Elecraft › [K2]

Q7 failure in K2

Classic

List

Threaded

6 messages Options

N9AUP

Q7 failure in K2

After 2 weeks or so of happy operating of my newly assembled K2, Q7 exploded with a loud snap. From what I read here, I'm going to have to replace the whole transmit transistor kit. I did see that this failure has happened to others, so I'd really like to hear if those who had the experience were able to identify the cause of the problem. Replacing the TX chain will be bad enough--I don't want to have to do it a second time.

What's crazy here is that I wasn't in transmit mode when it happened. Actually, I had gone over to change taps on my ladder line tuner to change bands. There was a considerable static build-up on the antenna and I got a small spark when I went to change a tap...and the Q7 let go at the same time. That could have been a complete coincidence. . .but it still seems strange. Any thoughts or experience regarding possible causes?

thanks,

George
N9AUP

Don Wilhelm-4

Re: Q7 failure in K2

George,

First, you do not have to replace the entire RF transmit chain, only the
two PA output transistors (Q7 and Q8) and the 2 2N2222As that control
the PA bias (Q11 and Q13). Replace Q11 and Q13 first, and then measure
the voltage on the PA transistor base solder pads during a TUNE - it
should be between 0.61 and 0.64 volts. If it is not in that range, do
not install the PA transistors until that condition has been corrected
or you will end up with a second set of failed PA transistors.

Yes, that static charge could have been a factor. The real failure
mechanism was either leakage in the Q7 PA transistor from the collector
to the base which will drive a lot of current into the base and saturate
the PA transistor. It is not designed for that kind of switching
service, so it will quickly explode with the high current and subsequent
heat inside the transistor housing.
The other possibility is that Q11 or Q13 failed and increased the bias
on Q7 and Q8 causing a swift increase in current through the PA
transistors - with similar results.
OK, so that is how it happened, one can only speculate on the real root
cause, but the static charge could have been a contributing factor.

Order the K2PAKIT from Elecraft - that kit has matched PA transistors.
Is this a K2/10 or is the KPA100 present? If the KPA100 is present, I
would suggest ordering 2 (or even 4) 1N5711 diodes (KPA100 D16 and
D17). Often with the KPA100 installed, static will destroy KPA100 D16
and D17, and then the power tries to go to maximum because the K2 MCU
thinks there is no power output - that attempt to drive at maximum power
often damages one or both of the base K2 PA transistors.
Bottom line - if the KPA100 is installed, make certain the KPA100
wattmeter diodes are good in addition to replacing the base K2 PA
transistors.

73,
Don W3FPR

On 2/18/2012 3:17 PM, N9AUP wrote:

> After 2 weeks or so of happy operating of my newly assembled K2, Q7 exploded
> with a loud snap. From what I read here, I'm going to have to replace the
> whole transmit transistor kit. I did see that this failure has happened to
> others, so I'd really like to hear if those who had the experience were able
> to identify the cause of the problem. Replacing the TX chain will be bad
> enough--I don't want to have to do it a second time.
>
> What's crazy here is that I wasn't in transmit mode when it happened.
> Actually, I had gone over to change taps on my ladder line tuner to change
> bands. There was a considerable static build-up on the antenna and I got a
> small spark when I went to change a tap...and the Q7 let go at the same
> time. That could have been a complete coincidence. . .but it still seems
> strange. Any thoughts or experience regarding possible causes?
>
> thanks,
>
> George
> N9AUP
>
> --
> View this message in context: http://elecraft.365791.n2.nabble.com/Q7-failure-in-K2-tp7297576p7297576.html
> Sent from the Elecraft mailing list archive at Nabble.com.
> ______________________________________________________________
> Elecraft mailing list
> Home: http://mailman.qth.net/mailman/listinfo/elecraft
> Help: http://mailman.qth.net/mmfaq.htm
> Post: mailto:[hidden email]
>
> This list hosted by: http://www.qsl.net
> Please help support this email list: http://www.qsl.net/donate.html
>

______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html

N9AUP

Re: Q7 failure in K2

Thanks Don,

I've now pulled the remains of Q7 and also removed Q8. I checked the bias and it is at .64 volts and present at both base pads. This is a K2/10, so does that rule out the static antenna charge theory? I have a sort of crude transistor check which consists of connecting ohm meter - to emitter and + to collector (NPN) then touching my finger across base and collector to see if resistance decreases. If that means anything in this case, then Q8 wasn't totally destroyed. Of course, I'll buy the K2PAKIT and install the matched pair.

