QDX went "bang", literally, Q10 blown


Tom C
 

My QDX had been going hard for a couple hours, when it made a frightening bang while transmitting.

After examination, I saw Q10 had a crack in it.  I imagine this was from excessive heat, plus the bolt being tightened down too much, putting excess physical stress on the plastic case of the transistor. 

How much do transistors expand when heated?  Perhaps what seemed like moderate pressure when I was tightening the bolt became extreme pressure when the transistor case expanded?



I'll be replacing all of the BS170s to be safe.  

I've been reading comments about thermal management using heat sinks, thermal tape, etc.  What about changing the shape of the transistor?  My thought is to sand down the rounded side of the transistor to make a flat surface that will be easier to connect to a heat sink, and will less susceptible to odd stresses from a compression force being applied to the rounded surface.

So far, I have sanded down over 1mm from that cracked BS170, with no sign of reaching any internal components.  It is now a nice rectangular shape.  If I can sand all 4 new transistors to the same thickness, adding a heat sink over the top of them will be easy.  Perhaps some soft copper tape as well, which will conform to the shapes and fill in any inconsistencies in my sanding.

Would this be worth experimenting with, or am I being silly?


Al Holt
 

Tom, 
Could you describe what "going hard for a couple of hours" entailed? What voltage do you operate the QDX and what's the typical current draw during transmit?

Thanks!

--Al
WD4AH


Tom G
 

"Going hard" was FT8, and the last 10 minutes I was doing CQ instead of responding to others.  Using a 13.8V supply, and it was putting out 5W.  I didn't have a current meter attached.  I'll do that once I get the QDX working again.


Adam
 

Tom G <tomscraiglistemail@...> wrote:

Using a 13.8V supply, and it was putting out 5W. 
I killed my finals with 13.8 V. I switched to 12.0 V and didn't have any
more problems (at least not with the finals).


Adam
 

Tom C <mel0torme@...> wrote:

How much do transistors expand when heated? Perhaps what seemed like
moderate pressure when I was tightening the bolt became extreme pressure
when the transistor case expanded?
Maybe a good idea would be adding a spring washer? If not overtightened it
should put a reasonable pressure regardless of the expansion...

So far, I have sanded down over 1mm from that cracked BS170, with no sign
of reaching any internal components. It is now a nice rectangular shape.
If I can sand all 4 new transistors to the same thickness, adding a heat
sink over the top of them will be easy. Perhaps some soft copper tape as
well, which will conform to the shapes and fill in any inconsistencies in
my sanding.

Would this be worth experimenting with, or am I being silly?
Try it -- if it works, it works :)


ve3ega
 

Don't overdo the sanding - just a slight amount to flatten the concave surface to adhere to a heatsink with some paste!


Lee
 

There is nothing like hearing a loud bang and having the remains of a component whiz by your ear as it attempts to reach orbital velocity.......................

Lee KX4TT


Brad Morris
 

I’m wondering about a copper crush washer under the flat washer provided? That would conform to the BS170 and provide better thermal transfer. 


Brad