Ian White, G3SEK wrote:
> Gary Schafer wrote:
>
>>
>>
>> Ian White, G3SEK wrote:
>>
>>> GW4FRX made some very useful tests using a borrowed analyser with a
>>> digital peak hold facility. By continuing to sweep across the whole
>>> signal for several minutes, holding the highest signal level found at
>>> each frequency step, the analyser builds up a good statistical
>>> picture of the overall bandwidth that the signal occupies.
>>> The composite spectrum is composed of IMD products from all speech
>>> frequencies, and it generally looks like a ragged triangle with the
>>> higher-order IPs disappearing below the noise. The longer you sweep,
>>> the better the statistics, and the smoother-edged the triangle becomes.
>>>
>>
>> Hi Ian,
>>
>> The problem with peak hold analyses is that a particular peak only has
>> to hit once and it becomes the same class of data as a frequency that
>> is almost constantly present.
>> It will tell you all the possible frequency "hits" that occurred but
>> may give a false impression as to problems.
>>
> It does give equal weighting to splatter that only occurs on transient
> peaks of the modulating waveform. But transient IMD could for example
> include splatter that occurs on almost every word, which most other band
> users would classify as annoying. Peak hold is a severe test, but not an
> unfair one.
Spectrum analyzers that have peak hold often have different means of
averaging also. Actually your 141 kind of averages when using the long
persistence. It can be done several ways. I am still not sure that it is
the best answer though.
Using peak hold will show a signal that has only an occasional hit (that
may be hardly audible) Just as bad as a signal that has many hits in the
same place. Comparing that way would show both signals to be as
offensive while in reality one may be much more offensive than the
other. Example: a station that only occasionally flattops compared to
another that is constantly flat toping.
>
>> Broadcast stations have different standards to adhere to than
>> amateurs. A broadcast station must have a certain "frequency contour"
>> that the signal conforms to. It must not transmit anything outside
>> that contour. Sample and hold analyses is how that is measured.
>>
> Do you have access to any standard test procedures? We need to avoid
> re-inventing the wheel here.
What I am referring to here is the audio spectrum of an AM station. I am
not that familiar with it but as I understand the modulation power is
limited to certain amounts as you get further from the carrier.
>
>>
>> Probably the simplest and most repeatable way to monitor inter mod is
>> to just switch your receiver to the opposite side band. Im products
>> are going to fall on the opposite side band as well as on the wanted
>> side band. On the opposite side band you don't have the wanted signal
>> components to interfere with hearing the IM products.
>>
> I thought we had agreed that levels of low-order IMD (which is what you
> find in the opposite sideband) give no indication of the levels of
> high-order IMD. That's why we need a test that goes outward to look for
> high-order IMD.
>
Not me. While I will agree that higher order products do not necessarily
track lower order levels, there is usually a pretty good relationship.
Some products can be generated at audio and some at RF. That is going to
make a difference in higher level relationships. Other factors may come
into play also.
But to try and establish a simple means of detecting IM products on a
signal I think that the low order products are the easiest to detect and
the most consistent. (especially where there are strong adjacent
signals) If there are low order products that are strong it is a good
bet that there are higher order products present with a fairly
consistent relationship in strength. I am not talking about real
accurate measurements here but ballpark levels to determine the severity
of the problems with signals.
One problem with trying to get others to clean up there signals is that
a dirty signal helps the guys receiver. The guy with the dirty signal
has much less qrm to put up with! It is a kind of "if you do it I will too".
73
Gary K4FMX
|