It is quite stupid but AP2700 has no THD test at all, THD+N only. The DSP section has some harmonic analyzer but it works maybe 1% to .01% range, not even for -110db levels. And actually, the internal ADC is ridiculously low-performance, so there is no reason to show harmonics <-110db 🤣
DAC output aka DGen is also awfully distorted but a lot faster vs AGen, and I see no reason to use AGen for the frequency response test.
The FFT is limited at 32k that's it, you can use AVG to make FFT look a bit better.
DAC output aka DGen is also awfully distorted but a lot faster vs AGen, and I see no reason to use AGen for the frequency response test.
The FFT is limited at 32k that's it, you can use AVG to make FFT look a bit better.
Last edited:
The FFT is processed in the (limited) hardware, APWIN just collects the data and displays it. That is true for basically everything the machine does, the software is just a remote control & display app.
That's why using external FFT on the "reading/monitor" output is an improvement.
RIAA measurement works with a table-based approach, setting frequencies and levels for the analog generator. You load the desired .adq file from the EQ presets folder.
I still don't understand why AP limited the FFT memory with 32k, I saw inside of analyzer pretty common DIM memory which was never expensive and easily expandable.
Well, one can tune the bandpass filter to track the N-th harmonic (up to the 5th) but that's a crude way for implementing narrowband selective voltage meter. Using external FFT is much easier, but you need some reference to get the levels right. I often use a marker frequency with known level for that, mixed into the signal to be analyzed. For a DAC, this is quite easy and straightforward.
I will try to connect also the sound card to reading monitor and use spectraplus.
I will try to see how table based approach works, thanks for the hint!
I've worked with the SYS-2700 at TI. The software is indeed awful. AP must have gotten a lot of flak for that as they rewrote the software completely with the APx500 software. Unfortunately, the new software doesn't work with the old hardware. It does work with ASIO compatible sound cards, though.
Tom
Thanks TOM, yes the SW has a quite awful interface, back then, this interface was innovative so now it has been improved.
It's completely wront their new SW works with Soundcards/ASIO and NOT with their System Two family.
Could you tell me what you didn't like about the SW specifically? This will give me an idea about the limitations. Also not sure what the difference is between APWIN and AP2700 SW, for me to install and work with that, I need to buy the AP USB adapter as the interface Duke has only works with APWIN.
It's completely wront their new SW works with Soundcards/ASIO and NOT with their System Two family.
Could you tell me what you didn't like about the SW specifically? This will give me an idea about the limitations. Also not sure what the difference is between APWIN and AP2700 SW, for me to install and work with that, I need to buy the AP USB adapter as the interface Duke has only works with APWIN.
Whew! I thought it was just me! I've had my 2722 a few weeks and have wondered why it was so convoluted to use the software.
Wait - the new software works with ASIO-comp. sound cards? Interesting.
This is the FFT of 1KHz 1Vrms
with the following setting (unfortunately I have to do a snip of the screen to capture the image, with APWIN I cannot see export to JPEG or anything like that.
Are these setting correct? Should I change something to obtain better performance?
Also, why on the pre-made tests I can see the generator has more options in terms of impedance and the analog analyzer selectable input impedance? How can I enable all those features?
Also, do I understand correctly that the max FFT frequency allowable is 32KHz? The machine is 96K so shouldn't it go up to 48KHz max BW?
One note, I have only loop cables between Analog in analyzer to Analog out gen, do I have to add monitor to the digital section ot some other conenction?
with the following setting (unfortunately I have to do a snip of the screen to capture the image, with APWIN I cannot see export to JPEG or anything like that.
Are these setting correct? Should I change something to obtain better performance?
Also, why on the pre-made tests I can see the generator has more options in terms of impedance and the analog analyzer selectable input impedance? How can I enable all those features?
Also, do I understand correctly that the max FFT frequency allowable is 32KHz? The machine is 96K so shouldn't it go up to 48KHz max BW?
