halves of a power pole turn on simultaneously shorting the 12V to ground). This logic failure could be caused by a piece of aluminum dust shorting out two pins on an IC chip elsewhere on the board.

So when you see power devices failed like this, think of it as an indicator that something else failed...

Of course, the failure could indicate a manufacturing defect or a design deficiency as well...

I hope this makes things a little more cloudy for you... Welcome my world...

Regards,

Mike

J.Warsoff 19-03-2011 17:34

Re: pic: Well there's your problem...
 
Quote:

Originally Posted by Mike Betts (Post 1042075)
Jared,

Actually, what you see could very well be the result of the failure and not the root cause...

When a power semiconductor fails, it almost always causes a resultant short circuit. This has to do with the inner structure of the device. The actual cause of the failure could be a voltage spike, over-current, loss of cooling (over-temperature), et cetera.

As an example: An over-current condition can be the result of a logic failure upstream of the device causing a "shoot-through" condition (the upper and lower halves of a power pole turn on simultaneously shorting the 12V to ground). This logic failure could be caused by a piece of aluminum dust shorting out two pins on an IC chip elsewhere on the board.

So when you see power devices failed like this, think of it as an indicator that something else failed...

Of course, the failure could indicate a manufacturing defect or a design deficiency as well...

I hope this makes things a little more cloudy for you... Welcome my world...

Regards,

Mike

That helps clear things up a little Mike. Thanks

One of the mentors on my team was the one teaching me about how jags worked, and I came up with the possibility of maybe the gate, source, or drain inside the MOSFETs coming out of alignment, which would mess up the circuit and cause an overheat.

And I am still looking at the jags and trying to figure out any other failures that caused it. Either way, my team will be sticking with the tan jags, since we've never had problems with those.

Tristan Lall 19-03-2011 19:15

Re: pic: Well there's your problem...
 
Quote:

Originally Posted by Aerosound (Post 1042085)
Either way, my team will be sticking with the tan jags, since we've never had problems with those.

I think you're the first person to express the opinion that the grey Jaguars (that's what they call them, though I agree they're not) are better than the black Jaguars.

Wasn't part of the reason for the redesign to improve fault tolerance? (With the other objectives being to use more TI components, and to add the CAN bridge.)

DonRotolo 19-03-2011 20:08

Re: pic: Well there's your problem...
 
Quote:

Originally Posted by Aerosound (Post 1042085)
and I came up with the possibility of maybe the gate, source, or drain inside the MOSFETs coming out of alignment, which would mess up the circuit and cause an overheat.

Considering that these sections of the MOSFET are one solid piece, that's impossible.

To make the Gate, Source and Drain of a FET, one takes a solid chunk of Silicon, and adds small amounts of special impurities by painting it on to the outside, then baking in an oven to get the 'paint' to soak into the silicon.

While the actual process is quite a bit more complex, this explanation will give you the right idea as to how it is done.

J.Warsoff 21-03-2011 08:01

Re: pic: Well there's your problem...
 
Quote:

Originally Posted by DonRotolo (Post 1042117)
Considering that these sections of the MOSFET are one solid piece, that's impossible.

To make the Gate, Source and Drain of a FET, one takes a solid chunk of Silicon, and adds small amounts of special impurities by painting it on to the outside, then baking in an oven to get the 'paint' to soak into the silicon.

While the actual process is quite a bit more complex, this explanation will give you the right idea as to how it is done.

Oh well thanks Don. Looks like I misunderstood you once again when you explained it to me. woops!


And Tristen, I haven't been able to compare the two different models or figure out the major difference, so that's a question I can't honestly give a good answer to. But I do know that we are switching to tan, since we have never had a problem with them.

Bandgeek80001 21-03-2011 12:15

Re: pic: Well there's your problem...
 
We had one of ours do that, too... We chalked it up to metal shards from some last-minute hole-drilling...

JByrd 22-03-2011 18:34

Re: pic: Well there's your problem...
 
You know we had that exact same problem with a victor except since a victor is open at that point it broke off and launched halfway across the room


All times are GMT -5. The time now is 10:46.

Powered by vBulletin® Version 3.6.4
Copyright ©2000 - 2017, Jelsoft Enterprises Ltd.
Copyright © Chief Delphi