It supports Win 7 and Win 10. We haven't specifically checked it with 8.1 but I suspect it will work. Last I saw, Windows 8.1 usage was below 7% and declining. IOW, there has not been a lot of incentive to investigate it.
We are not recording audio data to disc. We can save the measurement results to CSV format though.
-Jonathan
We are not recording audio data to disc. We can save the measurement results to CSV format though.
-Jonathan
This was a tradeoff made for size/cost/performance issues. There is no plan to increase the current hardware capabilities. As you get to higher ranges, things get bigger and thus more expensive. You want to stay in the linear range of components such as resistors and caps and at higher voltages, you have to use bigger parts or combinations of smaller parts to maintain the distortion characteristics. You also need to space things out more for insulation purposes. Our plan is to support the use of external attenuators within the software.
The ability to get the measured data stored to disk for offline processing would be a really useful feature.
Please consider adding it to your software.
Power amps also need Load Resistors.
Considering the range may be 16 to 2 Ohms, and 20 to 2,000+ Watts, and MUCH bigger than this box (and many experimenters already have a load-board) (and others never test power amps), it doesn't seem to be something that should be in the box.
If you can rig a load resistor, you can figure how to rig 10K+1K to divide-down and extend the range to much higher than any loudspeaker amp. You can fiddle the 1K to 1,111r to get exact 10:1, or if the software has a fudge-factor you can feed a 12VAC transformer, measure the actual voltage, and tell the software "this is 12.345V rms" and it will figure the correction. Store that in memory/file "CORR1", write CORR1 on the divider, done.
You need to be aware of the dissipation, temperature, and stray reactance. At few-K level, reactance errors may be "small" (for some definition of small). The dissipation of 150V in 10K is over 2 Watts, and all resistors have some tempco and life effects, so you want a low-tempco resistor(s) that is very over-size. Some of the 50 Watt wire-wound types are stable at 2 Watts and not expensive. More precision types are very available. I suppose DIYA and other folks will ponder and suggest divider designs; Averlabs might issue an app-note of suggestions.
Last edited:
PRR, thank you for bringing up the specific issues of high voltage divider circuits. Typical barrel XLR type of attenuators often use really small resistors (< 1/8 watt) and cannot handle such tasks. We had just discussed writing an app note on just this issue last week. There are just some practical issues to watch out for. It is not rocket surgery or brain science. Finding the time to get the note written is the real issue.
One simple approach to deal with the tempco issues as well as the voltage and current nonlinearities is to use parallel/series combinations of resistors instead of just a single resistor.
-Jonathan
One simple approach to deal with the tempco issues as well as the voltage and current nonlinearities is to use parallel/series combinations of resistors instead of just a single resistor.
-Jonathan
Any idea what this adapter could cost?
I am definitely interested, my 20+ year old Audio Precision System One is starting to have issues, which can be fixed, but at some point it will not be worth the trouble any more.
The DScope III price dropped to US$8K recently, so the Averlab at roughly the price of a used 20 year old Audio Precision sounds very attractive, provided the adapter is reasonably priced.
The Averlab & Dcsope hardware specs seem fairly comparable.
The DSCope software is extremely impressive, the US distributor did a web based demo for me a few weeks ago. How will your software stack up against it?
Lukas
Last edited:
Lukas, I'm not sure what adapter you are referring to. Perhaps you meant attenuator. We aren't planning to sell an external attenuator at this time. We will probably just write an app note on building your own. Jan Didden's autoranger may work too. I need to learn more about it.
The dScope software is fairly mature and in several ways an improvement over APwin. It's weakness tends to be the residual distortion which the last time I checked (albeit a long while ago) was among the highest of the commercially available audio analyzers. My other gripe with it was that there was only one graphical display. You could have 6 (I think that was the limit) unrelated traces on it but I prefer to have separate graphs for separate tests.
The dScope software is fairly mature and in several ways an improvement over APwin. It's weakness tends to be the residual distortion which the last time I checked (albeit a long while ago) was among the highest of the commercially available audio analyzers. My other gripe with it was that there was only one graphical display. You could have 6 (I think that was the limit) unrelated traces on it but I prefer to have separate graphs for separate tests.
