[Amps] Dummy load varying R

[Mike Tubby]

On 08/06/2011 16:22, Al Kozakiewicz wrote:
> Maybe not quite amps related, but this is the best forum I know of for RF expertise ...
>
> Without belaboring irrelevant details, I'm calibrating an LP-100A wattmeter I just built. With a brand new MFJ 300W dummy load I measure 50.1 ohms with my old Fluke DMM.  The dummy load is connected to the coupler with about 6 feet of RG-8U.  It appears as though the resistance of the dummy load increases with frequency, to the point where it is 65 ohms in the 20m band and the LP-100 can no longer compensate.
>
> After fiddling with the coupler with no joy I decide to connect my MFJ-259 to the cable and verify that the resistance is actually 50 ohms across all the HF bands.  Turns out that it's not.  In the 20m band the impedance at the end of the RG-8 has risen to 66+6j.
>
> This is as simple a setup as I can imagine - a 50 ohm resistor in a metal box and 6 feet of new 50 ohm coax cable.  Tonight I will test again without the cable if I can find a M-M UHF adapter.  But I'm puzzled as to what the issue could be.
>
> Any ideas?
>
> TIA
>
> Al
> AB2ZY

If its MFJ its likely to have been built down to a price and have a 
significant reactive component.

Several thoughts:

1. do an A-B comparison with a *real* dummy load, for example a Bird 
8080 for low power, Bird 8201 for 500W or even a Bird 8251 for 1000W 
(for what it's worth there's someone in California selling new Bird 
8251s on Ebay for sensible money)

2. change the length of coax and see how the impedance moves round the 
smith chart... if you're careful with your selection of length(s) you 
can probably find a point that hides the problem

3. *real* dummy loads have a single large resistor element surrounded by 
a double metal cone that is tapered to create a constant impedance 
taking in to account the length of the resistor - the screen has holes 
or slots in it to allow the oil to circulate

4. if/when you decide that your MFJ isn't all that you hoped it would 
be, either build a load or buy a good one...


Mike G8TIC

PS. While Bird officially state that their dummy loads VSWR at 1.1-1.5 I 
have never had a return loss of worse than 30dB for all frequencies up 
to and including 2m