[sdiy] Plotting the phase of a filter with FFT or a Phase Comparator?

[cheater00 .]

Hi Neil,
thanks for the info.
Given that I only want to do phase and amplitude comparisons, I'm
wondering if it isn't a better idea to first off get a Dranetz 305
plus an amplitude comparator and automate that with a DAQ. It'll
certainly be cheaper and I can also make sure the tests do what I
want. I already have a CV sig gen with an integrated counter and a
good external counter as well. Then again 1) it's not FFT, and 2) a
specan "just works". However, if I got THAT setup going, then I could
also use my Monacor 305 impedance bridge going with that, and then I
could also measure impedance of 4-terminal devices over frequency
which could be cool in its own right. That would only be halfway
automatized though (it needs some manual ranging, not sure if I could
somehow jig that up).

On Wed, May 8, 2013 at 2:34 PM, Neil Johnson <neil.johnson71 at gmail.com> wrote:
> cheater00 wrote:
>> Yeah - thanks! Living in Germany is really killing me.
>
> I hope not literally!!

Haha, well no, but it's really terrible what about the options for
test equipment here.

> You can use a stereo card for a basic network analyser: one channel
> samples the input, the other channel samples the output.
> But only useful over the audio range, and at typical audio amplitudes
> (there's no range switching like you get on a proper hardware
> analyser).
>
>>> Oh I don't know ... my HP 35660A can resolve down to microHz if you give it
>>> long enough.
>>
>> Yeah, I have looked at the 35660A and 3561A. Which one would you
>> recommend and why? Is there any "trick" to finding spectrum analyzers
>> that go this low in frequency?
>
> I first used a 3561A.  It's a portable analyser, but only single
> channel.  The 35660A is dual channel, but much larger and not very
> portable.  It is very much a bench machine.  Both have an upper range
> of 100kHz.  The 35660A splits that into 50kHz in dual mode, but you do
> get very precise phase information.  So it's fine for audio use, ideal
> for audio filters :)
>
> There's no trick as such, for very low frequencies you need to go for
> FFT instruments as there's no practical way for analogue to go sub-Hz
> - the settling time as the analyser filter is swept would give a sweep
> time measured in hours.

I wonder how the Tracor and Dranetz work then. Most likely digital.
You mentioned *mechanical* phase comparators... did I misread that or
was that a thing at some point in history?

>> Would you suggest the HP141T as a good spectrum analyzer?
>
> Only if you like ancient machines or want to look like Charles Atlas!
> It's fine for RF use, but for audio the lowest range is 30Hz, and even
> then the sweep times can get very long for narrow bandwidths.
>
>> Can you compare the 3580A to the 3561A? The 3561A does 125uHz-100kHz,
>> but not sure about much more.
>
> 3580A - analogue, slow, covers 1Hz to 50kHz
> 3561A - FFT, much faster, does computation, covers 125uHz to 100kHz.
> 3562A - dual channel version of 3561A
> 35660A - modern version of 3562A, but sample rate reduced to 50kHz in dual mode
> 35665A - as above, but sample rate still 100kHz in dual mode, plus a
> few other extra features


Why would one buy the 35660A instead of the 3562A?

> Also, price varies.
>
>> Again.. would you be able to tell the differences that 3562A brings
>> in? And the 35665A?
>
> Both are good machines.  The 3562A is the older one.  It depends what
> you can find for the price you're prepared to pay.
> As a rough guide, Stewarts of Reading have:
>  a 3562A for £1,500, and a 35665A for £1,850.

Hmm, I'll check them out. It's been some time since I've ordered from
good old UK. I know Stewarts, it's a name that keeps coming back. I
heard good stuff about them but also heard they're a bit pricey. But I
think at least you get your money's worth, which is not (always?) the
case with eBay.

Cheers,
D.