Author Topic: Analog Disovery as Audio Analyzer  (Read 72217 times)

0 Members and 1 Guest are viewing this topic.

Offline timofonic

  • Frequent Contributor
  • **
  • Posts: 904
  • Country: es
  • Eternal Wannabe Geek
Re: Analog Disovery as Audio Analyzer
« Reply #25 on: December 13, 2015, 01:19:16 am »
What about augmenting Sigrok capabilities based on your choice and research? It would be amazing to have these features on this multiplatform and ever expanding software developed by volunteers!
 

Offline rjeberhardt

  • Regular Contributor
  • *
  • Posts: 62
  • Country: fr
Re: Analog Disovery as Audio Analyzer
« Reply #26 on: December 13, 2015, 03:04:11 pm »
Excellent!  I must put an Analog Discovery on my Christmas present list.

Russell.
Retired Chartered Engineer
 

Offline jaxbirdTopic starter

  • Frequent Contributor
  • **
  • Posts: 778
  • Country: 00
Re: Analog Disovery as Audio Analyzer
« Reply #27 on: December 14, 2015, 04:46:58 pm »
Excellent!  I must put an Analog Discovery on my Christmas present list.

Russell.

jaxbird, that is awesome. If you keep this up, and release this software I think I will be buying an Analog Discovery and then send some $$$ your way.

Please make this a "poor man's AP setup" :)

Thanks, much appreciate the positive comments. :)


What about augmenting Sigrok capabilities based on your choice and research? It would be amazing to have these features on this multiplatform and ever expanding software developed by volunteers!

Maybe at some point, but tbh, I've never had a positive experience contributing to existing open projects.

Analog Discovery Projects: http://www.thestuffmade.com
Youtube random project videos: https://www.youtube.com/user/TheStuffMade
 

Offline jaxbirdTopic starter

  • Frequent Contributor
  • **
  • Posts: 778
  • Country: 00
Re: Analog Disovery as Audio Analyzer
« Reply #28 on: December 14, 2015, 04:55:42 pm »
Awesome is a good word for your audio analyzer suite. I was perfectly happy with the capabilities of my Analog Discovery, but now I'm really glad that I own one (although my trusty HP334A is looking a little nervous that it could be departing my main test bench soon to free up some much needed space).

Also, a quick thought about your IMD feature. If you see some value in this idea, could the amplitude of the two test tones be made independently adjustable so that you could, for example, test a vintage piece of audio gear using the older SMPTE standard (frequencies of 60Hz and 7KHz, mixed in a 4:1 ratio) if that was desired?

That could make an interesting way to compare the test results of the various IMD standards by supplying the appropriate test frequencies and amplitude ratios.

I tip my hat to you... keep up the excellent work.

Thanks, the HP334A is a nice instrument, but it does require a bit a patience to adjust it correctly.

I've made some minor changes that gives more flexibility on the IMD testing, the level of the 2nd tone is now set as a percentage of the level of the 1st tone. That will allow the tests you described.



Always open to suggestions.

Analog Discovery Projects: http://www.thestuffmade.com
Youtube random project videos: https://www.youtube.com/user/TheStuffMade
 

Offline jaxbirdTopic starter

  • Frequent Contributor
  • **
  • Posts: 778
  • Country: 00
Re: Analog Disovery as Audio Analyzer
« Reply #29 on: December 14, 2015, 05:10:27 pm »
Also added the Output Impedance feature. It requires 2 resistors, just use some within the range of realistic speaker impedance for best results. Could do open vs load, but I feel the two resistor method is better.

Here from measuring the same little class D board, using a log value scale because of the rise at high frequencies, but lin/log is optional.



Analog Discovery Projects: http://www.thestuffmade.com
Youtube random project videos: https://www.youtube.com/user/TheStuffMade
 

Offline bson

  • Supporter
  • ****
  • Posts: 2266
  • Country: us
Re: Analog Disovery as Audio Analyzer
« Reply #30 on: December 14, 2015, 07:03:33 pm »
I'd suggest an LTspice experiment with a simple op-amp driving an 8ohm resistive load; measure the THD+N.  Then replace the resistor with an inductor and parasitic resistor of say 0.1ohm that gives you |Z|=8ohm at 10kHz.  Check the THD+N.  Then repeat at say 2ohm and 200ohm.  This will teach you resistive loads are useless for design purposes and their only use with audio amplifiers is to generate fantastic-looking numbers for marketing collateral.  (Other than smoke testing of course, without risking destruction of real drivers.)
 

