Amps
[Top] [All Lists]

Re: [Amps] Checking for IMD

To: "'Tom W8JI'" <w8ji@w8ji.com>, <amps@contesting.com>
Subject: Re: [Amps] Checking for IMD
From: "Gary Schafer" <garyschafer@comcast.net>
Reply-to: garyschafer@comcast.net
Date: Mon, 14 Aug 2006 20:37:26 -0400
List-post: <mailto:amps@contesting.com>

> -----Original Message-----
> From: amps-bounces@contesting.com [mailto:amps-bounces@contesting.com] On
> Behalf Of Tom W8JI
> Sent: Monday, August 14, 2006 7:20 PM
> To: amps@contesting.com
> Subject: Re: [Amps] Checking for IMD
> 
> > The wrong things shouldn't be exaggerated then. We should
> > focus on what
> > really happens.
> 
> That's what I'm doing.
> 
> > The S meter doesn't measure "peak power". A more correct
> > statement would be
> > it measures the "average power of the peaks" (as in peak
> > envelope power).
> > But it is immaterial as it is the same meter that is being
> > used to measure
> > the fundamentals as is used to measure IM levels here.
> 
> Not any meter I've seen. They almost always work from the
> AGC.
> The AGC is generally a peak detection system with rapid
> attack and slow decay, with the exception that a good weak
> signal AGC will ignore very brief peaks.
> An AGC that used averages would be terible on CW and SSB, as
> would one that hung on very short duration peaks. There is
> often group delay in the AGC attack caused by time delay in
> filters so there might be some overshoot if the AGC controls
> stages before the narrow filters.

How the S meter works has little to do with the measurements being discussed
but it is interesting. I agree that most meters are driven from quasi peak
detectors from the agc but do not directly respond to true peak power. If
they did then only one rf cycle of energy could be reliably read. It takes
several cycles of rf energy to stabilize the meter to a point where it gives
any sort of representative information. In my opinion the nearest thing I
can relate it to is responding to peak envelope power which is an average
power reading.

> 
> I do agree the type of detection shouldn't be a major issue
> in most receivers if you reset the meter level to the same
> value and kept it low on the scale.
> 
> > "misses some of the third order and much of the higher
> > order products" is a
> > misleading statement too. In reality there is a sampling
> > of all of those
> > products with voice being used as it is a wide spectrum of
> > tones generated.
> > The resultant IM products will be represented in a rather
> > narrow band.
> 
> I don't follow that.

In a voice bandwidth of a 3000 Hz transmitter filter as an example there is
a chance for any set of frequencies with any spacing to be generated between
300 and 3300 Hz. As we discussed earlier the 3rd, 5th, 7th order products
can be produced from that bandwidth that can fall in the adjacent channel
(one filter spacing away). So I am saying that with voice modulation at
least some of each of those products will fall within that bandwidth and
will be heard.

> 
> > "probably isn't useful on the air except under certain
> > conditions". No test
> > of any kind is useful under all conditions. There are
> > always limitations.
> 
> That's what we need to learn.

The way it was stated you seemed to imply that there is slim to no chance it
is useful for anything. Or did I take that wrong?

> 
> >> Who's going measure the dynamic range of various radios
> >> using S meter movement?
> >
> > If someone measured their transmitter with their poor
> > performing receiver
> > and thought they had poor IM performance on their
> > transmitter maybe that
> > would be a good thing. That may prompt some further action
> > into
> > investigating why.
> 
> It could be a false good or a false bad reading.

A false good from a receiver with poor IM performance would be caused by??

> 
> > Seems better than to just throw up our hands and say "it's
> > too complicated
> > to try we're going to get the wrong answer".
> 
> I'm only saying it is necessary to know how the system
> behaves before saying it is OK or not OK. We have to know
> what it does and what it cannot do and explain it in a way
> everyone understands, including limitations. Why would that
> be a problem?

That would be no problem at all. That is why I am questioning some of your
broad statements of why it won't work. 

73
Gary  K4FMX

> 
> 73 Tom


_______________________________________________
Amps mailing list
Amps@contesting.com
http://lists.contesting.com/mailman/listinfo/amps

<Prev in Thread] Current Thread [Next in Thread>