instrument for audio signal analysis

[singapol]

They are different animals, to my mind (never had either, so I go by spec sheets etc.). I'd use the 3562A to visualize the content of the audio spectrum (invaluable, surely), while with the 8903A you can obtain similar information, but all numerical: distortion, S/N, RMS AC measurements, even DCV if one needs it, etc. They may look at the same content, but their output I see useful in fundamentally different ways. Also, it's true that the 3562A has a signal source, but if I go for another DSA/SA, it may not have it. Besides, I've learnt to appreciate having multiple instruments with overlapping duties, which is never truly superfluous. But seriously, never mention this to my wife.

I'm looking forward to feedback on this though; I am still figuring out my needs, including (instrumentally) through this conversation.

You might be interested in Standford Reasearch SR760/770 SA similar to HP3562a but only 1 channel and
-90db dynamic range vs -130db hp3562a but 770 has usb albeit older standard. I think SR760/770 does not have AM/FM modulation but it does have signal source. There is also SR780.

http://www.testequipmentdepot.com/usedequipment/hewlettpackard/spectrumanalyzers/3562a.htm

http://www.thinksrs.com/downloads/PDFs/Catalog/SR760770c.pdf

[deadlylover]

I'm not sure if the piece of gear I'm considering right now - HP 8903B - allows for this. In any case, I may be contemplating getting an audio analyzer (such as HP 8903B) and a DSA/SA....

I've been on the lookout for the HP 8903B, but not such luck here down under. I *think* a few simple mods to the oscillator section will make it perform quite decently.

Keep an eye out for the Panasonic VP-7722 because it's relatively unknown and you might score a deal, I freaking love mine and it's probably the best $300 I've ever spent on test gear. -112dB(0.00023%) THD+N @ 1kHz/30kHz BW, not too shabby at all eh for hobbyist use.

[davorin]

Hmm...from the HP 3562A datasheet it looks like it can't do THD and IMD measurements...or do I miss something?

Have you had a look at the QA400 from Quantasylum?

https://www.quantasylum.com/content/Products/QA400.aspx

Not going up to 100kHz due to 192kHz sample rate.
I bought one cause I found the combination of USB soundcard and commercial software in the same price range as the QA400.
The newer QA401 and QA405 will be more expensive though, but feature isloated inputs/outputs and differential signals...

[GK]

The AWA F242A is a reasonably sophisticated auto-tuning analyzer that pops up on ebay on a semi-regular basis; generally around the $200 - $250 mark. Since I've built my own analyzer with better performance I have no need for one, but have procrastinated over buying one a few times now after stumbling across examples on ebay.

Manual here:
https://isurplus.com.au/manuals/AWA%20F242A%20User%20Manual.pdf

I just nicked this image from a current ebay auction:

[Alex Nikitin]

As hardware analysers I use the Amber 3501 "Distortion and Noise Measuring System" and occasionally the HP3563A . The Amber has an excellent low distortion generator with THD about -115dB at midrange (400Hz) and the best SNR about 105dB in 30kHz BW and ~100dB in 80kHz BW. The most useful feature is the distortion only output for the scope, which allows you to see the distortion only waveform. Also it can be used as a good TRMS voltmeter and a selective amplifier.

Cheers

Alex

[Kjelt]

Is it my superstition or is it still bad to use a PC to sample analogue signals? Only the bad powersupply and noise inside would give me shivers.
I whould think it is a two stage job, input and signal generation hardware fully standalone analogue build and then the data transfer to a pc that does the math and presents the GUI.
Nobody on this forum build his own version of a audio precision system one or are the soundcards and pc's that good you can reach the same specs ?

[fivefish]

I have an 8903B, useful for getting numbers... not a spectrum analyzer though, so for frequency response visualization, I use FuzzMeasure and my MOTU 828mk2 audio interface.

[Rax]

I have an 8903B, useful for getting numbers... not a spectrum analyzer though, so for frequency response visualization, I use FuzzMeasure and my MOTU 828mk2 audio interface.

WhoaaAaaw!... It seems I just snatched one exactly like this. Somewhat unfortunate - though it probably helped with the price - with rear hookups like you have it ("option 001" I guess it is). How did you put together that installation there to bring them to the front? Looks pretty neat. It's briefly mentioned here: http://www.neurochrome.com/test-equipment/hp-8903/, but I will have to get the list of parts together in an order.

Thanks!

[Rax]

As hardware analysers I use the Amber 3501 "Distortion and Noise Measuring System" and occasionally the HP3563A . The Amber has an excellent low distortion generator with THD about -115dB at midrange (400Hz) and the best SNR about 105dB in 30kHz BW and ~100dB in 80kHz BW. The most useful feature is the distortion only output for the scope, which allows you to see the distortion only waveform. Also it can be used as a good TRMS voltmeter and a selective amplifier.

Cheers

Alex

Hi Alex,
I just discovered this (before reading your post), and I am quite impressed. It's sort of like an HP339a but better... One of the good things about it is that, even if in a tighter spec'd category, it retains the needle meter, which is useful at finding the minimum output when adjusting a filter, for instance, etc. There are some things easier done with an old school meter.

It also puts little damage in one's pocket, relative to what it does.

[fivefish]

>How did you put together that installation there to bring them to the front? 