Offline fpliuzzi

  • Regular Contributor
  • *
  • Posts: 157
  • Country: us
Re: Analog Disovery as Audio Analyzer
« Reply #31 on: December 14, 2015, 07:08:12 pm »
jaxbird, your change to the IMD test instrument allows great flexibility in setting up the test parameters for the various present, and possibly some future IMD standards (better than just having a couple of presets for the most commonly used ones). It also allows for experimentation, which is always interesting. Thank you for implementing this feature.
 

Offline fivefish

  • Frequent Contributor
  • **
  • Posts: 440
  • Country: us
Re: Analog Disovery as Audio Analyzer
« Reply #32 on: December 14, 2015, 07:31:55 pm »
jaxbird: Are you using Analog Discover or Analog Discovery 2? 

Side-Topic: What's the advantage of AD2 vs. original AD?  Will Waveforms 2015 run on the original AD?

I'm deciding what kind of AD to get to run jaxbird's software :)
 

Offline jaxbirdTopic starter

  • Frequent Contributor
  • **
  • Posts: 778
  • Country: 00
Re: Analog Disovery as Audio Analyzer
« Reply #33 on: December 15, 2015, 06:26:25 am »
I'd suggest an LTspice experiment with a simple op-amp driving an 8ohm resistive load; measure the THD+N.  Then replace the resistor with an inductor and parasitic resistor of say 0.1ohm that gives you |Z|=8ohm at 10kHz.  Check the THD+N.  Then repeat at say 2ohm and 200ohm.  This will teach you resistive loads are useless for design purposes and their only use with audio amplifiers is to generate fantastic-looking numbers for marketing collateral.  (Other than smoke testing of course, without risking destruction of real drivers.)


In essence what you are saying is that there is no point in standardized testing of amplifiers as using resistive loads are not equal to speaker loads.

I cannot say I agree, surely many valuable performance details are revealed using resistive loads. Based on these results it's even possible to make good predictions about the performance when combined with a specific speaker simply by looking at the speaker's impedance behavior, frequency response/dispersion pattern, sensitivity etc. And by using standardized tests with standardized loads the results are comparable.

Analog Discovery Projects: http://www.thestuffmade.com
Youtube random project videos: https://www.youtube.com/user/TheStuffMade
 

Offline jaxbirdTopic starter

  • Frequent Contributor
  • **
  • Posts: 778
  • Country: 00
Re: Analog Disovery as Audio Analyzer
« Reply #34 on: December 15, 2015, 06:28:01 am »
jaxbird, your change to the IMD test instrument allows great flexibility in setting up the test parameters for the various present, and possibly some future IMD standards (better than just having a couple of presets for the most commonly used ones). It also allows for experimentation, which is always interesting. Thank you for implementing this feature.


Thanks, it was a great suggestion  :-+. I will probably still add some general purpose spectrum analyzer screen with optional tone generator for more experimentation, but I like to have purpose specific tools for most standard tests.

Maybe at some point even add a scope/tone gen with the option to view signal with the fundamental notched out similar to what you'd get from many of the traditional distortion analyzers.

Analog Discovery Projects: http://www.thestuffmade.com
Youtube random project videos: https://www.youtube.com/user/TheStuffMade
 

Offline jaxbirdTopic starter

  • Frequent Contributor
  • **
  • Posts: 778
  • Country: 00
Re: Analog Disovery as Audio Analyzer
« Reply #35 on: December 15, 2015, 06:33:17 am »
jaxbird: Are you using Analog Discover or Analog Discovery 2? 

Side-Topic: What's the advantage of AD2 vs. original AD?  Will Waveforms 2015 run on the original AD?

I'm deciding what kind of AD to get to run jaxbird's software :)

Mine is the v1, but even v1 got different hardware versions, mine is revision C or D I think. But they are all compatible afaik, even with version 2. From a quick look I believe the major differences is that v2 got more analog bandwidth (30MHz+ vs 15MHz+) and a variable output power supply (v1 is fixed +/-5V, 50mA limited). Waveforms 2015 runs just fine with my v1.

I'd get a v2 if I need to get another one.

