It's a neat piece of kit and I'm still very much getting the hang of using it and I've run into a few surprises. I thought I'd do a post to update everyone and thought it might be a good idea to move it out of the TEA thread so that doesn't get clogged up, plus it might make finding information easier for anyone else encountering some of the surprises I encountered.
The first session with the QA403:
I fired up the venerable Hewlett Packard 8903B and set up the daisy chain of BNC cables and tees to feed the HP 3457A digital multimeter and Tektronix TAS485 scope and did a quick equipment check before adding the QA403 to the mix. The 3457A is great because regardless of what the sine source is and what front panel indications it might have, the 3457A can be used to check both frequency and amplitude using only one instrument. The existing bench setup is fine except for the usual 8903B issue of nothing present on the low output. So far so good, so it's time to bring in the QA403.
The first test was to move over to one side of one of the QA403's inputs. The BNC to banana plug with the wire jumper on it is serving as a shorting plug on the other side of the differential input. I need to come up with something more elegant than that. So far so good although I'm still getting used to how to operate the software for the best results.
Things started to go sideways with the next test I did which was loop the QA403's output to input and it looks ok at first glance except it's off frequency by a bit over 15 Hz. I would've expected it to be a lot closer than this. I ran a bunch of tests at different frequencies and it was off every frequency value I tried.
It was a bit difficult getting a picture but notice the ghost trace on the scope? When I saw that, I thought maybe the QA403 generator was acting up so I reversed things and used the QA403 to drive the 8903B's analyzer input and that's when things went from going sideways to completely off the rails. Both the 8903B and the 3457A couldn't make sense of what was coming off the QuantAsylum. Whatever was causing the ghost trace was causing the other gear to be unable to get any sort of a stable read on frequency, amplitude was bouncing around and the Hewlett Packard 8903B's distortion measurements were off the charts high, in total disagreement with what the QA403 was measuring in its own software.
It was at this point that I noticed the QuantAsylumn's software was shading the USB voltage readout and it was lower than 5.0 volts so I thought the QA403 might've been bottoming out and acting up on low USB voltage the way the manual warned was possible even though Windows never threw any warnings about excess USB current consumption. I decided to get a self-powered USB hub and a couple of other odds and ends at Sayal Electronics on my way home from work the next night since it's near the train station. I got to the store with about half an hour to go before they closed and then ran into total nightmare gridlock on the way home which put me back home seriously late that it wiped out any thought of doing more with the QA403 until the weekend.
The second session with the QA403:
I took another read through the built in manual, the quick start guide on the QuantAsylum website, and some of the discussion on their forum. One thought I had was that there might be some strange grounding situation that could be affecting the QA403's signal generator in a way that doesn't bother the 8903B's so I got out the Agilent U1602B portable scope and started charging it up with the idea of going out of the QA403 into that since that's a floating scope vs. the Tektronix which is ground referenced. I also got the self powered hub ready and redid the USB connections to have the QA403 connected to that and then in turn to the laptop.
The self powered USB hub cleaned up the voltage sag and the QA403 now runs on frequency. The ghost trace signal anomaly that only affected other equipment problem continued though and it didn't matter what oscilloscope I used to view the QA403's waveform. I tested with the TAS 485, the TDS 640A below it, and the Agilent U1602A and the results were consistent.
I remembered reading in the manual document that enabling the "Idle" button made the QA403 output a continuous sine when it wasn't in the "Run" mode acquiring samples. This worked and the scope trace and the values given by the 8903B and the 3457A were all reasonable. The ghost trace and haywire values issue was reproducible going between "Run" and "Idle" and after a while of observing the software running and the phantom trace appearing, it looks like the signal generator waveform on the QA403 is gated by the acquisition system when it's being used to read values.
That would explain why the QA403's software is not affected by this but it seriously messes with outboard test equipment being used to examine the signal at the same time.
I double checked the manual and it does say that with the "Idle" button enabled, it outputs continuous sine when it isn't acquiring samples - but it does not explicitly say the test signal output is non-continuous when it is acquiring samples. I don't know if there is a setting somewhere that can be adjusted to change this so the signal generator output is continuous during acquisition. If that isn't possible, it means toggling between "Run" and "Idle" depending on whether you're examining a test signal using the QA403's input side or outboard test equipment respectively; using the QA403 and external test equipment concurrently might not be possible.
I'm going to have to go back over the QuantAsylum forums and see if anything has been discussed about this. If most customers use the QA403 by itself, they'd never experience this issue. It would only show up when integrating a QA403 into an existing test bench where several pieces of test equipment are in use simultaneously.
That's where I had to leave it for today. I'm happy with the progress I've made so far. Getting the off frequency problem resolved and figuring out why the other test equipment was going haywire during signal acquisition when using the QA403's built in signal generator as the test signal source were two major hurdles that are now out of the way of putting the QA403 into productive use.
