Login  Register

Re: Funny power measurements

Posted by Jack Brindle on Jul 24, 2006; 5:58am
URL: http://elecraft.85.s1.nabble.com/Funny-power-measurements-tp392406p392408.html

The math is already more complex than most folks realize. The output  
of the diode detector is not RMS, but rather peak. So, you need to  
multiply it by SQRT(2) to get the real voltage to work with. On top  
of that, you should add in the diode drop, which is variable based on  
the current going through the diode, which, of course, is based on  
the RF voltage. 1N4148s have rather high voltage drop that can  
greatly alter the output voltage. We use 1N5711 diodes in our power  
meters since they have a much lower forward voltage drop than the  
1N4148s. Even still, at low power levels they add in some error to  
the calculations.

I suspect the load isn't as critical as others believe, as long as it  
is within a few percent of the "estimated" load value. Yes, the best  
accuracy will be with a purely resistive load at 50 ohms (if that is  
what you need the load to be), but remember that other component  
tolerances (including those in the voltmeter) will cause variances in  
the readings. Still, if you are careful, you should be able to get  
amazingly good results.

It appears that if Fran were to compare his calculated results with a  
well calibrated watt meter, he may find his results off by a factor  
of two, not including the diode drop!

On Jul 23, 2006, at 7:57 PM, Don Wilhelm wrote:

> Fran,
> Secondly, the accuracy of your power measurement is highly  
> dependent on the
> dummy load.  If you have a good antenna analyzer, check the dummy  
> load at
> all frequencies of interest - if the dummy load is truly good, it  
> will have
> a reactance of zero and a pure resistance of 50 ohms.  A precision  
> dummy
> load should be within 1% of those values - if yours is otherwise,  
> you can
> take steps to calculate the true power, but if there is any  
> evidence of
> reactance in the load, the math becomes non-trivial.

- Jack Brindle, W6FB
------------------------------------------------------------------------
---------------------


_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
 http://mailman.qth.net/mailman/listinfo/elecraft   

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com