[R-390] R-390 sensitivity measurements

Bob Camp ham at cq.nu
Wed Jan 19 20:19:08 EST 2005 

Hi

I have run into this problem when measuring commercial gear. Some of 
the FM IF filters have a "dip" in the middle of their pass band. Since 
you measure with a carrier at channel center that's the *worst* 
sensitivity number you can get.

The thing that always has bothered me is that the sensitivity measure 
assumes a 50 or 125 ohm system. In general both the receiver input and 
the antenna output will be significantly different than the assumed 
characteristic impedance.

To many numbers .....

	Take Care!

		Bob Camp
		KB8TQ



On Jan 17, 2005, at 10:45 AM, Lester.veenstra K1YCM wrote:

> One unexplored problem with all the signal generator based sensitivity 
> measurements is that they are dependent on a flat bandpass response 
> for the results to be meaningful. For example. take a 4 kHz filter 
> with 2 dB od pass band ripple. In terms of uV, a two dB difference in 
> measured performance, what would appear to be a significant 
> difference, as possible on the same receiver, as a function of where 
> in the pass band the signal generator is placed.
>
> The solution, although not the most practical in most shops, is to go 
> to a noise generator based system, which will be independent of both 
> IF bandwidth in which the measurement is made and independent of any 
> pass band ripple.
>
>     Les Veenstra
>     K1YCM/3
>      NNN0HWW
>
> Bob Camp wrote:
>
>> Hi
>>
>> As with many things, what matters the most in sensitivity 
>> measurements is that you define how you got your numbers.
>>
>> There are a number of ways to change the setup that will impact the 
>> result that you get. That's neither good or bad in it's self. What is 
>> a problem is a undefined method. If I use method A and you use method 
>> B we really can't be sure that the results are comparable to each 
>> other. If we each have fully characterized and defined (no small 
>> task) our methods then we may be able to guess at a comparison.
>>
>> A lot of restoring these radios comes down to "I got 0.001 uv on all 
>> bands, time to tell the world".  I think that has lead to a number of 
>> conversations about measurement technique ....
>>
>> A far more interesting question - does *any* single sensitivity 
>> number actually tell what's going on in the real world? I would claim 
>> that the answer is no and that is the root of the problem. At that 
>> point (no one number)  we more or less divide into those who give up 
>> on numbers and those who measure a whole bunch of numbers. That makes 
>> for interesting conversations ....
>>
>>     Take Care!
>>             Bob Camp
>>         KB8TQ
>>
>>
>>
>>
>> On Jan 16, 2005, at 10:46 PM, DJED1 at aol.com wrote:
>>
>>> Here's another bit of trivia which I hope will entertain.   In 
>>> looking at
>>> Dallas' other postings, he discusses a method of measuring receiver 
>>> sensitivity.
>>>  From my view, I'm not sure there is an absolute correct way, but 
>>> just for
>>> fun I looked up the MIL spec for the R-390 (MIL-R-13947B) to see 
>>> what the test
>>> requirement was.   It's pretty close, but not exactly what Dallas 
>>> discussed.
>>> The spec has two tests: AM and CW.   The AM is performed by turning 
>>> the
>>> generator modulation on and off, with 30% at 400 cycles modulation.  
>>>  The signal
>>> level is set for 10 mw audio out with modulation, and 1 mw with no 
>>> modulation.
>>> The only difference from his test is that it is done at 8 Kc 
>>> bandwidth, and a
>>> 125 ohm matching resistor is in series between the generator and the 
>>> receiver.
>>>   You still need to do the correction for the generator reading to 
>>> get the
>>> voltage at the receiver.
>>> For the CW case, the setup is the same, except the generator is 
>>> unmodulated,
>>> where signal is carrier on and no signal is carrier off (still in 
>>> the 8 Kc
>>> bandwidth).    The AM spec is 3.3 uv except in the 16-32 bnd, where 
>>> it is 4.4 uv.
>>>   The CW spec is 1 uv.
>>> This, I believe, was the final spec for the R-390 and was used for 
>>> acceptance
>>> testing.   I never did much like the concept of turning the 
>>> modulation on and
>>> off, but this gives that approach significant legitimacy.
>>> Ed
>>>
>>> _____________________________________________________________
>>> Win a new Icom IC-756PROIII and help QSL/QTH.net
>>> Details at: http://mailman.qth.net/index.html
>>> _____________________________________________________________
>>> R-390 mailing list
>>> Home: http://mailman.qth.net/mailman/listinfo/r-390
>>> Help: http://mailman.qth.net/faq.htm
>>> Post: mailto:R-390 at mailman.qth.net
>>> Unsubscribe: http://mailman.qth.net/mailman/options/r-390
>>>
>>
>> _____________________________________________________________
>> Win a new Icom IC-756PROIII and help QSL/QTH.net
>> Details at: http://mailman.qth.net/index.html
>> _____________________________________________________________
>> R-390 mailing list
>> Home: http://mailman.qth.net/mailman/listinfo/r-390
>> Help: http://mailman.qth.net/faq.htm
>> Post: mailto:R-390 at mailman.qth.net
>> Unsubscribe: http://mailman.qth.net/mailman/options/r-390
>>
>>
> _____________________________________________________________
> Win a new Icom IC-756PROIII and help QSL/QTH.net
> Details at: http://mailman.qth.net/index.html
> _____________________________________________________________
> R-390 mailing list
> Home: http://mailman.qth.net/mailman/listinfo/r-390
> Help: http://mailman.qth.net/faq.htm
> Post: mailto:R-390 at mailman.qth.net
> Unsubscribe: http://mailman.qth.net/mailman/options/r-390
>

More information about the R-390 mailing list