[Collins] 321B-4 watt meter, accuracy of

[Dr. Gerald N. Johnson]

On Fri, 2007-11-16 at 13:57 -0600, w9hak wrote:
> I have just compared my 312B-4 watt meter against my HP-410C with HP455A 
> rf adapter using an 50 ohm non reactive load. My watt meter reads low on 
> all bands of my 32S3 compared to the E2/50 values. Error values range 
> around 25%, with the watt meter reading low. The transmitter is putting 
> around 110-130 watts output depending on the band. The plate current is 
> being set at 230 ma.
> 
> The 312B-4 hasn't been abused and has been in service since new. It 
> reads reflected power if it's present. My initial thoughts are to leave 
> the unit alone. Absolute power output isn't high on my priority scale. 
> Having said that, what are the members of the reflector doing with their 
> 312B-4/5 control consoles? I would think the 1N82A diodes have seen 
> better days and wonder if anyone has replaced them and the results of 
> having done so.
> 
> Smith Bradford
> W9HAK

Then 1N82A was never a robust diode. Its a diode that can be damaged by
soldering it, surely could be damaged by too much RF or lightning.

Be careful in your measurement that the RF probe isn't changing the load
Z seen by the 312B-4 if the voltmeter is on the load and that the 312B-4
isn't changing the impedance at the voltmeter if the probe is before the
wattmeter.

Then check the calibration resistors, if they are carbon composition, be
very suspicious of them having drifted from age and humidity over the
years. And its not beyond possibility of the meter movement having
gotten stiffer with age. I see in my Yellow Book that the calibration
resistors were chosen at assembly to make the meters read correctly.
Changing diodes could mean needing to readjust all four of those
resistors and changing the type of diode could affect the meter scale at
all powers other than the one calibrated at.

My Yellow Book says the accuracy is 10% of full scale for the 302C
wattmeter. With the 200 watt scale of the 312B-4, that's 20 watts +/-.
Being 25 watts low at 100 watts isn't much beyond that 20 watt range.

I'd be happy with it.
-- 
73, Jerry, K0CQ, Technical Advisor to the CRA
All content copyright Dr. Gerald N. Johnson, electrical engineer