Thanks for the comprehensive replies, Jonathan.
Yes, I was talking about the external attenuator. I haven't followed the thread on Jan's attenuator in detail, but it seems to be well thought out autoranging with over voltage protection. Now, if your box could interface with it somehow with software so that your GUI knows the amount of instantaneous attenuation, we'd have a winner.
If I remember correctly, the DSCope residual is similar to my AP Sys 1, in the 0.000x % range, good enough for me
I like the AP method that allows multiple traces on one graph, e.g. left/right, also successive sweeps for something like a variable filter/EQ at different settings.
Also the DSCope software bargraph equivalents cannot be set for log, which I prefer for manual control of generator level & frequency for instance in order to vary quickly over a big range. But that is software and they might well implement it soon, or one could probably write some custom software with their API to do it.
I think you are trying to position yourself at the lower end of the pro users and the higher end of the amateur/DIY users, which could be a tough place to be
Lukas
Yes, I was talking about the external attenuator. I haven't followed the thread on Jan's attenuator in detail, but it seems to be well thought out autoranging with over voltage protection. Now, if your box could interface with it somehow with software so that your GUI knows the amount of instantaneous attenuation, we'd have a winner.
If I remember correctly, the DSCope residual is similar to my AP Sys 1, in the 0.000x % range, good enough for me
I like the AP method that allows multiple traces on one graph, e.g. left/right, also successive sweeps for something like a variable filter/EQ at different settings.
Also the DSCope software bargraph equivalents cannot be set for log, which I prefer for manual control of generator level & frequency for instance in order to vary quickly over a big range. But that is software and they might well implement it soon, or one could probably write some custom software with their API to do it.
I think you are trying to position yourself at the lower end of the pro users and the higher end of the amateur/DIY users, which could be a tough place to be
Lukas
We have a new take on this that I hope many will enjoy. AP requires that you specify ahead of time that you want to append results. We temporarily save all your test results with notations (by default we copy the generator settings into the notes) and then you can go back and select which tests you want to compare in an ad-hoc fashion. You can also name and save these results in the project which makes it convenient if you are trying to compare with tests you made on previous days.
-Jonathan
Jan, I had always regarded that feature as a work-around because one couldn't open up two distinct plots at once. Since APwin was also limited to a single plot, this feature was seen as an improvement.
In general, I find graphs with a single X-axis easier to read. I don't mind multiple Y axes because that can tell a story but when two overlaid graphs are distinct in both X and Y, it just creates confusion. Therefore, my preference is two distinct plots.
The only other benefit I saw in a combined plot was reduced screen size on lower resolution monitors. These days high resolution monitors are pretty standard. For instance, my laptops came with 2K and 3K screens and 4K screens are available on laptops under $1000 USD. Is there another benefit that I am missing or am I just undervaluing the need to preserve screen real estate? There is nothing in the architecture that would prevent us from doing this if there was a strong need to do so.
In general, I find graphs with a single X-axis easier to read. I don't mind multiple Y axes because that can tell a story but when two overlaid graphs are distinct in both X and Y, it just creates confusion. Therefore, my preference is two distinct plots.
The only other benefit I saw in a combined plot was reduced screen size on lower resolution monitors. These days high resolution monitors are pretty standard. For instance, my laptops came with 2K and 3K screens and 4K screens are available on laptops under $1000 USD. Is there another benefit that I am missing or am I just undervaluing the need to preserve screen real estate? There is nothing in the architecture that would prevent us from doing this if there was a strong need to do so.
I need a detailed plan of coming features in the bundled SW to decide between this HW and the Jensalyser (RTX some numbers) to make up my mind.
If I choose the Jensalyser then I can choose between a number of different already existing SW's, but if I choose the Averlab (very nice spec's and HW features btw) then I'm buying it "with good faith" (is that the right wording?).
Get my point ???
If I choose the Jensalyser then I can choose between a number of different already existing SW's, but if I choose the Averlab (very nice spec's and HW features btw) then I'm buying it "with good faith" (is that the right wording?).
Get my point ???
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.