Freq response chirp offset from analyzer measurement

1

Hello! I’m using a QA401 with Analyzer version 1.923. We are using the QA401 to automate effects pedal testing through Tractor - we have been using it for go/no-go tests using amplitude measurements for a while, but I want to add masks to the mix and am having trouble.

When using the FreqResponseChirp plugin, the result is offset by almost exactly +6dB from the Freq Response measurement done by the analyzer. The chirp amplitudes are both -30dB. I tried setting the external gain field to +6dB to nudge it back, but it also offsets the measurement that the QA401 is actively taking. Is this a bug, or is it user error on my part?

Thanks for your help!

2

Hi @john_eae

I just tried a simple test and it seemed to work as expected. Can you please try to replicate:

First, create a “simple.mask” file as follows:

# QuantAsylum Mask File
# Frequency, Minimum Pass Value (dB), Maximum Pass Value (dB)
100, -1.0, 1.0
20000, -1.0, 1.0

Then, pick File->Reset Startup Defaults to get the analyzer to a known state.

Next, to make sure the simple.mask file is correct go into the FR context menu and load it:

image
image390×560 13.4 KB

And then try it in loopback to make sure it displays correctly:

image
image800×480 46.8 KB

And then make a new project in Tractor and create a test as follows using that mask:

image
image786×762 18.2 KB

And then run it:

image
image800×480 53 KB

The 6 dB is a good clue. A few things that might be impacting your result: You have an input offset specified in the DBV menu, or the pre-analyzer input gain is non-zero in the Tractor settings, or you are making a differential measurement when single ended is expected (or vice versa).

Please try and let me know what you see.

1 Like
3

Hi Matt,

Thanks, that test did work. Based on this, I don’t think it’s the act of reading the mask file back into QuantAsylum (or Tractor) that is causing the issue. It appears that some setting for the chirp plugin is introducing the 6dB offset. I had a similar thought as you that maybe I had a gain error somewhere.

Here’s the analyzer screen:

image
image1494×965 28.9 KB

Here’s the same measurement from the plugin:

image
image820×456 8.53 KB

And just in case, here are my chirp settings:
image

I hope that helps explain it better.

4

OK, I think I figured it out. The attenuator setting persists through to the chirp test. Is there any way to control this when running the plugin?

5

This setting determines the input level for the Tractor test. It can be either 6 or 26.

image
image786×762 23.1 KB

6

I don’t think this problem was caused by my Tractor settings. I recorded traces for masks with the attenuator off, and they work perfectly whether I set the input level to 6dBV or 26dBV in Tractor. I think the attenuator setting is lost when generating the mask data. When I run the “AmpFreqResponseChirp.dll” plugin, it seems to compensate for the attenuator setting in a different manner from the Analyzer software. If I save traces at +6dBV, generate a mask in the graph tool, and import that back into Analyzer, it works. If I do this at +26dBV, the +6V offset appears.

7

Yes, OK, this has been reported previously I see now. We’ll make a plan to fix. Thanks