Elecraft › [K2]

K2 10W PA problem

Classic

List

Threaded

9 messages Options

Guido Tedeschi-2

K2 10W PA problem

A friend asked to me to repair his K2 10W which has a strange problem:
too high current in TX (1.2A even with 0.5W) and current that
increases during key-down until HiCurr protection.
Already changed the 2 x 2SC1969 PA transistors and re-winded the
output transformer, but nothing changed: the voltages on PA and driver
seem OK, the power from driver and the power output remain stable
during key-down, but the current increases...
What could be the issue?
Before spending more time on troubleshooting, has someone had and
solved the same problem?
Thanks and 73
Guido, ik2bcp
______________________________________________________________
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

NQ3RP

Re: K2 10W PA problem

You say the power output stays the same. Is this into a dummy load and measured with a good RF watt meter (like Bird)? If that is the case then due to power staying the same, the resistance the PA sees must be decreasing. Is this happening on all bands or just one or two? This could be related to the LPF causing it if only one or two, need to check those items in the LPF causing it. If it is all bands, I would check the items on the output side of T2 for correct value and soldering.

Don Wilhelm-4

Re: K2 10W PA problem

In reply to this post by Guido Tedeschi-2

Guido,

Did you also change Q11 and Q13? What I suspect is that the bias is
incorrect, and increases during keydown causing the current draw by Q7
and Q8 to increase.

Often Q11 and Q13 are damaged by a PA transistor failure, and if they
are not changed, they will in turn damage the replacement Q7 and Q8.

Where did you obtain the 2SC1969 transistors? I have heard reports of
Chinese transistors that do not meet specification and fail quickly.
Currently, the only source of 2SC1969 transistors that I trust is those
at Elecraft - order K2PAKIT, it includes the PA transistors,
replacements for Q11 and Q13 as well as R50 if needed, along with
mounting hardware for the PA transistors.

73,
Don W3FPR

Guido Tedeschi wrote:

______________________________________________________________
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

Kevin Hibbs

Re: K2 10W PA problem

In reply to this post by Guido Tedeschi-2

Guido,

I had a similar problem with a HiCurr a couple years back working on a
friends K2. The problem turned out to be R67, a 2% tolerance (blue)
resister that controls feed back to the RF power control; which was
burned for some reason. I replaced it and fixed the problem. If the
noise blanker is installed this resistor will be hard to see because
it sits under it. I did not have a problem with increasing current,
but it is worth a quick look just to make sure this is not the
problem.

Best of luck!
73,
Kevin
KG4TEI

On Mon, Jul 12, 2010 at 4:28 AM, Guido Tedeschi <[hidden email]> wrote:

> A friend asked to me to repair his K2 10W which has a strange problem:
> too high current in TX (1.2A even with 0.5W) and current that
> increases during key-down until HiCurr protection.
> Already changed the 2 x 2SC1969 PA transistors and re-winded the
> output transformer, but nothing changed: the voltages on PA and driver
> seem OK, the power from driver and the power output remain stable
> during key-down, but the current increases...
> What could be the issue?
> Before spending more time on troubleshooting, has someone had and
> solved the same problem?
> Thanks and 73
> Guido, ik2bcp
> ______________________________________________________________
> 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
>

W8JI

Re: K2 10W PA problem

In reply to this post by Guido Tedeschi-2

This is almost certainly thermal drift from heat.

> seem OK, the power from driver and the power output remain stable
> during key-down, but the current increases...
> What could be the issue?
> Before spending more time on troubleshooting, has someone had and
> solved the same problem?

Poor heatsinking of transistors will cause this, so make sure he used the
CORRECT very light amount of any heatsink compound and that the transistors
are clean underneath and bolted to correct tightness.

Also a very small upward drift in bias voltage will cause this. This can be
because of leaky series pass bias transistors or bias pass transistor
heating from inadequate heatsinking.

______________________________________________________________
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

Guido Tedeschi-2

Re: K2 10W PA problem

Tom W8JI wrote:

> This is almost certainly thermal drift from heat.
>
> Poor heatsinking of transistors will cause this, so make sure he used
> the CORRECT very light amount of any heatsink compound and that the
> transistors are clean underneath and bolted to correct tightness.
>
> Also a very small upward drift in bias voltage will cause this. This
> can be because of leaky series pass bias transistors or bias pass
> transistor heating from inadequate heatsinking.
>

Many thanks to Tom and the others: I've noticed that PA transistors are
not installed perfectly horizontal and it is possible that the
mechanical contact with the cover is not so good.
This poor heatsinking can explain the drift but, probably, not the 1.2A
with only 0.5W... Maybe, also the two transistors used for bias voltage
are not so healty...
Thanks and 73
Guido, ik2bcp
______________________________________________________________
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

Don Wilhelm-4

Re: K2 10W PA problem

Guido,

Since the normal base to emitter voltage on the PA transistors self
limits at about 0.63 volts, the best way to check the bias voltage is to
remove the PA transistors and measure it. With the PA transistors out,
if the base voltage during a TUNE is between 0.6 and 0.63 volts, then
Q11 and Q13 are doing their job and you have faulty PA transistors.

