Analog Disovery as Audio Analyzer

[Audioguru]

The little low power Trident amplifier was designed and patented 19 years ago. They called it Class-T instead of Class-D. It cuts the highest audio frequencies.
Trident went bankrupt 9 years ago so I do not know if the ICs are new-old stock or are made recently by a Chinese copy-cat company.

[jaxbird]

.....stuff deleted by jaxbird....

Just a couple of questions.

Presumably those plots were taken from the voltage on the amplifier output, not from the sound from the speaker.

If the speaker's impedance varies at different frequencies then doesn't the power dissipation and thus output vary significantly?

If you plotted the real power dissipated in the speaker, you'll probably find it varies hugely as the frequency is changed. Does that mean the output will be distorted by the same factor?

These two measurements were taken under the exact same conditions, except one was using a precision 4 ohm resistive load and one was using the speaker shown (a PSB alpha B 2 way speaker, manufactured in Canada, you can find a comparable impedance measurement done by stereophile in NY)

[jaxbird]

Hello,

I installed it on my laptop and could not find a RUN button, so I assumed it was a screen width issue (win7).  I installed it on my main PC and still could not find a run button (win7).  I then installed it on my win10 machine and still no RUN button...  am I doing something wrong?

Sorry. there was a problem with Win 10, it should be fixed now, please try again and let me know if you have any issues.

Thanks

[jaxbird]

This thread was recommended from another forum.  I'm excited to see where this leads...  Jaxbird, keep up the good work.

Thanks, appreciate the encouragement, new version available, enjoy.

[jaxbird]

Same issue here,no Run button. Tried on Win 10 x64 and Win7 x64. AD unit works fine with Waveforms on both machines.

Dave

My bad, it was some auto scaling issue, trying to comply with all these 4k screens and keep the same size, should be fixed, try update or new download.

Let me know if you still have issues.

Thanks

[jaxbird]

The little low power Trident amplifier was designed and patented 19 years ago. They called it Class-T instead of Class-D. It cuts the highest audio frequencies.
Trident went bankrupt 9 years ago so I do not know if the ICs are new-old stock or are made recently by a Chinese copy-cat company.

True, but it really is a bridged class D amplifier, not horrible, but as Tripath probably was taken over by Cirrus Logic before mine was produced, it is highly likely it is a copy, but hard to say whether the copy or the original performs better. I can say that it does seem to have it's switching frequency starting around 800khz

[jaxbird]

Anyone trying the software, please try restart, it should update your current version, I was only testing on Win 7 and it was not the same scaling as on Win10, I apologize for that and blame my former employer Bill Gates.

[jaxbird]

as an added bonus for your great mood and bravery, the general purpose spectrum analyzer should be functional in the latest version.

[jaxbird]

Let me know if you still cannot run the basic software, thanks.

The current version should say 1.0.0.62, if not, or if you still have issues with this version, please let me know.

[Mark]

Still no joy jaxbird...    I tired it on Win 10 tablet and win 7 desktop. 
It must be a resolution issue, the text along the top is truncated, for example "Total Harmonic Distortion and"
Bottom left text is truncated to "Spectru" and Bottom right text is truncated to "Scop D AC" and still no run button. 

[Mark]

Screenshot attached from Win7 x64 desktop.  My screen resolution is 1920 x 1200. 

Anything I can try? 

Thank you!

[mtdoc]

Same issue here. Windows 7, x64 laptop. Screen resolution 1366 x 768.

[jaxbird]

Still no joy jaxbird...    I tired it on Win 10 tablet and win 7 desktop. 
It must be a resolution issue, the text along the top is truncated, for example "Total Harmonic Distortion and"
Bottom left text is truncated to "Spectru" and Bottom right text is truncated to "Scop D AC" and still no run button.

Same issue here. Windows 7, x64 laptop. Screen resolution 1366 x 768.

