The point! We have all missed the point! :-)
I do not dispute the science.
The question was not "what is the correct science", rather, "what is
expected by the FCC (or any other spectrum authority) for successful
processing of the submittal. It became apparent that the science did not
match the requirements. Is the requirement scientifically pure and
correct? -- no. Is the requirement specified? -- yes.
I noticed no one commented on the impedance of free space in the far field
specified for use in the reactive near field given as another example of
"convention". Again is the convention correct? -- no. Specified? -- yes.
Submittal convention is not about correct application of science; it's about
following specified rules, for whatever the reasons. The reasons usually
have to do with simplifying the submittal process while providing repeatable
results, this being more important than totally correct science. Allowances
have been made for the lack of perfect science.
As a partial explanation of the origin of the current instruction being
debated, the following is an excerpt from another response regarding duty
cycle reporting for a Part 15.209 device:
> the FCC reviewer had explained to me that he had problems
> > in the past with the interpretation by applicants for using averaging
> detectors so he preferred to mathematically arrive at
> > the average voltage. He asked for the peak detector output in units of
> uV to be multiplied
> > by the duty cycle to provide the "mathematical equivalent
> > of the average detector" in his words. I have been doing this ever
> since
> > for various reviewers and no submissions have come back.
>
2.38 mV pk detector, 0.5 duty cycle on an average detector = 1.66 mV
per vbw averaging per the experiment in this thread below.
2.38 mV x 0.5 (duty cycle) = 1.19 mV per the mathematical process
specified.
The 2 methods are clearly not equivalent. But FCC convention requires me to
submit "1.19 mV" data, which is also arrived at by the 20 log() operation
(hence the confusion about the sanity check earlier). I don't create the
convention; I just follow the submittal instructions (requirements).
It is always good to have the correct science at your fingertips, but in the
end providing what is requested is what counts. For those still in the dark
about 20 log () or 10 log (), my suggestion is to go straight to the source
-- ask the FCC what they specify for your situation.
Don Umbdenstock
Sensormatic
> ----------
> From: Ken Javor[SMTP:ken.ja...@emccompliance.com]
> Sent: Monday, October 29, 2001 2:41 PM
> To: umbdenst...@sensormatic.com; emc-p...@majordomo.ieee.org;
> dmck...@corp.auspex.com
> Subject: The Trouble with Convention, The Final Chapter
>
> In the face of all the responses I and others gave last week showing the
> MATHEMATICAL RULES for calculating logarithms and average and peak power,
> and the rationale and math behind pulse desensitization calculations,
> apparently it is still not clear that power is averaged, not potential.
> In
> the interest of stopping the flow of incorrect test reports to the FCC and
> their apparent approval, I submit the following, "A single test is worth a
> thousand expert opinions." For those who are confused and don't know what
> to believe, here is the simple test. I have run it and have pix of the
> spectrum analyzer display I can send to those who are interested (no
> attachments allowed on general mailings).
>
> Tune an rf source and your spectrum analyzer to a common frequency, say 10
> MHz. Set up a baseline rf output, say -40 dBm. Use linear mode on your
> analyzer with a reference level of -37 dBm. Measure in peak detect mode
> (vbw greater than or equal to rbw). Measure again while averaging, using
> a
> 300 Hz vbw. You will get the same result. Apply a 50% duty cycle pulse
> modulation at 1 kHz. Peak detect will show the same -30 dBm, but video
> averaging will show a 3 dB decrease to -33 dBm. Try it - I did and it
> works. On the analyzer I used, linear mode defaults to a millivolt
> reading.
> I got 2.38 mV in peak mode, 1.66 mV in average mode - you do the math.
>
>
>
> ----------
> >From: umbdenst...@sensormatic.com
> >To: emc-p...@majordomo.ieee.org, dmck...@corp.auspex.com,
> ken.ja...@emccompliance.com
> >Subject: The Trouble with Convention
> >Date: Mon, Oct 22, 2001, 4:46 PM
> >
>
> > Similarly, it appears the same issue of convention is the basis of
> certain
> > FCC clauses, for example, the reporting of the output of an averaging
> > detector as called for by 15.209 and other clauses for some frequency
> bands.
> > The FCC is looking for field strength, a voltage representing the output
> of
> > the averaging detector. The FCC is aware that there are different
> > implementations of "averaging" detectors and linearity issues so they
> > provided instructions to arrive at the reporting level by mathematical
> means
> > for consistency. The instruction was to multiply the peak detector
> reading
> > by the duty cycle and report this value in terms of the limit units, uV,
> as
> > the equivalent of the output of the averaging detector.
> >
> > So, what is this unit they asked for? It appears to be the function of
> > averaging a voltage signal, i.e., if the signal is X and is on for y,
> 0<y<1,
> > then the value of the signal to be reported is y*X (uVolts).
> >
> > When it is desired to address multiple factors (distance correction,
> > antenna, cable, preamp, etc.), the process is simplified by converting
> to
> > log terms. By the relationship between P and V, we have developed the
> > expression for V in log terms to be 10*log V^2 or 20*log V. Addressing
> the
> > entire expression above, we have 10*log (y*X)^2, or 20*log(y*X). This
> can
> > also be expressed as 20*log (y) + 20*log(X). From this expression we
> see
> > that duty cycle is expressed as 20*log (y) for this situation.
> >
> > It appears in this case the FCC is looking for average voltage, not
> average
> > power.
> >
> > Speaking of power, don't forget the power of the sanity check :-)
> >
> > Best regards,
> >
> > Don Umbdenstock
> > Sensormatic
>
-------------------------------------------
This message is from the IEEE EMC Society Product Safety
Technical Committee emc-pstc discussion list.
Visit our web site at: http://www.ewh.ieee.org/soc/emcs/pstc/
To cancel your subscription, send mail to:
majord...@ieee.org
with the single line:
unsubscribe emc-pstc
For help, send mail to the list administrators:
Michael Garretson: pstc_ad...@garretson.org
Dave Heald davehe...@mediaone.net
For policy questions, send mail to:
Richard Nute: ri...@ieee.org
Jim Bacher: j.bac...@ieee.org
All emc-pstc postings are archived and searchable on the web at:
No longer online until our new server is brought online and the old
messages are imported into the new server.