If you get one you might be interested in this other project I did based on the Analog Discovery:
https://www.eevblog.com/forum/projects/impedance-analyzer-build-and-experiments/135/

Analog Discovery Projects: http://www.thestuffmade.com
Youtube random project videos: https://www.youtube.com/user/TheStuffMade
 

Offline jaxbirdTopic starter

  • Frequent Contributor
  • **
  • Posts: 778
  • Country: 00
Re: Analog Disovery as Audio Analyzer
« Reply #36 on: December 15, 2015, 07:13:39 am »
And just for fun, why not :) ... this is a single ended mono amplifier, need two for stereo, capable of something like 5-10W output. Analog Discovery for scale and my little class D test amplifier board next to it.

I don't actually use these amplifiers as they don't produce much watt and I don't have any appropriate high efficiency/gentle impedance speakers for them, but to be honest, their coolness factor is hard to beat, so they do have a prominent display in my living room. I originally got them quite cheap as they were not working well, so I've done some work and modifications to get them into good working order. It was during this repair work I thought I needed something to easily measure and diagnose an amplifier resulting in the software I'm now doing for the Analog Discovery.



Analog Discovery Projects: http://www.thestuffmade.com
Youtube random project videos: https://www.youtube.com/user/TheStuffMade
 

Offline mtdoc

  • Super Contributor
  • ***
  • Posts: 3575
  • Country: us
Re: Analog Disovery as Audio Analyzer
« Reply #37 on: December 15, 2015, 08:45:36 am »
jaxbird: Are you using Analog Discover or Analog Discovery 2? 

Side-Topic: What's the advantage of AD2 vs. original AD?  Will Waveforms 2015 run on the original AD?

I'm deciding what kind of AD to get to run jaxbird's software :)

Mine is the v1, but even v1 got different hardware versions, mine is revision C or D I think. But they are all compatible afaik, even with version 2. From a quick look I believe the major differences is that v2 got more analog bandwidth (30MHz+ vs 15MHz+) and a variable output power supply (v1 is fixed +/-5V, 50mA limited). Waveforms 2015 runs just fine with my v1.


I suspect there is no significant difference between hardware versions.  Scope bandwidth on V2 is advertised as 20 MHz with BNC adapter on the NI website and 30MHz on Digilent's site. The spectral characteristics diagram for V2 on it's wiki page is the exact same as the one on page 16 of the V1 technical reference manual. In fact the V2 wiki looks to just be a web adapted version of this manual.  Digilent made a silly marketing decision with V1 to market the scope as 5 MHz because they "did not want to confuse students" who might not understand the concept of a -3dB bandwidth.  You can read this on page 16 of the technical reference manual.

I think once Digilent was bought by National Instrument, NI decided to put the AD in a new case and call it V2 with a more accurate bandwidth advertised. (though I question the 30 MHz bandwidth given the 100 MSPS for both scopes ADC, regardless of the -3dB measurement) 

I'll be happy to be proven wrong but I think the only additional hardware of note is the addition of external power input to augment the power supply - the least useful feature of the AD IMO.    Basically I think V1 = V2 just without the fancy case, ext power input and marketing fluff.
 

Offline fivefish

  • Frequent Contributor
  • **
  • Posts: 440
  • Country: us
Re: Analog Disovery as Audio Analyzer
« Reply #38 on: December 15, 2015, 09:02:26 am »
jaxbird, big kudos to you on this project.

More questions: I assume the software will work with line level signals?

How about what if I need to measure DUT gear with balanced input/output, I'm thinking I'll need to convert the signals from the AD to balanced and vice versa before feeding it to the DUT gear. Have some kind of breakout box with the balanced converter circuits before the AD box? will that screw up the readings?

-- how about a "normalize" function, say measuring freq response, have the software automatically normalize the resulting curve to 0dB reference.

 

Offline bson

  • Supporter
  • ****
  • Posts: 2266
  • Country: us
Re: Analog Disovery as Audio Analyzer
« Reply #39 on: December 15, 2015, 09:36:58 am »
In essence what you are saying is that there is no point in standardized testing of amplifiers as using resistive loads are not equal to speaker loads.
Yes.  Did you follow the suggestion to run a simulation with different kinds of loads?
 

Offline jaxbirdTopic starter

  • Frequent Contributor
  • **
  • Posts: 778
  • Country: 00
Re: Analog Disovery as Audio Analyzer
« Reply #40 on: December 18, 2015, 05:53:18 am »
jaxbird, big kudos to you on this project.

