[Spooks] Signal strengths measurements reliability

[Chris Smolinski]

This is in reply to a posting on the e2k mailing 
list. The owner of that list does not allow any 
postings larger than a few lines, since one or 
two of the list members are technically 
challenged and try to read messages using a cell 
phone, so all must suffer.

The Spooks list does allow messages longer than 
two or three paragraphs, so it can be posted 
here. Hopefully those from the e2k list who are 
interested will find the message here to read.

>Hello group,
>
>Whilst preparing the HTML edition of E2K NL 23, a question came to my mind
>about S-metres reliability, and what trust we can give to them.

None. See below ;-)

>It is clear that with the same RX at the same QTH and the same aerials,
>S_metre readings give an accurate idea of signal strength evolution over
>time.

Fair enough.

>Now, if in the middle of a transmission, at the same QTH, we switch-off the
>RX and replace it by a different model of a different brand, will the
>S-metre give the same value ? Will S5 stay S5 and not become S3 or S7 ?
>S-metre values are linked to the electric field generated by the signal
>(usually in the µV/m order) but how caring are manufacturers to respect the
>equivalence ? Can we trust S-metres values at different QTHs in order to
>deduce what direction (if any) a numbers station is beaming its signal to ?

Short answer - no you cannot trust an S-meter.

Long answer:

The S unit system is a logarithmic based system. 
Each S unit is supposed to be 6 dB apart. That 
is, S9 is 6 dB more than S8, etc. This is rarely 
the case in most radios. Not only is it usually 
not 6 dB, it often varies as the signal varies 
from S1 to S9 and beyond.

S9 corresponds to 50 microvolts across the 
antenna input terminals of the radio. But some 
radio manufacturers use different standards for 
S9.

From this, you can deduce the input levels for 
any given signal strength in S units, and 
vice-versa.

Above S9, of course, we generally refer to a signal as "S9+10dB", etc.

>Any eperiment made ?

Yes, take a look at 
http://www.seed-solutions.com/gregordy/Amateur%20Radio/Experimentation/SMeterBlues.htm

Short answer - S meters are good for *relative* 
signal strength measurements. Directly comparing 
readings between two different radios, even if 
using the same antenna, is risky at best. The S 
meter apparently is something added onto the 
radio at the last moment by the manufacturer. 
Most work by measuring the AGC voltage in the 
radio, not by directly measuring the input signal 
level.

Long answer - you could calibrate your S meter, 
much as the author of that web page did. Also 
note from his tests that on a given radio, the s 
meter sensitivity varies from band to band. This 
is to be expected, since the sensitivity of most 
radios is not constant across all of HF.

Now, you can pull out your RF signal generator 
(calibrated against a NIST reference I hope!) and 
calibrate your radio's S meter on all bands, for 
all signal levels. Then do the same for any other 
radio you have.

But also bear in mind that the RF voltage at the 
antenna input terminal of the radio is also a 
function of the impedance of the antenna 
connected to the radio, not to mention the actual 
efficiency of the antenna itself.

There are a large number of variables at work 
here. Hence, an S meter is best used as a 
relative measurement of the signal strength, not 
an absolute one.


-- 

---
Chris Smolinski
Black Cat Systems
http://www.blackcatsystems.com