Still no joy jaxbird...    I tired it on Win 10 tablet and win 7 desktop. 
It must be a resolution issue, the text along the top is truncated, for example "Total Harmonic Distortion and"
Bottom left text is truncated to "Spectru" and Bottom right text is truncated to "Scop D AC" and still no run button. 

I thought I had it fixed, but new version available 1.0.0.63, hope it really is fixed this time. I believe it was caused by my laptop being set to scale 125% during compile.

Thanks for having patience everyone.

[mtdoc]

That fixed it!  At least on my machine. Thanks for all your efforts jaxbird!

[fpliuzzi]

  Just a quick update to state that v1.0.0.64 is up and running on the Win XP laptop that's sitting on my bench (nice to see that elusive RUN button). Many thanks for this very useful tool. It's going to be used quite frequently.

[Selectech]

Working fine now on Win 7 X32 with ver 1.0.0.64

I'll try iit on Win 10 X64 in the morning.

A great piece of test software - thanks. It will get a lot if use.

[Mark]

Working fine now on Win 7 X32 with ver 1.0.0.64

I'll try iit on Win 10 X64 in the morning.

A great piece of test software - thanks. It will get a lot if use.

Same here but Win 7 x64, I'll try win 10 on my tablet tomorrow. 
Thanks jaxbird! 

[jaxbird]

Just a quick update to state that v1.0.0.64 is up and running on the Win XP laptop that's sitting on my bench (nice to see that elusive RUN button). Many thanks for this very useful tool. It's going to be used quite frequently.

That fixed it!  At least on my machine. Thanks for all your efforts jaxbird!

Working fine now on Win 7 X32 with ver 1.0.0.64

I'll try iit on Win 10 X64 in the morning.

A great piece of test software - thanks. It will get a lot if use.

Same here but Win 7 x64, I'll try win 10 on my tablet tomorrow. 
Thanks jaxbird! 

Great!... thanks for the feedback, good to hear it's working.

I have added a page for reporting bugs and requesting features: thestuffmade.com/audioanalyzersuite/bugs

If anyone wants to contribute to a better help file, feel welcome to volunteer .. all I can offer is credit though, as it is a free piece of software.

Let me know if anyone got success with the AnalogDiscovery2 version, should work afaik, but always nice to get a confirmation.

Next I plan to add some more flexibility on the configuration when I have some time, currently most is fixed drop down selection, I'll add free text in those with capability to use short form like 2k or 2M etc and some more error checking on inputs etc.

Cheers

[hugocoolens]

OK, here is a download link:

http://www.thestuffmade.com/audioanalyzersuite/publish.htm

I'm interested in knowing about any bugs you find, pm me or just post here in this thread. I haven't fool proofed the software yet, so if you write e.g. "dave" in start frequency, it will likely blow up. Or if you try run multiple measurements at the same time it will blow up, but in general it should be usable.

I am only able to test with my version of the Analog Discovery, so any reports on compatibility are appreciated.

Note: the spectrum analyzer is not yet implemented. If that annoys you, I can hide the tab. Feature should be available before new year.

Dear Jaxbird,
Is it also possible to run your software using Linux?

kind regards,
Hugo

[jaxbird]

Dear Jaxbird,
Is it also possible to run your software using Linux?

kind regards,
Hugo

It will probably happen at some point, it's c# code, so should be possible to build against the Mono framework for Linux, but I'll have to rewrite the native wrapper for the Analog Discovery API, so definitely significant work involved.
 
Maybe Wine, I'm not sure what state it's in these days, but might be possible to use under Linux. If you find a way let me know.

