Hi @matt , can you please check my request below?
Thanks,
Alain
Hi @DeAI66, just to confirm, your goal is to measure THD and THDN? That would be recorded by the record head, delayed by 165 to 330 mS, and the captured by the playback head, is that correct?
In this case, I think the Edit->Settings->Latency Compensation is what you want. Start at 48k, and then increase pre-buffer to give you 0.51 seconds of system latency. What this will do is play and extra 0.51 seconds of tone at the front end of the burst, and then capture the tail end. Your 330 mS of delay will be handled here. This is primarily for testing bluetooth, but should work equally well here.
And once confirmed at 48k, you can bump up to 96k. You will again need to adjust the pre-buffer size to get back to >330 mS.
And in both cases, just as you’ve done above, you can verify the time domain plot is correct. In the MISC section of the settings dialog, there is a checkbox for Show Truncated Burst in Time Domain. When this is ticked, you will see only the burst that is used for the FFT. It should be gapless when ticked.
When the checkbox is unticked, you’ll see gaps in the front-end of the time-domain burst that are related to the delay you measured above. That is OK, since those gaps aren’t considered for the measurement.
If time permits, please share some waveforms. I think the interest in reel to reel continues to grow. We can work through getting a triggered chirp and frequency response measurement done next if you are interested.
PS. Sorry for the delay in responding. I missed this first time around.
Another question… I’ve been using it for FR plots for a while now. I just leave the expo chirp running and the display updates quick enough to allow me to tweak EQ pots or whatever in almost real time. Using a 32k FFT and the default (I believe) 2048 pre buffer. Works fine on 15 and 7.5 IPS. Distance between rec and repro heads is about 1.2" According to the post above it shouldn’t work with 7.5IPS if I did my math correctly.
Hi @ReelAudio, the expo chirp has a bit more room at the trailing edge. If you look at time domain (in loopback) of a 64K FFT expo chirp, note that the chirp is at the front of the burst. The chirp finishes around 350 mS, but the total burst is about 725 mS long. And so, a 64k burst can just barely cope with 725-350 = 400 mS of delay. The latency setting shown above only works for the sine generation. If the latency is more than 400 mS and you want to do a chirp, I think you’ll need to export a triggered chirp, play that onto tape, and then play that back and capture it via triggered.
Thanks Matt, that makes sense. It explains why the FR chirp even works with cassette decks at 1-7/8 IPS. It’s because the heads are only like .25" apart. I’ll never has a 400ms delay so I’m good.
Hi @matt , yes, my goal is what you described. Thanks for the instructions.
I will try those settings during the record bias/EQ adjustment of the following tape deck and report back with some waveforms.
The few QA403 THD results with tape decks so far, were lower than what my analogue distortion meter measures (Leader LDM-178).
This LDM-178 uses steep high pass filters with a cut-off point just above the fundamental frequency setting of 315Hz or 1kHz. So in fact not only THD is measured but also the noise above the fundamental frequency, so more like THD+N.
It’s no surprise that tape noise is significant within THD+N.
With the QA403 I could measure a THD of only 0,2% @ 315Hz and recording level 0VU (tape speed 7.5 ips and fluxivity 250nWb/m) on a good performing and well adjusted tape deck, while the factory spec’s mention 0,8% average, but most likely including the tape noise.
Kind regards,
Alain
1 Like
Hi,
I am new here on this forum and I should received my QA403 today. I bought it to complete my testing gears. I am an electronic technician and do R2R and cassette decks refurbishing, including of course the calibration of them.
I just found this very interesting thread and I am very open to see future new features on the software of the QA40x, but I have some opinion about not going to far. As for example, any tools to calibrate the azimuth of a tape gears. IMO I don’t think that the effort worth it. I bet that most people that own a QA40x gear probably already have a scope. The XY Timebase Mode of a scope does very well the job for calibrating the azimuth. You can use a generator or the QA40x generator to start a 1 kHz sine-wave on both channel and look for the phase of the scope Lissajous figure. That is fast, easy and very precise. And if you have the chance to measure both channel level on a millivoltmeter at the same time, you can read their level and try to achieve the maximum level on both while looking at the scope phase. With these, you will never be out of phase. Then you increase the frequency to 5 kHz, 10 kHz 12 kHz and 16 kHz. You will never fail!
I just got my 403. Can I plot the Freq response of a recorded sweep already on tape? Many decks I have a combined REC/PLAY head and can not monitor the actual recording in real time. When I play the tape into the 403, I can see the sweep going across the software but for the life of me, I can’t figure out how to plot the peak line with it. Currently I’ve been using a software plug-in Analyzer on my DAW (computer audio workstation) using my RME audio interface to bring in the recorded signal from the tape and plotting out on the software analyzer. It actually works pretty well and gives a good visual plot that I can screenshot, but I would like to be able to do this with the QA403. Seems like it should be easy enough no?
Thanks!
Just reading through this thread in bit more detail, Would a possible way would be to record the chirp to tape, play back the recorded “chirp” and capture it in the time window then check the freq display to see the response? Not home right now so I can’t quick check to see if that would work. Getting to know this QA403 baby is actually very educational!
I tried recording the FR ExpoChirp, in this case to track 1 of a Tascam 688 8 track cassette deck. Then did playback while FR is still running on the 403. Took a few tries but once I got a nice full capture in the time screen I looked at the Freq input screen and got this result. Curious from the experts here if this is a legit FR? Thank you all in advance for your tips and mentoring.
Hi @drfonta, take a look at Triggered Sweeps
The QA40x software will export a wav with a special trigger sequence at the beginning. You need to get that onto tape, but once it’s on tape, you can righ click on the Run button and select the Triggered Sweep tab and wait for the chirp to happen. You can adjust the trigger amplitude as needed.
And yes, what you’ve captured above looks legit. Note your time domain plot clearly shows the start if the chirp. You can see some amplitude variation over time, which means FR variation, and you can see the clear roll-off at the end of the chirp. Based on your freq plot, the rapid rolloff is probably beyond 20k. But you can shift to 96k sample rate and learn if the rolloff is the tape machine or the analyzer.
if you try the triggered sweep, please report back on how it goes!
Thanks for the info on the triggered sweep. Will try that in the next day or two when I can get back to it. I’ll post results here. Definitely like this method of testing FR. Still amazed at how good a response these Tascam multitrack decks were/are able to achieve on cassette!
That works! Exactly what I was looking for. The QA403 detects the trigger and then plots the result. Will post some screenshots later.
Here’s a couple of tape recorded triggered sweeps. I set the range up to 22Khz to see the roll off. Not bad!
Nice! This is 8 tracks on a standard cassette tape? I’ve used 2" 24-track Revox machines which are ~12 tracks per inch, and also Fostex 4 track on cassette machines. I didn’t know there was 8 tracks on a cassette, however. This is ~64 tracks per inch. Do you know if the low and high-end rolloff is due to track width? In other words, is 8 tracks in 1/8" tape about at the limits of what is useful?
Yes that is correct. How Tascam managed to get 8 tracks on a tiny cassette with fairly decent quality is beyond me. They also have a model 238 rack mount version that produces very similar results using dbx as well. Prior to digital these were great devices for budget recording for artists/bands.
Well you have 50dB per division, so anything will look good! Try to standardize on plots like these, 10 or 5dB/division is pretty universal.