[Test-Equipment] PTS250 - was General Radio module

[Fuqua, Bill L]

This is a basic mathematical principle. That also is why Heisenberg’s Uncertainty Principle is a principle and not a law.  Time and frequency measurement folks understand this very well given that they have to deal with it every time they do a frequency measurement test.
I have some demonstrations that I present to my students to help them understand this. Mostly using a modulated signal and bandpass filters. But no matter how you do it you can’t know for certain what a frequency is until you have measured it (observed it) for a minimal amount of time. 
If you think you know a frequency given a short measurement time interval you are only guessing because you can’t see into the future. 
If you instantaneously change frequency, phase or amplitude of a sinusoidal signal  it becomes broad, we experience this in terms of key clicks. It does not become well defined until a time later and that time is determined by how well you are trying to resolve its frequency.
73
Bill wa4lav
 

________________________________________
From: test-equipment-bounces at mailman.qth.net [test-equipment-bounces at mailman.qth.net] On Behalf Of gandalfg8 at aol.com [gandalfg8 at aol.com]
Sent: Saturday, December 31, 2011 12:40 PM
To: test-equipment at mailman.qth.net
Subject: Re: [Test-Equipment] PTS250 - was General Radio module

I'm still away from home and using an unfamiliar PC that's having some formatting problems with AOL mail so not sure how this might display, but I'm either missing something here or the "basic mathematical principle" suggested below is just not relevant in this instance.

Just as modern frequency counters certainly don't require anywhere near a 10 second gate time to measure to 0.1Hz accuracy, albeit for different reasons, a frequency synthesiser such as those in the PTS series does not require 10 seconds to generate a signal to 0.1s resolution.
I believe PTS quote something like 20microseconds for frequency changes to stabilise to within 0.1 radians, and although I can't confirm this I can be sure that to all intents and purposes it's nigh on instaneous.

Similarly, the HP 3336, to quote just one further example, offers frequency resolution into the microHertz region but again frequency changes settle very quickly.

So far I can see no obvious reason why one or more extra DM modules couldn't be inserted into the PTS synthesiser chain but if there is a limiting factor I expect it to be other than suggested below, as that would also limit the units as they stand.

regards

Nigel
GM8PZR




I just went to the office for a few minutes and checked. I have two PTS250
anuals there (they belong to the NMR lab) and they are older ones using DM1000
odules.
  The question gets down to a very basic mathematical principle.
eltaF*DeltaT>1 .
or example to resolve 1 Hz you must count for 1 or more seconds. For .1Hz you
ust count for 10 or more seconds.
nd you can say that the actual frequency resolution is limited by the time that
he signal exist. That is a signal that
xist for 1 second has a minimal bandwidth of 1 Hz. It can't have a bandwidth of
ess than a Hertz.  Simply because it
as not been around long enough to have enough zero crossings. A difficult thing
o wrap your mind around at first but
fter studying it for a while you will understand.
his relates to another one "HUP"  DeltaE*DeltaT>h
ou get the second from applying the first and  E=hF.
ither you will recoginize the relationships or not.

______________________________________________________________
Test-Equipment mailing list
Home: http://mailman.qth.net/mailman/listinfo/test-equipment
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:Test-Equipment at mailman.qth.net

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html