Release 1.180

Release 1.180 (June 2023)

• Firmware versions 58 (unchanged)
• UI updates after every HTTP REST call.
• Lots of internal structuring for REST calls (added sync and async methods for all QA40x calls).
• Changes to A weighting, based on Wiki response curves. Previously A weighting had used a similar transfer function
• Added C weighting, based on Wiki response curves
• Added two user weightings (User1 and User2) which can be applied concurrently. See updated REST documentation for changes there too.
• Change in application close logic to avert USB hangs
• Added "Idle Gen" to the status bar when the idle generation is active
• Added AMP Crosstalk plugin. Docs are HERE. This should permit a sweep of an amp to generate crosstalk plots.

Release 1.178

Release 1.178 (June 2023)

• Fixed bug related to missed data at 192Ksps and larger FFT sizes. The solution here was giving the OS more overlapped buffers. The prior releases had 2 32K buffers given to the OS for both transmit and receive packets. This release increases that to 5 buffers.
• Cleaned up Wow and Flutter plug-in to better work at lower frequencies. It now works down to 400 Hz.
• Restored operation of cursor buttons in Oscilloscope Visualizer, added more cursors options, right channel display and the ability to trigger on zero crossing. This will allow a non-shifting display on async signals.
• Released plugin for MISC RLC Automated Test. Wiki for the page is located HERE
• Numeric keypad (small keypad on right side of keyboard) should now work for direct entry of amplitude.
• MISC Path Delay plug-in correctly reports 0 or 1 sample delay. Sometimes it would wrap around and report a 0-sample delay into a max delay

Release 1.176

Release 1.176 (May 2023)

• Fixed some display bugs where the RTHZ tracking axis that was added in 1.175 wasn't always showing up.
• Add a new feature to dramatically reduce DC offsets in acquired ADC data. The issue at hand here is that ADC offsets can normally be 2mV or so. The ADC inputs are AC coupled. However, the source of the offsets occurs after the input capacitors--the input buffers, the gain stages and the ADC. This doesn't really impact AC measurements, but the offsets are bothersome when looking at very small signals in the time domain because the signals aren't exactly where you think they will be.

The way the feature works is as follows:

1. Bring the analyzer up to operating temperature. This will probably take 15-20 minutes or so. The offsets are dependent on the operating temperature.
2. Short the 4 inputs (L+ and L-, and R+ and R-). These can be direct shorts to ground, or 50 or 75 ohms.
3. With the analyzer stopped, go to File->Generate ADC DC Offset File. This will result in the prompt:

4. Click yes. Measurements will be made for each full scale input setting, as the voltage offset present is different for each. This process will take several minutes.

5. When completed, a new file will be written in the MyDocs/QuantAsylum/QA40x/CalibrationData folder. This file will have the name '1234.dcOffsets'. But instead of '1234' it will be the serial number of your Analyzer.

6. Inside the file you will see the measured offset for each input range:

7. After the measurements are made and the file is written, you'll be asked if you'd like to apply the offsets to measurements.

8. You can also elect to apply the corrections in the Edit->Settings dialog.

9. Check the time domain to see if the compensation is working as expected. Here's a Analyzer without DC offsets applied. In the plot below, we can see the left channel (blue) has about 2.8mV of offset, and the right channel (red) has bout 1.2mV off offset. This is 0 dBV full scale input.

10. Now, we go into the settings, tick the "Apply DC Offset to Acquisitions" and run the time-domain trace again. This time we see the "DC Offsets" annunciator in the upper left. We can also see the offsets are now about 120uV for the left, and 40uV for the right.

11. Note that since the offset file is different for each Analyzer you might connect the PC, you can create an offset file for each QA40x analyzer, and the correct offset one will be applied when plug the Analyzer in.