More questions: I assume the software will work with line level signals?

How about what if I need to measure DUT gear with balanced input/output, I'm thinking I'll need to convert the signals from the AD to balanced and vice versa before feeding it to the DUT gear. Have some kind of breakout box with the balanced converter circuits before the AD box? will that screw up the readings?

-- how about a "normalize" function, say measuring freq response, have the software automatically normalize the resulting curve to 0dB reference.

The Discovery got 2 differential inputs accepting up to +/-25V with 2 attenuation levels, to measure larger amplifiers more attenuation is needed.

For balanced signals, the differential inputs are fine, but for balanced output it would be best to add something like this:



You'd just need to check the Ref 1kHz box to set reference level at 1kHz level, otherwise it displays the voltage gain referenced to input level.
Analog Discovery Projects: http://www.thestuffmade.com
Youtube random project videos: https://www.youtube.com/user/TheStuffMade
 

Offline jaxbirdTopic starter

  • Frequent Contributor
  • **
  • Posts: 778
  • Country: 00
Re: Analog Disovery as Audio Analyzer
« Reply #41 on: December 18, 2015, 06:14:22 am »

I suspect there is no significant difference between hardware versions.  Scope bandwidth on V2 is advertised as 20 MHz with BNC adapter on the NI website and 30MHz on Digilent's site. The spectral characteristics diagram for V2 on it's wiki page is the exact same as the one on page 16 of the V1 technical reference manual. In fact the V2 wiki looks to just be a web adapted version of this manual.  Digilent made a silly marketing decision with V1 to market the scope as 5 MHz because they "did not want to confuse students" who might not understand the concept of a -3dB bandwidth.  You can read this on page 16 of the technical reference manual.

I think once Digilent was bought by National Instrument, NI decided to put the AD in a new case and call it V2 with a more accurate bandwidth advertised. (though I question the 30 MHz bandwidth given the 100 MSPS for both scopes ADC, regardless of the -3dB measurement) 

I'll be happy to be proven wrong but I think the only additional hardware of note is the addition of external power input to augment the power supply - the least useful feature of the AD IMO.    Basically I think V1 = V2 just without the fancy case, ext power input and marketing fluff.

I wasn't even aware that NI had bought Digilent, but I guess they fit in nicely in the low end compared to Ni's more high end products.

I think you are right, they could have made some minor adjustments to the front end (just component values as the AD8066 should have enough bandwidth).

The added power supply features are probably worth the extra $20. I just had a look at microchipdirect.com, they used to sell the original for just over $200 to non students, but now it's the same $259 as everywhere else.

Analog Discovery Projects: http://www.thestuffmade.com
Youtube random project videos: https://www.youtube.com/user/TheStuffMade
 

Offline jaxbirdTopic starter

  • Frequent Contributor
  • **
  • Posts: 778
  • Country: 00
Re: Analog Disovery as Audio Analyzer
« Reply #42 on: December 18, 2015, 07:48:12 am »
In essence what you are saying is that there is no point in standardized testing of amplifiers as using resistive loads are not equal to speaker loads.
Yes.  Did you follow the suggestion to run a simulation with different kinds of loads?

You seem persistent, so let's do a little experiment, instead of a simulation, to find out how much difference there really is and what it looks like.

Here is a comparison between using a typical small 2 way speaker with an impedance between 5 Ohm and 20 Ohm, compared to a 6 Ohm resistive load.

2 Way speaker's impedance for reference:



6 Ohm load resistor's impedance: (not very exciting)



Distortion measurement with Speaker load:



Distortion measurement with Resistive load:





The only difference here is that the amplifiers distortion is actually overall lower with the speaker load ;) Especially at the frequencies where the speakers impedance is higher.
Analog Discovery Projects: http://www.thestuffmade.com
Youtube random project videos: https://www.youtube.com/user/TheStuffMade
 

Offline timofonic

  • Frequent Contributor
  • **
  • Posts: 904
  • Country: es
  • Eternal Wannabe Geek
Re: Analog Disovery as Audio Analyzer
« Reply #43 on: December 18, 2015, 08:30:32 pm »
269USD? I thought it would cost 100USD as much...

Is his really interesting?
 

