[Elecraft] K3/100 input voltage and TX IMD

[Jan Erik Holm]

Rick, and others

Just to clarify things a bit.
First of all when I did get the "bad IMD" report on the
air I was at a different QTH and used a another PS, in
this case I do not know exactly what voltage K3 got or
voltage drop.
Then I did bring K3 home to a PS that had 13.9V RX and
13.5V at TX, didn´t reflect much about this. I then started
to measure IMD on the K3 and it didn´t look all that
bad up to 80W but at 120W it was much worse.
After that I started to "rant" about different things here
on the reflector, did get more or less zilch help from
Elecraft sorry to say.
Received some good ideas from some nice people here on the
reflector. That together with some studies of my own
led to bias adjustment on the KPA100 and also I did
turn up PS voltage to 14.9V at RX. Also made area of cable
between PS and K3 much much larger, at TX I now have
14.6V at the K3 display. My IMD measurements now looks
very good up to 100W and at 120W only slightly slightly
worse.
Since I didn´t get much help from Elecraft I don´t feel I owe
them anything but still to be fair to Elecraft it actually
looks slightly better now compared to the "known good"
FT-1000D. You can look at these analyzer pictures,
Ft-1000D at 200W and K3 at 100W. It actually looks very
good.
http://sk3w.se/sm2ekm/view_photo.php?set_albumName=SM2EKM-pictures&id=FT_1000D
http://sk3w.se/sm2ekm/view_photo.php?set_albumName=SM2EKM-pictures&id=P1010014
Sorry but it´s difficult to hold the camera still.
Now I have to test it on the air. Thanks a lot to the
individuals that gave a helping hand, I appreciate it.
I hope this is the end of the story.

73 Jim SM2EKM
--------------------
Rick Shindley wrote:
> A solid state linear RF PA will generate more IMD if the supply voltage 
> sags on current peaks. How much distortion vs voltage change depends on 
> the amplifier design and components used.
> 
>  
> 
> Someone reported here that he increased the supply voltage a bit and 
> noted a 1 to 2 dB reduction in TX IMD.  He also reported a 400 mV drop 
> in the supply at the back of his K3/100 during Tx.  Increasing the 
> supply voltage but still having a 400mV “droop” and yet seeing a slight 
> reduction in TX IMD hints at the sensitivity of the relationship between 
> supply voltage and TX IMD.
> 
>  
> 
> I measured the voltage drop at the Power Pole connector at the back of 
> my K3/100 while transmitting 100W CW out on 14.350 MHz into a dummy 
> load. I found a 270mV drop due to the standard length power cable from 
> Elecraft (IR loss) and 20mV due to the supply (Astron RM-35) load 
> regulation capability for a total “droop” of 290mV.    
> 
>  
> 
> So on SSB voice peaks or CW at 100W there is at least a 290 mV (and some 
> say 400mV) drop in the supply voltage to the radio.  It may even be a 
> bit more inside the rig at the PA due to connector contact resistance 
> and trace routing on the PCB. The slight drop in voltage will contribute 
> to the amount of distortion in the output.  Could this be the cause of 
> the reported variations in measured TX IMD among rigs?    
> 
>  
> 
> Rick
> 
> KC0OV
>