[halej1]

Jaxbird- this is Awsome! I'm new to EEVblog-- and have seen some of the Audiophoole/Audio Geek bashing. Some of it is well deserved but frankly much of the electronics world is represented by audio and audio analysis. which is why I'm here. Actuallyout of necessity I'm am an Audio guy with limited funds (like all of us) and my gear is aging and failing, so I've had to start learning repairs and testing (albiet I'm new to that part).  Really interested in your work and want to use it to analyse some of my gear (sub amps, amps, processers). I have recently purchased (ebay cheap) a Keithly 2015. two reasons- 1) A great bench multi meter 2) the ability to Do THD measurements.  Not so much for the ansolute measure but to see if my amps are providing rated power.  Sorry such a long lead-in but my question is:

When you connect your output signal to test equipment do you attenuate the signal?  Basically input would be 0-2V ish and output may get to high voltages for high output amps.

Anyway glad there are audio folks on the Blog as well.

Thanks

[jaxbird]

Jaxbird- this is Awsome! I'm new to EEVblog-- and have seen some of the Audiophoole/Audio Geek bashing. Some of it is well deserved but frankly much of the electronics world is represented by audio and audio analysis. which is why I'm here. Actuallyout of necessity I'm am an Audio guy with limited funds (like all of us) and my gear is aging and failing, so I've had to start learning repairs and testing (albiet I'm new to that part).  Really interested in your work and want to use it to analyse some of my gear (sub amps, amps, processers). I have recently purchased (ebay cheap) a Keithly 2015. two reasons- 1) A great bench multi meter 2) the ability to Do THD measurements.  Not so much for the ansolute measure but to see if my amps are providing rated power.  Sorry such a long lead-in but my question is:

When you connect your output signal to test equipment do you attenuate the signal?  Basically input would be 0-2V ish and output may get to high voltages for high output amps.

Anyway glad there are audio folks on the Blog as well.

Thanks

Thanks, the goal for this project was to provide a cheaper alternative to the already reasonably affordable older HP etc audio analyzers, but adding more features. Performance level should be about the same, it will never get anywhere near the Audio Precision equipment level, but certainly good enough to tell whether an amplifier is faulty and help locate any faults. IMO in reality these 0.000001% specs don't really matter, I am confident no living person can successfully identify the difference between 0.01% and 0.000001% distortion in a true blind test.

There is a big difference between audiophoolery and actual audio engineering, many inexperienced people are confused by this difference, thinking everything is audiophoolery and hoping to gain community recognition by calling it out.

The immediate limitations are that the Analog Discovery will start clipping at +/-25V input, so that does limit the usability with large amplifiers, but I can easily add a feature to use any division factor, as in add any attenuation (couple of resistors) at the inputs and just enter the factor into the software to scale everything. I can make a board with auto attenuation up to 1000W if anyone is interested.

Other limitation is that the outputs are limited to near +/-5V peak (single ended), anything higher than that will need external amplification, but I think most amps will never require more than for full wattage output. Balanced output will also require some external components.

Great thing is that the inputs are differential (not floating, but within limits), so measuring things like bridged amplifiers is not an issue.

This solutions is much more useful than a Keithley 2015 for amplifier characterization.

Cheers

[jaxbird]

Here is a good example of using this software to check the performance of an amplifier, a slightly ridiculous 300B single ended tube amplifier in this case:

[halej1]

Jaxbird - still awesome! and thank you for the input to my question. Your software/hardware solution looks really good!  I watched your video and was wondering --- you are able to look at the graphs and pretty easily tell if things are good or not so good. Obviously pretty simple on something like THD or THD+N but can you share or direct me to where I can find out, for example -- what is acceptable output impedance (tube, solid state) or input impedance (tube, Solid state) etc...  I guess what I'm asking is lot's of people can probably set the system up and get graph but the skill comes with the interpretation.

It may be a little while before I spring for an Analog Discovery and your software -- cause I just blew my budget on an Oscilloscope (Rhode and Schwarz 1202) -- Yeah a bit overkill for this type of analysis but I wanted to treat myself (the FFT function is supposed to be very good, though)

Cheers
John

[ci11]

Hi,

I can't get the download to work - got a 'Fatal Error - Unable to find "dwf.dll"'. What am I doing wrong? My PC in Win 7 x64, .Net 4.6.

Thanks.