You mention SCPI commands. Would I need to write some complex code to talk to a GPIB interface or just add a string to send? Supporting a GPIB interface with simple command strings would enable support for things like the HP/Agilent/Keysight switch boxes that can be pretty cheap on eBay. (E.G. Keysight (Agilent) 3488A HPIB Switch / Control Unit Mainframe | eBay) Of course the problem is which GPIB controller.
Thanks!
Can I make a couple of suggestions? Firstly, make the binaries available directly from your downloads page, it will help. The "releases" link is not obvious. I do commend you on making the source available though.
And finally, tractor looks perfect! All it needs to make it universal in my opinion, is an "external command" operation that can be run as part of the test sequence, running some user defined external command to activate whatever external hardware needs activating. It shoudl be simple from your side, and give the users freedom to implement whatever it is they need.
One last question, does the db connection support a local MySQL database?
Can I make a couple of suggestions? Firstly, make the binaries available directly from your downloads page, it will help. The "releases" link is not obvious. I do commend you on making the source available though.
And finally, tractor looks perfect! All it needs to make it universal in my opinion, is an "external command" operation that can be run as part of the test sequence, running some user defined external command to activate whatever external hardware needs activating. It shoudl be simple from your side, and give the users freedom to implement whatever it is they need.
One last question, does the db connection support a local MySQL database?
Last edited:
My suggestion to Matt was to add a simple "external command" feature to Tractor, where you call a user defined external command eg "./sendscpi.exe 'SOURCE
OWER:START 0DBM'" then it is completely flexible and just relies on you having some small binary you can install locally to talk to your hardware, whatever it may be!
Last edited:
Yes, got it, makes sense rszemeti and that can go in for next release. MySql isn't supported currently, but it's on the list of wants for sure.
Demian, yes, to use scpi there would have to be code written if you needed to interact with the instrument. If you just wanted to change the state of the instrument, then I think the mechanism rszemeti proposes is good. That would also allow you to change the state of a home-grown instrument too.
Demian, yes, to use scpi there would have to be code written if you needed to interact with the instrument. If you just wanted to change the state of the instrument, then I think the mechanism rszemeti proposes is good. That would also allow you to change the state of a home-grown instrument too.
Tractor release 0.9 went on Github last week (including source). Today some more bugfixing went up in release 0.91.
Releases * QuantAsylum/Tractor * GitHub
Release 0.91 of Tractor and Release 1.730 of the QA401 application were used for the blog post below. This is a walk-through on how to automate testing of a small 5W guitar amp previously measured
Automated Amplifier Testing with Tractor
Towards the end of the post is a video of the automated test running. The test includes operator entry of serial number, operator setting knobs is the required test position, the audio tests, and then the test playing a test WAV file out of the QA401 so that the operator can verify all the pots work as expected.
This release should also let you save results to the cloud and query them later. For example, if you make a test to measure gain at 1 KHz, and then you run 100 products, you can then query "what is the mean and sigma on the products we've run so far" and get that data back from the cloud.
Also, release 0.91 of Tractor has a command called "ShellA00" that should meet rszemeti's needs. This allows you specify a process name, the arguments, a timeout and whether or not you want to run it minimized. So, if you have an external EXE that will change some relays, you can call it via this ShellA00 command. If the ShellA00 command doesn't complete in X seconds, the testing is aborted.
Docs are in process for Tractor, but not ready yet.
Releases * QuantAsylum/Tractor * GitHub
Release 0.91 of Tractor and Release 1.730 of the QA401 application were used for the blog post below. This is a walk-through on how to automate testing of a small 5W guitar amp previously measured
Automated Amplifier Testing with Tractor
Towards the end of the post is a video of the automated test running. The test includes operator entry of serial number, operator setting knobs is the required test position, the audio tests, and then the test playing a test WAV file out of the QA401 so that the operator can verify all the pots work as expected.
This release should also let you save results to the cloud and query them later. For example, if you make a test to measure gain at 1 KHz, and then you run 100 products, you can then query "what is the mean and sigma on the products we've run so far" and get that data back from the cloud.
Also, release 0.91 of Tractor has a command called "ShellA00" that should meet rszemeti's needs. This allows you specify a process name, the arguments, a timeout and whether or not you want to run it minimized. So, if you have an external EXE that will change some relays, you can call it via this ShellA00 command. If the ShellA00 command doesn't complete in X seconds, the testing is aborted.
Docs are in process for Tractor, but not ready yet.
Last edited:
I have an issue. My qa400 sees itself just fine in loopback. But when I plug into my amp I see nothing. There appears to be lots of noise with a scope referencing case and .25 VDC, but qa400 does not see the amp.
The coax input and output ohms out well. I don't know why....the measure only working in loopback mode. Any ideas? I'm fresh out.
The coax input and output ohms out well. I don't know why....the measure only working in loopback mode. Any ideas? I'm fresh out.
First, make sure you don't overload and damage the input on the QA400. Check the manual/directions for max input.
Then start a continuous tone on the QA 400 and see if its at the input of the amp with your scope. (A check for shorted or open connections/cables). A BNC T at the QA400 output should give tou the same signal that is going to the amp. Or better scope at the input connection inside the amp. Then check the output of the amp if its powered on looking for signal.
The loopback confirms the QA400 is working with cables that connect. I have spent time looking for missing signals that proved to be damaged or incorrect connectors on the test cables. it happens to everyone.
Then start a continuous tone on the QA 400 and see if its at the input of the amp with your scope. (A check for shorted or open connections/cables). A BNC T at the QA400 output should give tou the same signal that is going to the amp. Or better scope at the input connection inside the amp. Then check the output of the amp if its powered on looking for signal.
The loopback confirms the QA400 is working with cables that connect. I have spent time looking for missing signals that proved to be damaged or incorrect connectors on the test cables. it happens to everyone.
D'Oh! Of course I reversed in/out. Thanks.
Current measurement of Left channel of this F5T is THD+noise < 0.00126. -78 dbV, THD alone -88 dbV. The second harmonic is the largest spike.
Noise on the chart never exceeds -90 dB.
Input signal is -23dBV, peak is -0.15 dBV. Into an 8 ohm load -- so about 1 watt.
Woot!
Where is the snapshot data stored for the QA400? I'd like to save/rename and compare various trials.
Also, when overlaying, the overlay seems to be dropped about 20 dB?
Never mind about the overlay dropping.... that was a real measurment.
Sorry for the trouble.
Current measurement of Left channel of this F5T is THD+noise < 0.00126. -78 dbV, THD alone -88 dbV. The second harmonic is the largest spike.
Noise on the chart never exceeds -90 dB.
Input signal is -23dBV, peak is -0.15 dBV. Into an 8 ohm load -- so about 1 watt.
Woot!
Where is the snapshot data stored for the QA400? I'd like to save/rename and compare various trials.
Also, when overlaying, the overlay seems to be dropped about 20 dB?
Never mind about the overlay dropping.... that was a real measurment.
Sorry for the trouble.
Hello everybody,
long thread with lots of posts. Search function couldn´t really help, so sorry
folks if i might be getting on someones nerves.... but is there a possibility to
use the QA401 on Mac OS? And i don´t mean using windows with parallels
or the like. Or is the software only for Windows?
Greetings
ulf
long thread with lots of posts. Search function couldn´t really help, so sorry
folks if i might be getting on someones nerves.... but is there a possibility to
use the QA401 on Mac OS? And i don´t mean using windows with parallels
or the like. Or is the software only for Windows?
Greetings
ulf
Back to this testing. I ordered a USB make and Female connector over ebay from China. They arrive a week or so ago. This stuff takes time. I assembled a little breakout to see the voltage at the USB B connector.
Good cable 4.99.
Bad cable 4.75V
Really bad cable 4.5V
At 4.5V the device doesn't run. For the long test I could not get any of the 28AWG cable to work. The device dropped link the moment the stream started and the voltage dropped to 4.5V
The bad and really bad cables had USB if approval etc. If you can see a printwheel on the cable it will usually call out the gauge of the wire inside. 28 AWG + 28 AWG will not cut it but is very common. Look for 28 AWG (the data pair) and 24 AWG or larger for the power conductors.
Separate issue- When making these hirez FFT's (262K) I ran into the inablity to get the software to respond once the plot was started. The only way to stop it was to kill the program. This may be a side effect of a Windows update but be forewarned. 131K works fine.
The capture below is what you can get when everything is working with at least 4.7V at the QA401. Its a fully balanced loopback.
