[R-390] Measuring sensitivity

[Mark Huss]

Flowertime can probably back me up on this, he has worked on a LOT more 
R-390's than I have. By spec, the R-390A should have a 10dB S+N/N at 4 
uV, 30% modulation, 8 kHz bandwidth, MGC, RF Gain Max, using the little 
Antenna Adaptor/Matcher that comes with the URM-25. In a Training 
environment, where we had the time to fool around with the occasional 
recalcitrant one , the R-390A on the bench for an Annual PM did not 
leave until it did better than 1 uV. That was an easy spec to meet. 
Anything worse than that, there was usually something wrong with it. 
Ether a component going bad, or a weak tube. I have seen traceable 
measurements down around 0.2 uV (probably 2khz bandwidth or less, it was 
not specified). Of course, the problem here is that most of the readings 
you see on the Internet are not traceable (to NIST standards), because 
calibration is expensive. We usually rely on the attenuator being in 
spec and leakages being in spec because we just do not have the 
equipment to calibrate voltages that low properly. Making it worse, 
there are at least three methods to measure S+N/N ratio. All valid. All 
giving different results everything being equal. I learned the Army way, 
I.E. disconnect generator from receiver, measure audio power, reconnect 
generator, dial generator output down to 10 dB greater than first 
reading, Note Generator output level. On the bench, we had a special 
coax switch that in one position, connected the receiver to the 
generator. In the other, it connected the generator to a 50 ohm load, 
and the receiver to another 50 ohm load. It helped reduce the URM-25 
leakage to manageable levels. As good as a URM-25 is as far as leakage 
is concerned, we often had problems getting a good S+N/N reading on the 
lower bandwidths because leakage would screw the readings. Then we would 
finish off by adjusting the IF Gain using Tangential Sensitivity 
readings. A few times I remember going back to do the standard 10dB 
S+N/N test to see if I could measure the improvements, but by then the 
R-390A noise was so low that leakage from the test setup was just too 
great to get reliable readings. Even for a modern receiver, the R-390A 
is equal or better in sensitivity than most general coverage receivers 
out there.

n4buq at knology.net wrote:
> On Mon, 12 Mar 2007 10:19:47 -0400, Mark Huss <mhuss1 at bellatlantic.net> 
> wrote :
>
>   
>> Good point. Being picky about the actual output level of the signal 
>> generator is only important in the following respects;
>>
>>    1. Making valid comparisons between yourself and others.
>>    2. Establishing the receiver is working properly the first time, see 
>>     
> #1.
>   
>> Other than that, it really does not matter if the meter reads 1 uV when 
>> it is actually 3 uV.
>> Sort of like having a twelve inch ruler that is eighteen inches long. 
>> The boards still come out the same length unless you try to mix the 
>> regular ruler with the long one.
>>     
>
> My issues tend to fall in both categories.  Other folks claim to get 
> sensitivities around the 1uV range and the specs say something more like 
> 3uV.  My initial tests were looking like I was getting much less than spec, 
> but without knowing if I was testing it correctly, I wasn't sure if 
> something was wrong with the radio or my tests.  Still not quite sure, but 
> I think it was the later.
>
> Thanks,
>
> Barry - N4BUQ
>
>   


-- 
----------------------------------------------------------------------
Some people are like a Slinky .. not really good for anything,
but you still can’t help but smile when you shove them down the stairs.
----------------------------------------------------------------------