It's just a DIY 1u front panel with BNC pass-through connectors, and some short BNC patch cables connected to the rear of 8903B then to the 1u front panel.  It adds a few decimal points in the readings (but almost negligible), but the convenience it brings outweighs that.

[kc8apf]

Hmm...from the HP 3562A datasheet it looks like it can't do THD and IMD measurements...or do I miss something?

It can definitely do THD.  See pages 19 and 8-18 of HP 3562A Operating Manual.  The signal source isn't capable of two-tone generation so they don't include an IMD feature.  If you use an external two-tone generator, you'll definitely see the IMD visually and can use the markers to measure it.

[vk6zgo]

The AWA F242A is a reasonably sophisticated auto-tuning analyzer that pops up on ebay on a semi-regular basis; generally around the $200 - $250 mark. Since I've built my own analyzer with better performance I have no need for one, but have procrastinated over buying one a few times now after stumbling across examples on ebay.

Manual here:
https://isurplus.com.au/manuals/AWA%20F242A%20User%20Manual.pdf

I just nicked this image from a current ebay auction:

Back in the day,somebody thought it would be a nice idea to replace  our manual AWA N&Ds with HP auto units.
Well,the HPs were a major disappointment!

Probably good in a Lab,but when you have limited time to do multiple tests,as is common in Broadcasting,their slo-o-o-www nulling made it a hassle.
The only way to use them was to manually null,then let the auto do the last few "tweaks".

We thought all our birthdays had come at once when we got the AWAs---fast auto null,equal or better specs.
The HPs got shut away in the equipment cupboard!

[Rax]

OK, time for some updates.

Since my last post on this tread, I received my 8903B and got a chance to test it and all that. I must have ran into the nicest seller on the entire bidding site, because they not only replaced my purchased unit (they have a large stock at hand) with one with front hookups (no option 001), but also with one having all filters I wanted (400Hz, A-weighing, 30kHz, and 80kHz - all available for this unit that are most useful in my opinion), in pretty much perfect shape, and - best of all - with current calibration until spring of next year!... They even sent me the long form calibration certificate. This is in fact much better than I expected. The price was far from excessive also.

Anyway, my decision to get this is in a juncture with getting myself geared up for doing the audio spectrum analysis on my 17" laptop (I gave in on this one...). I am thinking that maybe later I will be able to afford an HP 3562a or similar to have this analysis done in a more calibrated manner, so to speak. For now, my task is to get myself a suitable USB sound card and maybe also build Pete Millett's interface kit to work with it, and I think I have this nailed down for now.

Thanks all for your input - extremely useful guiding for me in this endeavor.

[fivefish]

Congrats. If yours didn't come with the hardbound operation and calibration manual, I'd suggest also buying one. 

[Cubdriver]

Excellent - glad to hear that you found something!

And I'll second the suggestion to get the manual.

-Pat

[Rax]

Thanks a lot, fivefish and Cubdriver!

I am confused on your recommendation, though, as the Keysight website has all of the manuals for this instrument (and pretty much all historical HP instruments) available for free as PDFs. I have the one I need the most (Op. & Calib.) printed out.

[rsjsouza]

I found interesting nobody mentioned the Keithley 2015. Just curious, as I am not an expert in the audio analysis field: how different is the Keithley from the animal you bought?

[Cubdriver]

Thanks a lot, fivefish and Cubdriver!

I am confused on your recommendation, though, as the Keysight website has all of the manuals for this instrument (and pretty much all historical HP instruments) available for free as PDFs. I have the one I need the most (Op. & Calib.) printed out.

As long as it's legible you're good to go, but I've found that a lot of those scans are less than optimal as far as quality goes - especially for things like the schematics, which in the factory manuals are often multi-page foldouts and color coded, and display screen reproductions (which don't really apply here upon further reflection, since the 8903B doesn't have a display screen, just digital readouts and an analog meter).  Don't get me wrong - I think it's wonderful and a great thing that Agilent/Keysight have the scans for so many instrument manuals posted and easily available, and have downloaded plenty of them myself for my own instrumentation, but I find it (for me at least) best to try to get a real hardcopy of an original manual if possible.  I find doing cut-and-pastes with the multi-page schematic printouts to be a royal PITA

YMMV.

-Pat

[Rax]

I found interesting nobody mentioned the Keithley 2015. Just curious, as I am not an expert in the audio analysis field: how different is the Keithley from the animal you bought?

Not a specialist either, but the first observation would be that the Keithley seems to kind of hit a different price bracket. Not by much, and on some sites where you can find both used they may really not differ much at all, but enough to be substantive for a decision (at least for me). Additionally, I think it's easier today to find a good deal on an HP 8903B than you typically can on a Keithley 2015.

From the specs, I think the HP offers a bit more consistent performance across the board and is a bit better in some critical points (the Keithley is better in some other points). It may also be geared more specifically towards audio measurements.

Then again, certainly, the Keithley could replace a bunch of instruments on my - chronically lacking room... - bench.

Maybe guys having used both may pitch in here with more hands on information than what I have.

[RNA]

I have been using the transfer function a lot on my hp 3582a.  In this video I use it to examine the hippos filter mod for the GSSL bus compressor.  Very useful.