One note, I have only loop cables between Analog in analyzer to Analog out gen, do I have to add monitor to the digital section ot some other conenction?
Last edited:
Definitely not you. I tried for over an hour to get a THD+N vs output power sweep going. It wasn't until an AP rep or someone more experienced with the software showed me that it needed to be set up as a nested sweep that I was able to get it going. Maybe I didn't try hard enough, but it was certainly anything but intuitive. The hardware is very capable, though. The 2700 is a smidge better than the APx525 I used to have.
In the Bench Mode of the APx500 software you select the sweep tab and then select the generator amplitude as the swept variable. You then get THD+N vs Measured Amplitude as an option to graph and can change the unit to W (into the specified reference impedance).
The APx500 software is a true joy to use, especially considering how incredibly complex it is. And that's not praise I dole out willy-nilly.
It does and has for a few (5?) years now. The licence is around $3k AFAIR. You then get the same functionality as with the APx500-series, obviously within the limitations of the hardware. If I understand correctly, that also includes the smart speaker test functions and such they've added recently.
My only gripe about the APx500 software is that it relies on Microsoft .NET. I use version 7.1, which broke when the latest .NET was rolled out. So I have to run it on a PC that isn't fully updated (and, hence, not connected to the internet). I have to fork over $3k USD for an upgrade, even though I need none of the fancy new features. That's pretty disappointing.
Tom
I thought the APx was the latest and greatest how can the 2700 be better? and I wonder if there is any significant difference in performance between the 2722 and the 2522.
Nice 3K for the full SW, nevertheless, it won't work with system two anyway. I think you can setup THD sweep with fixed amplitude sweeping frequency, or fixed frequency sweeping the amplitude fairly "easy" even on APWIN, but sweeping power into a given load, I don't see that option in here and I don't even know what nested sweeps are yet. I agree this is not user friendly.
Per your point if you spend $3K on extra SW and you need to continue working on it, then it becomes a bit disappointing.
Nice 3K for the full SW, nevertheless, it won't work with system two anyway. I think you can setup THD sweep with fixed amplitude sweeping frequency, or fixed frequency sweeping the amplitude fairly "easy" even on APWIN, but sweeping power into a given load, I don't see that option in here and I don't even know what nested sweeps are yet. I agree this is not user friendly.
Per your point if you spend $3K on extra SW and you need to continue working on it, then it becomes a bit disappointing.
Thank you for that information, Tom. Having to pay to use it knocks me out of the game - I'm not in the business so this would be even more silly than me buying the 2722 (though it also came with the additional AUX-0025 and the Serial Interface Unit).
I must say, now that I've figured out the software a bit better, it is a pleasure to use and very powerful.
I must say, now that I've figured out the software a bit better, it is a pleasure to use and very powerful.
Because it is... The SYS-27xx was the state of the art at one point. I'm thinking the APx525 (~$12k) was less expensive than the SYS-27xx. Then the APx555 came along and made the SYS-27xx obsolete. AP just discontinued support for the SYS2700-series.
I received this from AP when I was talking to them about these things.
You can compare the performance of you 2522 with the plot above and find out. I suspect the 2722 is quite a bit better, but that's my guess.
I don't know if this generalizes to other AP models, but the performance of the APx525 is limited by the THD of the generator. So by filtering out the HD one can get down around -125 dBc THD (not THD+N!). I was able to measure near -140 dBc THD by then also notching out the fundamental from the DUT output. Unfortunately, I could not get it to work reliably across a wider range of voltages, so I ended up punting and getting the APx555 instead.
Tom
I thought they only had the APx555. I cannot find any documentation regarding the performance of a 2722 vs. 2522. I am wondering if they were different like in this case, they would have published it, but I am not sure. 2722 was too expensive and I don't know what benefits it would have offered.
From what I see the 2722 is an excellent performer. I do insist the interface is terrible and now I know what you mean about being buggy. I had to restart the box and the software as I was running FFT and no matter what I did it did not change and was displaying an older scan I took. SMH!!!!