QuantAsylum QA400 and QA401

I've been drooling over the QA400 for a number of months now for electronics measurement, but then I realized that it would be great if I could also hook up a mic (through a preamp of course) and make speaker measurements as well. That way I don't have to get a separate (and very similar) audio interface to do the speaker work.

I can't find any information on the feasibility of speaker measurements with the QA400. Can anyone tell me if one is limited to the QA software or if it is possible to use something like REW with the QA400 for measurements?
 
Member
Joined 2004
Paid Member
The QA400 is specifically not recommended for speaker work. Its not setup to deal with acoustics so you can get a lot of misleading results. Look for a setup with chirps, windowing and mike cal files as basic to measurements of speakers.

QA has mentioned supporting speaker measurements but its a lot of software stuff to do. The box won't talk to other software so its not useful except with the QA software.
 
The QA400 is no better than a good sound card. QuantAsylum's aim was to remove the uncertainty of windows audio. The QA software bypasses windows audio eliminating the possibility of a user or other software from making level changes or altering the QA400's calibration. The QA400 front end is just a digital audio codec. Nothing special. The rest is in software and the communication between hardware and software, which is most likely handled by an FPGA.
 
Headphone THD measurement

Just a couple of THD measurements I did with a QA400 connected to a headphone amp, they indicate quite clearly that the QA400 cannot realiably measure anything below -100dB as has been noted earlier in this thread.

The upper table shows the QA400 connected to a blameless-style headphone amplifier. The load is a divider made from a 560ohms resistor and a 39 ohms resistor. Output level on the QA400
was -7dBV; the amplifier has a gain of 16 and puts roughly 7,15Vrms into 600 ohms; the divider attenuates the signal to approx. -7dBV again.
Second table shows the QA400 in loopback-mode.

The third table shows the difference of each harmonic calculated as "loopback minus amplifier".
Positive numbers thus indicate that the harmonic coming out of the amplifier was actually *lower* than in loopback mode whereas negative
numbers show distortion that is presumedly produced by the amplifier. The rightmost column is just the sum over all substracted harmonics (note that @3kHz the amplifier output seems to have significantly lower THD
than the DAC that feeds it - now how's that possible?)

Amplifier distortion and internal distortion in the ADC/DAC seem to partially add or cancel. -120dBV would be a mere 1uV... I wonder if the QA400 can really measure that accurately.

The headphone amp has a class-A output stage and produces neglible distortion (<0.00001% @20kHz) in LTSpice.
I was hoping to verify its performance... maybe a bit naive to think that a $200 product would allow such
precise measurements :)
On to the next project then - a deep notch filter with a x100 gain amplifier to bring those distortion
product up to the QA400's usable range.
 

Attachments

  • QA400BS.JPG
    QA400BS.JPG
    111 KB · Views: 380
Just a couple of THD measurements I did with a QA400 connected to a headphone amp, they indicate quite clearly that the QA400 cannot realiably measure anything below -100dB as has been noted earlier in this thread.

The upper table shows the QA400 connected to a blameless-style headphone amplifier. The load is a divider made from a 560ohms resistor and a 39 ohms resistor. Output level on the QA400
was -7dBV; the amplifier has a gain of 16 and puts roughly 7,15Vrms into 600 ohms; the divider attenuates the signal to approx. -7dBV again.
Second table shows the QA400 in loopback-mode.

The third table shows the difference of each harmonic calculated as "loopback minus amplifier".
Positive numbers thus indicate that the harmonic coming out of the amplifier was actually *lower* than in loopback mode whereas negative
numbers show distortion that is presumedly produced by the amplifier. The rightmost column is just the sum over all substracted harmonics (note that @3kHz the amplifier output seems to have significantly lower THD
than the DAC that feeds it - now how's that possible?)

Amplifier distortion and internal distortion in the ADC/DAC seem to partially add or cancel. -120dBV would be a mere 1uV... I wonder if the QA400 can really measure that accurately.

The headphone amp has a class-A output stage and produces neglible distortion (<0.00001% @20kHz) in LTSpice.
I was hoping to verify its performance... maybe a bit naive to think that a $200 product would allow such
precise measurements :)
On to the next project then - a deep notch filter with a x100 gain amplifier to bring those distortion
product up to the QA400's usable range.


The problem with distortion cancellation which you sighted can just as easily happen with a $25k analyzer as with the QA400. The question is can any analyzer measure accurately at its residual. The answer is no. You can improve the accuracy of the QA400 by suppressing the fundamental by either the null method or a notch filter. The problem is not so much with the digital as it is with analog input circuitry.
 
I'd like to observe that David made salient points in both his posts. As to speakers, Paul Klipsch and I did good distortion and response measurements on speakers using an HP 3580A swept spectrum analyzer. Of course this was in Arkansas where we could work outdoors for much of the year, yielding a quasi-anechoic environment, with a solid ground plane. The QA400 will certainly do as well, so I wouldn't be so quick to write off good speaker measurements given the ability to work outdoors in a large open space.

Knowing the equipment limits is crucial to good results and then they can be worked with instead of being worked against. If you have to work in a small space then some form of TDS is needed for good results at low frequencies, with the limit set by room dimensions. Nevertheless, I believe the QA400 can do a good job at higher frequencies in ordinary rooms.
 
Just did another test with the headphone amp replaced by a LME49860 "ultralow-distortion" opamp into the same 600ohms load. Same result: there is serious rise in harmonics at 1 kHz whereas THD at 2 kHz and 3 kHz is actually lower than in loopback-mode. At least this confirms the distortion is not from my amplifier.
 
Only just found out about the existence of the QA400 from a link placed by Tom.

I was going to ask if it's any good.

The answers are all here, especially the fact that so many of our renowned Members have bought the product.

I think I'll do a bit more reading and probably end up buying one.

I think I'll also buy their QA190 differential probe. Unless someone says there is a better value product elsewhere.
 
Last edited:
Just wanted to say I recently found this thread and the QA400 looks like a great deal for experimenting with audio. Did anyone ever build an auto ranging attenuator?
I built a switchable attenuator.
2 @ -20dB and progressively lower steps down to -0.05dB.
12 "T" attenuator switches (double pole double throw) and each is 600ohms.
This allows me to "dial in" from -0dB to -61.05dB in steps of 0.05dB.
Great for testing the gain of an amplifier. and finding the -0.1dB roll off frequency, or any other, -1dB, -3dB etc.
 
I built a switchable attenuator.
2 @ -20dB and progressively lower steps down to -0.05dB.
12 "T" attenuator switches (double pole double throw) and each is 600ohms.
This allows me to "dial in" from -0dB to -61.05dB in steps of 0.05dB.
Great for testing the gain of an amplifier. and finding the -0.1dB roll off frequency, or any other, -1dB, -3dB etc.

Thanks for the info. Do you mean something like this? Logarithmic Attenuator Calculator

I was checking out old HP distortion meters but this looks like a better investment.
 
I built a switchable attenuator.
2 @ -20dB and progressively lower steps down to -0.05dB.
12 "T" attenuator switches (double pole double throw) and each is 600ohms.
This allows me to "dial in" from -0dB to -61.05dB in steps of 0.05dB.
Great for testing the gain of an amplifier. and finding the -0.1dB roll off frequency, or any other, -1dB, -3dB etc.
Sounds very useful.
What input and output resistances ?.
Can you give a schematic please ?.

Dan.