Release 1.175

Release 1.175 (May 2023)

• Firmware Version 58 (no change)
• Added direct entry for amplitude for Gen1 to see how it feels on touch screens. Shortcut is CTRL+1
• Added ability to adjust Gen1 level by +1 dB (CTRL+Up arrow) or -1 dB (CTRL+Dn arrow)
• Measurement tiles now preserve user ordering on save/load
• TIME button will now scale axis based on full scale input
• Default import/export directory is now /TraceData off the /QA40x dir in MyDocs
• Export Frequency Series no longer exports the bin count as the first line of data. However, that info is still in the comment section
• Removed DAC Underflow message as this is too pessimistic. More changes will be coming here to help with 192Ksps on lower-spec machines
• Fixed bug where Nyquist wasn't accepted for THD(N) Measurement Stop settings.
• Fixed bug where THD measurements weren't being shown. This was related to Nyquist-referenced Measurement Stop Frequency and FFT size
• Fixed bug where displayed noise spectrum wasn't correctly taking windowing into account. This resulted into graph of noise density appearing 1-2 dB off when non-rectangular windowing was selected.
• RtHz display now activates a right-hand linear scale that tracks dB scale on left.

The Direct Entry Dialog
In production setups, there often isn't a mouse--maybe just a keyboard in a drawer with a trackpad. And engineers can be called to a production test bay to try and figure out what is happening with a particular test. And often this means some quick sanity checks. By hitting "CTRL+1" you can bring up a direct entry touch screen and quickly enter a level. That screen appears below:

Given the above, to enter "-10 dBV" you'd touch the 1 key, the 0 key, the NEG key (to make it negative), and then the dBV button. We can let this percolate for a bit to see how it feels and then add a similar dialog for Gen2 amplitude and also for frequency entry. This should make amplitude and frequency entry very quick and easy on a touch-screen device.

GEN1 Amplitude Adjustment
The Gen1 amplitude (tone only) can be quickly adjusted using the "CTRL+Up Arrow" (+1 dB) and "CTRL+Down Arrow" (-1 dB).

RtHz Linear Scale
When Rt Hz is activated, a linear scale will appear on the right axis of the primary graph. This will track the dB scale on the left side of the graph. From here, you can more readily read the linear value of the Rt Hz measurement. For example, in the plot below, we can see the noise density displayed is flat and around 2uV/RtHz across the spectrum.

Release 1.173

Release 1.173 (March 2023)

• More QA404 code merged
• Allowed PWR Output Impedance max load to be as high as 1Mohm
• Fixed bug where PAUSE (See Settings dialog) between acquisitions wasn't being applied in all the plugins. The PAUSE is now applied to all plug-ins. A change was made so that PAUSE was also applied to REST acquisition accesses.

Release 1.171

Release 1.171 (Jan 2023)

• Firmware version 58 for both QA402 and QA403
• More QA404 code added
• Fixed regression where model wasn't being saved on close
• More cleanup on Filter Visualizer. This still has a way to go as the graph display is too busy.
• Removed the Bartlett window and replaced with Blackman-Harris
• Name of settings file is shown on app title bar
• Added N+D measurement (Noise and Distortion). Normally, noise does include distortion components. The existing N-D measurement captures the RMS energy of the noise minus the distortion productions (aka known as "just the noise"). The added N+D measurement captures the RMS energy of the noise AND distortion components.
• Added ability to export captured data as WAV (File-Export->Export as WAV). This can be useful if you want to export a captured waveform and then load it into different analysis software for comparison.
• Fixed bug where user weighting wasn't loading upon startup even though it would be indicated on the display.
• Added PC Mirroring. See Edit-> Settings to enable. The "Freq/Amp Pause (mS)" setting allows you to set the amount of time to pause after the QA40x app changes the frequency and/or amplitude of Gen1. The mirroring only works for Gen1 settings, and if Gen1 is enabled. So, if you enable mirroring in the settings file, set Gen1 to -10 dBV and 1 kHz, and then enable Gen1, you should hear a 1 kHz tone coming from your primary PC sound card at -10 dBFS. Amplitudes in excess of 0 dBV are clipped to 0 dBFS when mirroring on the soundcard. With the addition of Mirroring, there are now 3 sources that can introduce a pause between acquisitions. First is the settling of relay. This is hard-coded based on the hardware. The second is the user setting (Edit->Settings) "Pause Between Acquisitions" (aka User Pause). This setting can be useful if you are testing a high-powered amp with a thermally limited load and you wanted to limit the duty cycle. And finally, the the Mirror pause. The Mirror pause delay will most often manifest in the automated test. So, if the DAC isn't settled to the new freq and/or amplitude before the measurement starts, you can bump up the Freq/Amp pause for the mirroring. The pause that occurs before the new acq starts after a freq/amp change will be the longest of 1) hardware relay changes, 2) User Pause and 3) Mirroring.

