K2/100 - low power 15m only

[aj4tf]

Using a 50 ohm dummy load.
In tune mode, with KPA100 bypassed (power control to 10W out),  power out is OK on all bands (~10W).  

When the KPA100 is engaged, all bands have full power output except 15m (21 MHz)... it tracks up to about 50W and peaks there.  The other bands on the LPF in the KPA100 (10m and 12m) are OK, getting around 95W out at 100W setting.  

Confirmed with external wattmeter,  it agrees with K2's power reading in tune mode.

I haven't opened the case up yet,  this one has me scratching my head... any ideas out there?

AJ4TF

[Don Wilhelm-4]

If you had the KPA100 simply bypassed, but still in place, try the base
K2 again but turn the power up to 15 watts (to properly bypass the
KPA100 power the base K2 only through the coaxial power jack or turn the
PA menu setting to OFF (secondary menu).
If you get a HiCur message on 15 meters with that setup, check the base
K2 17/15 meter LPF - check for an additional turn on L21 or L22.

If you have full 13 or more watts output (and no HiCur) powering only
the base K2, then look at the KPA100.  Look at all capacitor values in
the input side of the PA transistors as well as the T-R switch area.

73,
Don W3FPR

On 7/6/2012 2:24 PM, aj4tf wrote:

> Using a 50 ohm dummy load.
> In tune mode, with KPA100 bypassed (power control to 10W out),  power out is
> OK on all bands (~10W).
>
> When the KPA100 is engaged, all bands have full power output except 15m (21
> MHz)... it tracks up to about 50W and peaks there.  The other bands on the
> LPF in the KPA100 (10m and 12m) are OK, getting around 95W out at 100W
> setting.
>
> Confirmed with external wattmeter,  it agrees with K2's power reading in
> tune mode.
>
> I haven't opened the case up yet,  this one has me scratching my head... any
> ideas out there?
>
> AJ4TF
>

______________________________________________________________
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

[aj4tf]

Don,

With PA turned off, 15W out on all bands (except 10m, it was about 13W).  No warnings.

I'll take it apart tonight and look at it.

I forgot to mention,  it had been working in the past, so I don't suspect any inductors have changed the number of turns on them.. :-)  but I'll look it over and report what I find (if anything).   Thanks,

73, de AJ4TF

[aj4tf]

OK, here is what I can see so far...

I visually examined the low pass filter components on the KPA100,  between K9 and K10, nothing seems out of the ordinary, nothing cracked, broken, or burned that I can see.

I'm not quite ready to remove the PCA and the power transistors from the heatsink yet.

Measurements:  no change.   50 Ohm load,  tune mode.  Power output on 15m tracks linearly with the power control,  up to about 50W, and then it peaks, even if the knob is turned up further.   The DC current draw is about normal for 50W,  around 10A,  similar to the other bands.  The current never goes any higher as the power control is turned up.

On the other bands, power goes up to full as the power control is turned up.

Any other suggestions before I disassemble it any further?

[aj4tf]

I finally had time to disassemble the circuit board from the heat sink on the KPA100.   I haven't yet found anything suspicious on the 10/12/15 LPF, but what I did find was a 'hot trace' on the +12PA line going to the collector of Q4 (provides the bias to the power transistors).  I wiped the trace with my finger and it peeled off the board.  

I can fix the trace with a white wire, but it makes me wonder if I have more than one problem...

73, AJ4TF

[aj4tf]

After white-wiring a burned trace to the collector of Q4, and resetting the bias current,  it appears to be working on all bands.  It's weird that the apparent lack of bias current caused only one band (21MHz)  to have low power.  
Thanks to Don, and Gary @ Elecraft for all of your suggestions and assistance.
 
73,  David AJ4TF