(I could use a real USB-C PD tester but those are real $$$)
more testing-
1) 3 USB 2 hubs. The QA401 enumerates correctly only on one (which aslo has a serial and parallel port on it)
2) using a good bench supply and raising the supply to 5.25V (max legit) at the USB-A end and still get the really lousy results with bad cable and good results with good cable.
Now to check for obvious screwups in the cable fab, like short between shield and ground. . .
more testing-
1) 3 USB 2 hubs. The QA401 enumerates correctly only on one (which aslo has a serial and parallel port on it)
2) using a good bench supply and raising the supply to 5.25V (max legit) at the USB-A end and still get the really lousy results with bad cable and good results with good cable.
Now to check for obvious screwups in the cable fab, like short between shield and ground. . .
Hi Demian,
I have been able to read lower current draw on the odd bad cable. Since these only cost me about $2.99 each, they have paid for themselves many times over. Having adapters would be great. I guess you can't run them backwards? I have never tried.
-Chris
For the most part I agree with you. I have a pair of them since they don't all fit in the same spaces.
I have been able to read lower current draw on the odd bad cable. Since these only cost me about $2.99 each, they have paid for themselves many times over. Having adapters would be great. I guess you can't run them backwards? I have never tried.
-Chris
Hi Demian,
This may be a dumb question and I may be showing my ignorance here, but what is DiAna and how is it being used with the QA401?
Cheers,
Bob
Bob I'm shocked you haven't seen this. It brought end to all my software projects. Simply can't do better. DiAna, a software Distortion Analyzer
Thanks to someone mentioning it here I resurrected my Focusrite and tried their latest driver which fixed the problems. To my surprise my 19 yr. old shareware version of Cooledit suddenly works full duplex at 24/96. I have not found an ASIO driver that was recognized by Cooledit since XP.
No, It's sort of funny in a way. Cooledit had ASIO support built in 20yr. ago and it was reawakened without any intervention needed at all.
Still my preferred app for simple looking at stereo waveforms as data, in fact I used it at work with a sound card modified to do DC rather than Labview and an NI DAS.
QA401 Release 1.726 posted yesterday on GitHub
This release continues to focus on streamlining workflow and making it as easy as possible to build some fairly sophisticated graphs directly the in app and minimize the time you need to spend in other apps importing/massaging/exporting data.
The basic idea is that you make a measurement sweep, generate a graph, make a tweak, make another sweep and then push the output of that sweep to the previously created graph. And then the graph tool let's you interactively pretty everything up (min/max on axis values, trace widths, titles, notes, colors, etc)
The "THD FFT Size" plot below is a look at THD versus FFT size using this technique. Sweep, change FFT size, sweep, change FFT size, etc. At each step, keep pushing the output to the same graph.
This post HERE shows the process while making measurements on TI's TPA3255 2x300W class D amp.
This post HERE reviews the types of sweeps you can make in the 1.726 release.
If you need to generate project documentation from your measurements, then the easiest way to do so is to run all the sweeps you want, pretty up the graphs in the QA401 graphing tool, and then copy them all as same-size bitmaps and paste them into PowerPoint. Then arrange as desired, and then group them and re-size them all at once to fit the page. Below is a summary on a Class D amp based on TI's TDA7498 from a company called Fosi. The total time to make the measurements, build the graphs, and group them in PowerPoint and add the text, drop shadows, banner, etc, is around 20 minutes.
Comments on what seems confusing or broken and where to improve are always welcome
Thanks, Matt
This release continues to focus on streamlining workflow and making it as easy as possible to build some fairly sophisticated graphs directly the in app and minimize the time you need to spend in other apps importing/massaging/exporting data.
The basic idea is that you make a measurement sweep, generate a graph, make a tweak, make another sweep and then push the output of that sweep to the previously created graph. And then the graph tool let's you interactively pretty everything up (min/max on axis values, trace widths, titles, notes, colors, etc)
The "THD FFT Size" plot below is a look at THD versus FFT size using this technique. Sweep, change FFT size, sweep, change FFT size, etc. At each step, keep pushing the output to the same graph.
This post HERE shows the process while making measurements on TI's TPA3255 2x300W class D amp.
This post HERE reviews the types of sweeps you can make in the 1.726 release.
If you need to generate project documentation from your measurements, then the easiest way to do so is to run all the sweeps you want, pretty up the graphs in the QA401 graphing tool, and then copy them all as same-size bitmaps and paste them into PowerPoint. Then arrange as desired, and then group them and re-size them all at once to fit the page. Below is a summary on a Class D amp based on TI's TDA7498 from a company called Fosi. The total time to make the measurements, build the graphs, and group them in PowerPoint and add the text, drop shadows, banner, etc, is around 20 minutes.
Comments on what seems confusing or broken and where to improve are always welcome
Thanks, Matt
Attachments
Newbie question.... I've been away from this for such a long time, I've re-qualified.
I've built a resistor ladder 7 ohms/1ohm, with fukushima MPC74 1ohm/5 watt resistors. to aample across the 1 ohm resistor nearest speaker ground, but that means more adaptors, and an output level needs to be chosen so as not to overdrive the input.
It occurred to me that a BNC to RCA for input and a dual banana plug to BNC for the output, putting the 8 ohm resistor across the amp outputs would do, but with a different input signal level.
If the device under test has a gain of 22 dB ( it is an F5T ), then, should the output of the QA400 output a -32dB signal to get a -10dB signal at the input? Is that how that works?
Thanks in advance.
I've built a resistor ladder 7 ohms/1ohm, with fukushima MPC74 1ohm/5 watt resistors. to aample across the 1 ohm resistor nearest speaker ground, but that means more adaptors, and an output level needs to be chosen so as not to overdrive the input.
It occurred to me that a BNC to RCA for input and a dual banana plug to BNC for the output, putting the 8 ohm resistor across the amp outputs would do, but with a different input signal level.
If the device under test has a gain of 22 dB ( it is an F5T ), then, should the output of the QA400 output a -32dB signal to get a -10dB signal at the input? Is that how that works?
Thanks in advance.