I note that the receiver mode seems to be working just fine. I pulled the controller board and turned power on and off to do a relay reset per the manual. 40 through 10 meters are open at the moment, the relays are clicking and I have reception across the board so the relays/filters must be ok.

Can you suggest anything else to check?

73,
George

Don Wilhelm-4

Re: Q7 failure in K2

George,

OK, do check D9 for possible damage, or if you have the KAT2 installed,
there are wattmeter dides in that option that may have failed and could
produce the same thing as damaged diodes in the KPA100.

I am not saying the above is your problem, but I am saying, "do check" -
if you have a damaged power measurement component (RF Board D9 or KAT2
wattmeter diodes), you could end up wiith a repeat failure (and that
would not be nice).

Put the new 2N2222s in anyway and then check the base solder pads of Q7
and Q8 before installing Q7 and Q8.

If you do not have good de-soldering equipment - waste the component(s)
by clipping off the leads and remove the leads one at a time. Clean up
the solder pads with solder wick and if the holes are not open, use a
wooden toothpick or a stainless steel needle to open the holes if required.

73,
Don W3FPR

On 2/18/2012 5:39 PM, N9AUP wrote:

> Thanks Don,
>
> I've now pulled the remains of Q7 and also removed Q8. I checked the bias
> and it is at .64 volts and present at both base pads. This is a K2/10, so
> does that rule out the static antenna charge theory? I have a sort of crude
> transistor check which consists of connecting ohm meter - to emitter and +
> to collector (NPN) then touching my finger across base and collector to see
> if resistance decreases. If that means anything in this case, then Q8
> wasn't totally destroyed. Of course, I'll buy the K2PAKIT and install the
> matched pair.
>
> I note that the receiver mode seems to be working just fine. I pulled the
> controller board and turned power on and off to do a relay reset per the
> manual. 40 through 10 meters are open at the moment, the relays are
> clicking and I have reception across the board so the relays/filters must be
> ok.
>
> Can you suggest anything else to check?
>
> 73,
> George
>

N9AUP

Re: Q7 failure in K2

OK Don,

I pulled one leg of D9 and saw 54K forward and infinite reverse on the DVM (no KAT2) so I'm guessing that's ok for a Schottky diode. I just now checked the tuner taps and the antenna was charged up pretty good. . .that's amazing since there's hardly a breeze and it isn't below freezing out. I suppose my 35 foot linear loaded vertical dipole (with 6 foot capacitance hats on on the far ends) is a pretty good static picker-upper. Do you think I could get away with a choke to ground, say 50 MH on each leg of the ladder line on the antenna side of the tuner? I guess I better do the math to figure out what the impedance would be on each of the bands 40 thru 10 --although heaven only knows what the antenna impedance would be on each band. I guess I could always put the bleeder on the nominal 50 ohm side of the tuner.

73,

George

Don Wilhelm-4

Re: Q7 failure in K2

George,

I would advocate 100 uHy rather than 50 mHy - but for 40 thru 10 meters,
50 uHy is OK.

If you really meant 50 millihenries, then yes that would be quite fine
for providing a DC path for the lower bands, but there may be strange
effects on the higher bands due to stray capacity.

73,
Don W3FPR

On 2/18/2012 7:02 PM, N9AUP wrote:

> OK Don,
>
> I pulled one leg of D9 and saw 54K forward and infinite reverse on the DVM
> (no KAT2) so I'm guessing that's ok for a Schottky diode. I just now
> checked the tuner taps and the antenna was charged up pretty good. . .that's
> amazing since there's hardly a breeze and it isn't below freezing out. I
> suppose my 35 foot linear loaded vertical dipole (with 6 foot capacitance
> hats on on the far ends) is a pretty good static picker-upper. Do you think
> I could get away with a choke to ground, say 50 MH on each leg of the ladder
> line on the antenna side of the tuner? I guess I better do the math to
> figure out what the impedance would be on each of the bands 40 thru 10
> --although heaven only knows what the antenna impedance would be on each
> band. I guess I could always put the bleeder on the nominal 50 ohm side of
> the tuner.
>
> 73,
>
> George
>
>