Amps
[Top] [All Lists]

[Amps] Solid state amp advice needed

To: <amps@contesting.com>
Subject: [Amps] Solid state amp advice needed
From: rakefet@rakefet.com (Vic Rosenthal)
Date: Mon, 25 Nov 2002 07:52:14 -0800
Jeffrey Madore wrote:
> 
> I recently had problems working on a solid state final. The transistors
> idled fine but the minute I applied drive the collector current went way up
> and the output was low. I too, let the smoke out of some expensive
> transistors.
> 
> What I found, through careful inspection, was a bad chip capacitor in the
> collector circuit. This final was also subjected to a mismatched load which,
> I believe, caused the cap to fail.

In this case, I first swapped out the whole board for a spare (and lost its
transistors).  Then, I put new transistors in the old board with the same result
(it really sounds dumb, doesn't it!).  So I don't think it's something on the
board.  I looked very carefully at the combiner and splitter, but they are very
simple and I can't see anything there to fail.

My present thinking is the following:  this amp has had a lot of abuse, even
before the most recent failure (the one caused by transmitting into the wrong
filter).  When I went to adjust the bias after installing the two new
transistors, the bias pots were all set to the same point.  I found that the
three other boards had idle currents that ranged from 200-275 ma per board, and
the new one about 375 ma.  I did adjust them to all draw about 250 ma, but I
think that the new transistors probably had much higher gain than the old ones,
and perhaps this caused them to 'hog' most of the drive, which is what destroyed
them.

Does this make sense?

What I am planning to do now is to replace all 8 transistors, adjust the bias,
and then carefully apply drive a little at a time, watching the current to each
board and looking at the RF voltage on the output of each board with a scope. 
That should show up potential problems before they become actual ones.

73
Vic K2VCO

<Prev in Thread] Current Thread [Next in Thread>