For proper heatsinking, yes the PA transistors must fit flat against the
heatsink. Before soldering the PA transistors in place, I position them
on their mounting hardware and secure them temporarily with a nut (no
lockwasher, no thermal pad). That places the tab parallel with the RF
board.
Solder the transistors and remove the nuts - add the thermal pads and
mount the heatsink.

73,
Don W3FPR

Guido Tedeschi wrote:

> Tom W8JI wrote:
>
>> This is almost certainly thermal drift from heat.
>>
>> Poor heatsinking of transistors will cause this, so make sure he used
>> the CORRECT very light amount of any heatsink compound and that the
>> transistors are clean underneath and bolted to correct tightness.
>>
>> Also a very small upward drift in bias voltage will cause this. This
>> can be because of leaky series pass bias transistors or bias pass
>> transistor heating from inadequate heatsinking.
>>
>>
> Many thanks to Tom and the others: I've noticed that PA transistors are
> not installed perfectly horizontal and it is possible that the
> mechanical contact with the cover is not so good.
> This poor heatsinking can explain the drift but, probably, not the 1.2A
> with only 0.5W... Maybe, also the two transistors used for bias voltage
> are not so healty...
>
>

Guido Tedeschi-2

Re: K2 10W PA problem

I removed the PA transistors and measured the bias, which is 0.64V and
this should be fine...
Probably, it was only an heatsinking problem, unless the TX current is
wrong: without the PA transistors, with RF Power set to 5.0W (with any
value greater than 0.1W is the same), the TUNE current is 0.62A (0.24A
in RX)
Is it ok 0.62A?
Thanks and 73
Guido, ik2bcp

Don Wilhelm wrote:

> Guido,
>
> Since the normal base to emitter voltage on the PA transistors self
> limits at about 0.63 volts, the best way to check the bias voltage is
> to remove the PA transistors and measure it. With the PA transistors
> out, if the base voltage during a TUNE is between 0.6 and 0.63 volts,
> then Q11 and Q13 are doing their job and you have faulty PA transistors.
>
> For proper heatsinking, yes the PA transistors must fit flat against
> the heatsink. Before soldering the PA transistors in place, I
> position them on their mounting hardware and secure them temporarily
> with a nut (no lockwasher, no thermal pad). That places the tab
> parallel with the RF board.
> Solder the transistors and remove the nuts - add the thermal pads and
> mount the heatsink.
>
> 73,
> Don W3FPR

Don Wilhelm-4

Re: K2 10W PA problem

Guido,

I would like to see the base bias voltage a little bit lower - 0.63
maximum, .62 nominal. I recommend replacing Q11 and Q13.
Yes, the current draw without Q7 and Q8 installed is within the normal
range.

73,
Don W3FPR

Guido Tedeschi wrote:

> I removed the PA transistors and measured the bias, which is 0.64V and
> this should be fine...
> Probably, it was only an heatsinking problem, unless the TX current is
> wrong: without the PA transistors, with RF Power set to 5.0W (with any
> value greater than 0.1W is the same), the TUNE current is 0.62A (0.24A
> in RX)
> Is it ok 0.62A?
> Thanks and 73
> Guido, ik2bcp
>
> Don Wilhelm wrote:
>
>> Guido,
>>
>> Since the normal base to emitter voltage on the PA transistors self
>> limits at about 0.63 volts, the best way to check the bias voltage is
>> to remove the PA transistors and measure it. With the PA transistors
>> out, if the base voltage during a TUNE is between 0.6 and 0.63 volts,
>> then Q11 and Q13 are doing their job and you have faulty PA transistors.
>>
>> For proper heatsinking, yes the PA transistors must fit flat against
>> the heatsink. Before soldering the PA transistors in place, I
>> position them on their mounting hardware and secure them temporarily
>> with a nut (no lockwasher, no thermal pad). That places the tab
>> parallel with the RF board.
>> Solder the transistors and remove the nuts - add the thermal pads and
>> mount the heatsink.
>>
>> 73,
>> Don W3FPR
>>
>
> ______________________________________________________________
> 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
>
>