SINAD measurement procedure

[kcbooboo]

I ran some tests on a UHF MSF5000 today. I connected an ACVM to the speaker output on the control jack and set the volume level with PL disabled so the meter indicated 0 dBm. I then fed an unmodulated signal into the receiver and noted the signal level required for various amounts of dB quieting (usually called dBQ). My measurements exceeded the specs in the Motorola manual: 0.5uV for 20dBQ and my receiver only needed 0.335uV for 20dBQ. This same receiver breaks squelch, with a PL-modulated signal, around 0.110uV. This station has the internal duplexer (filter) and all of these measurements were taken with the sig gen feeding the station's antenna connector, NOT the raw receiver input.

I do NOT have a SINAD meter, but I believe to make those kinds of measurements I need a modulated signal, so I set the sig gen for 3 kHz deviation of a 1 kHz tone, cranked the signal level way up, and used that as my reference level. I then removed the modulation and made the same quieting measurements as before. I ended up with exactly the same signal levels as I did when I used just squelch noise as the reference level.

It is my understanding that a SINAD meter notches out the 1 kHz tone and essentially measures what's left. Since I removed the modulation at the source, I think this is why I got exactly the same readings in each test.

Can anyone expand on the way a SINAD meter really works, and is there a preferred method of measuring SINAD without one? In other words, did I come up with a reasonably good alternative here?

Thanks.
Bob M.

[Hartley]

Hello Bob,

Yes and no..

You are correct about how a SINAD meter works - it measures the total audio in the output, then notches out the 1000 Hz input modulation and then compares what's left to the original - which comes not only from noise, but also distortion products introduced by the IF filters, etc. This is why it is considered a better measurement than a simple quieting measuerment - and, if you have a SINAD meter, rather easier to perform. Using a SINAD meter and adjusting the deviation of the source signal allows you to easily see the effect as you approach the edge of the IF filters.

But this means that without such a meter, you cannot make the same measurement by measuring the degree of quieting. You can see that by making the bandwidth of a receiver narrower, you could get very good quieting measurements, but lousy SINAD (and actual performance!)

73 DE Hartley

[xmo]

"...You can see that by making the bandwidth of a receiver narrower, you could get very good quieting measurements, but lousy SINAD (and actual performance!) ..."
_______________________________________________________________

Hartley's observation is exactly why they came up with the SINAD procedure for two way radios. That reasoning was very valid at the time, however today's radios have virtually no alignment of the critical low IF stages, so the chances are the 20dBq measurement procedure is all you will ever need for maintaining something like an MSF or Quantar.

Today we are really looking to verify proper operation of the unit - and possibly align the receiver's front end. Then you listen to the radio's performance on a tone modulated test signal. It doesn't take long before you become pretty good at determining if there is a problem.

If you want to play with SINAD measurement - the cheap entry is a SINAD meter from Helper Instruments. Motorola also used to make one. They show up pretty often on ebay.

Most service monitors [except really old ones] have SINAD capability built in.

[Wowbagger]

SINAD and quieting are trying to measure much the same thing, but in different ways. Each has its plusses and minuses.

Quieting can be done with nothing but an RF generator and a voltmeter - thus it is easier to do with simpler equipment. However, it involves turning a signal on and off, so it is harder to make quick measurements, and harder to automate.

SINAD requires a notch filter and an audio signal generator - so it takes a bit more gear to do (however, most service monitors have all this built in.) However, since there is no switching of a signal on or off, it can be done more rapidly, and is more ameniable to automation.

For example, if you are adjusting a radio and trying to monitor the changes in sensitivity, it is easier to set up a SINAD measurement, and then focus on tweaking the radio, and watching the updates on the meter. Doing a quieting test would require you to:
1) Tweak
2) Make signal on measurement
3) Turn off signal.
4) Make signal off measurement.
5) Repeat as needed.

IMHO, that is why you see quieting tests speced on older gear, and SINAD on newer gear - the older gear's proceedures were writting for the tech with a VTVM and RF sig gen, the newer proceedures assume a service monitor.