Release 1.170

Release 1.170 (November 2022

• Linux/Mono: Fixed exception from clicking in graph region. This was a regression from 1.167 related to build
• Linux/Mono: App Close should now cleanly shut down app in nominal cases without exception or hang
• Linux/Mono: Worked on out-of-memory issue. This was related to a library used by Mono. See wiki for more information.

Release 1.168

Release 1.168 (November 2022)

• Firmware version 58 for both QA402 and QA403
• Fixed crash when clipboard was busy on graph context menu copy.
• Got rid of splash screen under mono as it was obscuring important startup information that needed user confirmation (appeared app had hung)
• Added logging option (-l) to -headless
• Linux/Mono: Explicitly closed USB to help with shutdown. Without this, the next run will not see the device.

Release 1.167

Release 1.167 (November 2022)

1.667

• Firmware version 58 for both QA402 and QA403
• Fixed REST doc bug related to ATTEN levels
• Explicitly closed web server on app close.
• Added -headless mode for command line. This will display just the console, which might be useful for factory testing
• Fixed bug where Linux version was generating an exception when the active graph was clicked
• Certain functions that rely on Win32 native calls were disabled in Linux. You'll see a message when this happens. This is things like "copy clipboard as EMF." This is because these Win32 calls aren't available under Mono.
• Fixed bug where not all logging info was flowing to Linux console
• Re-worked Filter Explorers in Visualizer. Now Bandpass and Lowpass are supported. More work is coming here.
• change USB buffer sizes to try and help Linux data underflow issues
• Switched to source version of Zedgraph
• Added ability to specify desired hardware model on the command line (eg -QA402). This is useful for headless mode.
• Added more support for QA404 (not yet enabled)

Filter Explorer

The Filter Explorer remains experimental. It's accessible from the Visualizers menu. The aim is to make it easy to determine key performance metrics of a filter, including passband ripple, 3 dB points, Q and roll off rates.

In the plot below you can see the response of a low-pass filter with a notch at roughly 1 kHz.

We want to know the center and Q of the notch. First we make sure the Bandpass filter type is selected. Then we press the "Mark bandpass peak" button, followed by "mark -3 dB points" button. Then we drag to zoom and we can can see the center frequency is 1.019 kHz, with a Q of 4.6.

Next, we want to learn about the low-pass aspects of the filter. So, we switch to the Low Pass filter type. Then we manually drag the markers to specify the lower and upper edges of the passband. The press the "mark -3dB point" button, and then drag the "Fc 20dB" marker to a region beyond the rolloff. This location needn't be precise. The visualizer will calculate the roll-off rate automatically. Then drag to zoom and we get the following plot:

From the above we see the digital filter has about 0.22 dB of passband ripple and a roll-off around 12 dB/octave, which approximates a second order filter.

Release 1.165

Release 1.165 (October 2022

• Firmware version 58 for both QA403 and QA402.
• This release fixes two bugs in 1.164. The first bug was that an exception was thrown if you tried to open a .Settings file without hardware being plugged in. The second bug was that a malformed settings file didn't give much clarity as to why it refused to open. Both have now been fixed.

With release 1.164, a change was made to the measurement LPwrWatts. That measurement was re-named to LPkPwrWatts, in order to better distinguish it from LRmsPwrWatts. But the problem here is that causes a break to previously saved settings files if the settings file was using the LPwrWatts measurement.

You will know if you encounter this error because you'll see an error as shown below:

Note in the above error that the LPwrWatts is flagged as being invalid. When this happens, the settings file you tried to load will have the current date/time appended as a suffix, and a new Settings file will be created.

If you need the old settings file, you can preserve it by locating the LPwrWatts tag inside the file and changing it to LPkPwrWatts as shown:

The file can then be saved with a normal .settings extension.