Offline miguelvp

  • Super Contributor
  • ***
  • Posts: 5550
  • Country: us
Re: Analog Disovery as Audio Analyzer
« Reply #44 on: December 18, 2015, 08:34:58 pm »
There is also the NI edition for students only:

http://sine.ni.com/nips/cds/view/p/lang/en/nid/213328

It does come with Multisim Student Edition as well.
 

Offline jaxbirdTopic starter

  • Frequent Contributor
  • **
  • Posts: 778
  • Country: 00
Re: Analog Disovery as Audio Analyzer
« Reply #45 on: December 19, 2015, 12:26:00 am »
269USD? I thought it would cost 100USD as much...

Is his really interesting?

I agree, it's a good chunk of money, IMO for the features it's worth the cost, it's got 14 bit resolution on the 100Msps ADCs, DACs, plus 100Msps 16ch logic in/out. It's like a little Swiss army knife, with some patience it will do most jobs, but it can not compete with a dedicated instrument at any specific task.

And of course, like any instrument, it's only worth it if you got a use for it, like e.g. you don't already have an arbitrary function gen or you don't already have a logic analyzer/generator or high res scope, low frequency spectrum analyzer or network analyzer etc.

If it's just another toy that will start collecting dust after a week, then it's probably not worth the money :)

Analog Discovery Projects: http://www.thestuffmade.com
Youtube random project videos: https://www.youtube.com/user/TheStuffMade
 

Offline jaxbirdTopic starter

  • Frequent Contributor
  • **
  • Posts: 778
  • Country: 00
Re: Analog Disovery as Audio Analyzer
« Reply #46 on: December 19, 2015, 12:29:45 am »
Last feature for now, I think, input impedance measurements:




Then I just need to clean it up, do a bit of testing and bug bashing.

Analog Discovery Projects: http://www.thestuffmade.com
Youtube random project videos: https://www.youtube.com/user/TheStuffMade
 

Offline fivefish

  • Frequent Contributor
  • **
  • Posts: 440
  • Country: us
Re: Analog Disovery as Audio Analyzer
« Reply #47 on: December 19, 2015, 02:41:14 am »
Looking forward to the release of this software jaxbird. Then I'll get an AD.

Excellent work.

Hey, how hard is it to do a "normalize"... i.e. the software takes a base reading, normalize it to flat response, then put the DUT, and let the software do it's thing (relative to the normalized base reading)?
 

Offline jaxbirdTopic starter

  • Frequent Contributor
  • **
  • Posts: 778
  • Country: 00
Re: Analog Disovery as Audio Analyzer
« Reply #48 on: December 23, 2015, 02:19:04 pm »
Looking forward to the release of this software jaxbird. Then I'll get an AD.

Excellent work.

Hey, how hard is it to do a "normalize"... i.e. the software takes a base reading, normalize it to flat response, then put the DUT, and let the software do it's thing (relative to the normalized base reading)?

Thanks, but you should only invest in the Analog Discovery if you really feel you have a use for it, no obligations :)

I'm putting a lid on new features until a release is available, after that I'll start tracking bugs and feature requests.

I've been preparing the software for distribution, primarily to make it easy to distribute and update, so I've gone with a web deployment strategy with update check on startup. It will still run fine offline, but if online and there is a new version available on startup it will suggest to update. Not that I expect it to be full of bugs, but I'm sure there will be many updates initially so easy updating is fairly high priority.

Also it needs a bit of explanation on how to configure connections for the various measurements, some measurements require various shunt resistors. So I'm compiling a document for reference and perhaps a video.

Anyway, I expect to have an install ready within the next couple of days.

Analog Discovery Projects: http://www.thestuffmade.com
Youtube random project videos: https://www.youtube.com/user/TheStuffMade
 

Offline jaxbirdTopic starter

  • Frequent Contributor
  • **
  • Posts: 778
  • Country: 00
Re: Analog Disovery as Audio Analyzer
« Reply #49 on: December 24, 2015, 09:53:32 am »
Made a couple of minor changes:

The power vs THD+N now includes the input level. It's nice to know what input level is required e.g. when clipping occurs. I'm considering adding a scope view as it's nice to know the balance, as in, if one rail is clipping prematurely.



Also, I've referenced the frequency response at 1kHz level by default, it will still show the voltage gain at cursor position:




Release will be shortly after Christmas (once I get past the traditional hangover :))

Analog Discovery Projects: http://www.thestuffmade.com
Youtube random project videos: https://www.youtube.com/user/TheStuffMade
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf