The biggest jump was from version 5 to 6. They completely redid the UI - for the much better IMO. It has a few quirks, but it has definitely become a mature product.
Interesting. I ran the QA403 from a battery for a bit. That made no difference as far as I could tell.
Tom
I meant I only run the PC (laptop) in battery mode i.e. I unplug the adaptor. The 401 runs ok via the PC USB port. I also found it important to ensure the amplifier chassis and the 401 and the load were all bonded together at 0V. This allowed me to push 50 Hz noise peak below the noise floor (see the ax-Amp plots in audioXpress). Without doing that, the 50Hz noise spike pops out of the noise floor. That said, most folks have their own set-up config and interconnect methodology so YMMV.
Impressive performance. Do they have a box that has both ADC and DAC where the clock is shared between the two? It's advantageous to have them on the same clock as the FFT is sensitive to clock wander.
Tom
The ADC has a firmware option to put out an SPDIF signal on a pin, which then can be used to sync a DAC like ADI-2 Pro.
Not ideal, but better than nothing for now.
I would think sooner or later @IVX, will offer an integrated DAC+ADC device running from the same internal master clocks, as this is really important for a lot of measurements.
Well,
You mean coherent measurements. Even here the the master clock changes within the sample time and FFT & FFT Window may hide here the issue too.
So it will show the inherited performance of the two (ADC & DAC).
RME showed on an early doc, as on real, where NONE coherent measurements where used as two independent HW as one ADC the other DAC.
Bee just surprised how crappy clocks will perform.... sometimes, for reference purpose, a low performing is used 😀
Attached a later measurements to beet as using costume made ADC AK5572 & DAC AK4493 using coherent MCK clock 😀
Attachments
How about building yourself an ADC with similar performance and possibility for syncing DAC.
https://www.diyaudio.com/community/threads/diy-adcs.419922/
Yes, I'm following this interesting thread already.
Such an ADC+DAC combo is already in the works (in my day job), albeit for RaspberryPi environment because that's going to be the host platform.
Personally I'm also looking at direct USB implementation (rather than making an rPi look like a USB device) but haven't put my nose into the controller part. A STM uC is certainly an option and likely a better one than the usual XMOS but I have zero experience with STM controllers, only other brands. USB drivers (on the controller side) are also not my field of expertise and one would need a dual purpose driver, one for the audio transfers and a second one for control interface (MIDI SysEx control would be my preferred choice)... lot of open questions...
This may be an easy low overhead solution for the USB to ADC/DAC interface. https://www.diyaudio.com/community/...tichannel-input-output-interface-york.413504/
For the Cosmos I suspect one could find the master cloc input to the ADC and kludge a bugffer/trasformer output that could be used on a DAC or vice versa. With USB its more complex but its done in the pro world.
For the Cosmos I suspect one could find the master cloc input to the ADC and kludge a bugffer/trasformer output that could be used on a DAC or vice versa. With USB its more complex but its done in the pro world.
Myself being completely irrelevant to USB to I2S conversion, I was looking for a ready available solution and my understanding is that besides synchronous clocking, low latency is a requirement depended on the microcontroller chip.
My 2c on the Cosmos stuff. I bought three of his items in the process of trying to upgrade my test lab and while they do indeed perform astronomically well they come with the imo standard budget issues. For the ADC it's input impedance (too low), maximum voltage (also too low), UI (opaque), and pinouts (really? custom mini-trrs cables? you ever try to build one of those?).
To be more precise, I then had to add on preamps and scalers and still watch things like a hawk.
Real life: I bought the Auto-Scaler and then ... To begin with who makes a piece of hardware with a single important dial (gain) that's unmarked such that you have to count the clicks? Anyway, I did a sine test then a multitone test and duh forget that the peak voltage goes up. The scaler literally lasted a day as it self-perma-destructed on an overvoltage. I found myself spending so much time only testing up to X or dealing with low impedances or ... the QuantAsylum is like a breath of fresh air.
So, I'm a real fan of the specs but not so much of the real life use model.
To be more precise, I then had to add on preamps and scalers and still watch things like a hawk.
Real life: I bought the Auto-Scaler and then ... To begin with who makes a piece of hardware with a single important dial (gain) that's unmarked such that you have to count the clicks? Anyway, I did a sine test then a multitone test and duh forget that the peak voltage goes up. The scaler literally lasted a day as it self-perma-destructed on an overvoltage. I found myself spending so much time only testing up to X or dealing with low impedances or ... the QuantAsylum is like a breath of fresh air.
So, I'm a real fan of the specs but not so much of the real life use model.
That's generally how I feel about actual test equipment versus items cobbled together to mimic test equipment.
Tom
Yeah. Especially since my last non-cobble device had a max 400V auto-scaled-input. Well, lesson learned. I spent, in dribs and drabs, ~$1000 overall and then gave up and bought the QA403.
Hey @tomchr
You might take a look at this program I'm working on. Compared to the stock app, well...
See: https://github.com/MZachmann/QA40xPlot
You might take a look at this program I'm working on. Compared to the stock app, well...
See: https://github.com/MZachmann/QA40xPlot
If I had a QA403 I'd take a look. I sent it back to the owner. Below are my comments based on the screen shot.
First impression: Much cleaner than the stock app. Nice work!
I would suggest a single sweep tab where you can select if you want to sweep frequency or amplitude. What's the difference between Freq Response and Bode Plot? As far as I'm aware, Frequency Response = Amplitude Response + Phase Response. A Bode plot is a piece-wise linear approximation for these.
If you prefer to keep it as-is, I suggest capitalizing THD as a minimum. FFT is capitalized as well. It would also be nice with Amp(litude) and Freq(uency) spelled out. I'd also spell out the window functions. Does it add that much to add 'ing' to Hanning? We're not in the 80s where every bit and every pixel counts.
I would rather have a dropdown menu for the attenuation. What does the current format add?
I'd argue it should be Amplitude rather than Voltage for the generator setting. Indicate the units either Frequency (Hz), Amplitude (V) or add the Hz and V after the input field. Why are those drop-down menus?
How about Y-axis Range (dB) instead of "dB range"? Why are the end points for the frequency axis drop-down menus? Shouldn't they be text/number entry fields?
Some may argue that I nitpick, but I think the precision of the instrument should be reflected in the precision of language used in the software.
I think it's a big step in the right direction, though.
Tom
First impression: Much cleaner than the stock app. Nice work!
I would suggest a single sweep tab where you can select if you want to sweep frequency or amplitude. What's the difference between Freq Response and Bode Plot? As far as I'm aware, Frequency Response = Amplitude Response + Phase Response. A Bode plot is a piece-wise linear approximation for these.
If you prefer to keep it as-is, I suggest capitalizing THD as a minimum. FFT is capitalized as well. It would also be nice with Amp(litude) and Freq(uency) spelled out. I'd also spell out the window functions. Does it add that much to add 'ing' to Hanning? We're not in the 80s where every bit and every pixel counts.
I would rather have a dropdown menu for the attenuation. What does the current format add?
I'd argue it should be Amplitude rather than Voltage for the generator setting. Indicate the units either Frequency (Hz), Amplitude (V) or add the Hz and V after the input field. Why are those drop-down menus?
How about Y-axis Range (dB) instead of "dB range"? Why are the end points for the frequency axis drop-down menus? Shouldn't they be text/number entry fields?
Some may argue that I nitpick, but I think the precision of the instrument should be reflected in the precision of language used in the software.
I think it's a big step in the right direction, though.
Tom