Attachments
Hey great detective work! The 28/24 AWG are amazingly hard to find. Finally located some...
28/24 AWG USB cables
and more on Amazon...
28/24 AWG USB cables 6 ft
28/24 AWG USB cables 3 ft
28/24 AWG USB cables 1.5 ft
Last edited:
Release 1.79 posted last week, which allows complex mask testing via chirp stimuli. The basic premise is:
1) Make 3 or more measurements of the production DUT
2) Generate a mask using statistical descriptions of the DUT (sigma, max/min, etc).
3) Export the mask
The mask can then be used from inside Tractor as part of an automated test. It takes about 1.5 seconds to run the chirp and verify the mask as pass/fail
Post on automating an EQ measurement is here.
Post on the release is here
Also, some measurements on a 1 kw Behringer commercial amp here. This amp is interesting because it's a discrete design with automatic compensation for load.
1) Make 3 or more measurements of the production DUT
2) Generate a mask using statistical descriptions of the DUT (sigma, max/min, etc).
3) Export the mask
The mask can then be used from inside Tractor as part of an automated test. It takes about 1.5 seconds to run the chirp and verify the mask as pass/fail
Post on automating an EQ measurement is here.
Post on the release is here
Also, some measurements on a 1 kw Behringer commercial amp here. This amp is interesting because it's a discrete design with automatic compensation for load.
With patience you can often get things to work that shouldn't. Attached is a loopback using Praxis software + VB HIFI cable + QA401 ASIO + QA401.
Praxis is really a windows XP program that happens to work on current Win 10, however it has no ASIO interface. The QA401 has the thrird party ASIO interface that works quite well. The VB cable thing VB-Audio Virtual Apps hooks them together and seems to work OK with all of this. End result is pretty great performance at 192/24.
Praxis is really a windows XP program that happens to work on current Win 10, however it has no ASIO interface. The QA401 has the thrird party ASIO interface that works quite well. The VB cable thing VB-Audio Virtual Apps hooks them together and seems to work OK with all of this. End result is pretty great performance at 192/24.
Attachments
Hello All, Just found you guys here in this thread. Thinking and playing with noise measurement. I did the RMS noise measurement setup as 1Audio articulately spells out in post number 2624. With the QA401 inputs shorted with a 50Ohm BNC cap, RMS noise reads -114.7dBv, no weighting filter applied. With a 30dB gain amplifier attached (think QA470 when it is released) with the input shorted with bare wire RMS noise reads -126.2dBV, a -11~ dBv improvement over the QA401 by its self. With 680Ohms across the input of the 30dB gain amplifier the RMS dBV is reduced to -123.3dBV, 2.9dBv more than with the input shorted with bare wire. As close as I could get to 3dB with the resistors in my bin. The EIN of the QA401 plus 30dB gain amplifier is right in there at 0.469uV on my bench. Thermal Noise Calculator
Thanks
DT
Thanks
DT
Attachments
Hello 1Audio,
Looks like Matt has been busy. Open your QA401 application and click on Ctrl + .dBV. The dBV Options box will open. There is now a rtHz check box. This may be the root/Hertz measurement you were looking for. Matt is now shipping his new QA470 with a 30dB gain low noise amplifier. This is the low noise amplifier that I used for the plot in my post above. In my hand I have a pmillett Opamp experimeters PCB. Ready to solder on the little PCB is a low noise OPA1612 Opamp. This little low noise amp will be a gain of 100 or 40dB's. This project will look a lot like a TI post amp. http://www.ti.com/lit/ug/tidu016/tidu016.pdf
Thanks
DT
Looks like Matt has been busy. Open your QA401 application and click on Ctrl + .dBV. The dBV Options box will open. There is now a rtHz check box. This may be the root/Hertz measurement you were looking for. Matt is now shipping his new QA470 with a 30dB gain low noise amplifier. This is the low noise amplifier that I used for the plot in my post above. In my hand I have a pmillett Opamp experimeters PCB. Ready to solder on the little PCB is a low noise OPA1612 Opamp. This little low noise amp will be a gain of 100 or 40dB's. This project will look a lot like a TI post amp. http://www.ti.com/lit/ug/tidu016/tidu016.pdf
Thanks
DT
Last edited: