A pot would be hard for the balanced output. A pair of jumpers could work, but then I have to figure out to keep the level indication in sync.
The level indication is I believe important to see at a glance what goes into the sound card, even if the 'official' level is determined by the test software.
Demian we discussed whether you need a level indication on the autoranger at all, or only the attenuation setting. Any other views out there?
Jan
Personally, I prefer savory to sweet!
Sweet spot? What makes it so sweet?
Personally, I prefer savory to sweet!
If we are examining the output of a vinyl RIAA preamplifier there is not much purpose in looking at noise or distortion products 10 or 20 dB’s below the expected noise floor of the source.
If we are examining the output of a CDRom there is not much purpose in looking at noise or distortion products 10 or 20 dB’s below the CD program dither.
If we are interested in picking the fly poop out of the pepper, as in examining the distortion produced by a metal film resistor we will need more amplification and resolution. (Plus notching or nulling of the fundamental test frequency with external circuits.)
Jan,
I think that many users would prefer user adjustable output to select their own sweet spot for their particular application. Rather than a pot, perhaps a two or three step switch. ~1.0 volt out, ~0.5 volt out and ~0.25 volt
Added
In terms of “a level indication on the autoranger” and or db gain / attenuation indicator, please provide whatever is needed by the software to compute dBfs dBr and dBV values on the FFT’s.
I may be using external amplification, it would be nice if the autoranger could display level so this variable can be pasted into the software for calculation of dBfs dBr and dBV values.
Thanks DT
Sweet spot? What makes it so sweet?
Personally, I prefer savory to sweet!
If we are examining the output of a vinyl RIAA preamplifier there is not much purpose in looking at noise or distortion products 10 or 20 dB’s below the expected noise floor of the source.
If we are examining the output of a CDRom there is not much purpose in looking at noise or distortion products 10 or 20 dB’s below the CD program dither.
If we are interested in picking the fly poop out of the pepper, as in examining the distortion produced by a metal film resistor we will need more amplification and resolution. (Plus notching or nulling of the fundamental test frequency with external circuits.)
Jan,
I think that many users would prefer user adjustable output to select their own sweet spot for their particular application. Rather than a pot, perhaps a two or three step switch. ~1.0 volt out, ~0.5 volt out and ~0.25 volt
Added
In terms of “a level indication on the autoranger” and or db gain / attenuation indicator, please provide whatever is needed by the software to compute dBfs dBr and dBV values on the FFT’s.
I may be using external amplification, it would be nice if the autoranger could display level so this variable can be pasted into the software for calculation of dBfs dBr and dBV values.
Thanks DT
DT, the ' sweet spot' is related to the measurement soundcard which the autoranger is supposed to feed. If the soundcard performs best with, say, 0.5V input, you'd like the autoranger set up to provide a nominal 0.5V signal to the soundcard whatever the signal is from your amp or preamp going into the autoranger. So for best measurement results you'd want the autoranger to feed the soundcard at its sweet (oir savory) spot.
Your other point are interesting. You would anyway rely on your test software to get the actual signal levels, and having the actual attenuation/gain displayed on the autorager would help to scale the final result.
At this point the AR displays the attenuation/gain in dB as a dimensionless ratio. Isn't that enough? Suppose you feed it with 50VRMS from your super duper power amp, and it ranges to 0.5V for the soundcard input and shows 'atten: -40dB' that's fine then, no?
Jan
Your other point are interesting. You would anyway rely on your test software to get the actual signal levels, and having the actual attenuation/gain displayed on the autorager would help to scale the final result.
At this point the AR displays the attenuation/gain in dB as a dimensionless ratio. Isn't that enough? Suppose you feed it with 50VRMS from your super duper power amp, and it ranges to 0.5V for the soundcard input and shows 'atten: -40dB' that's fine then, no?
Jan
So we are talking RMS, not Peak2Peak or Average?
The thing is, the connected measuring device (sound card) will have a maximum P2P input voltage, so the question is: how does this correlate? e.g. P2P, AVG or/and RMS to each other is clear, but not in relation to sweet spot.
Please let me know if this makes no sense.
https://youtu.be/S1mwCs1SKIY
Now I'm confused, surely we are talking about the output of the DUT and the DUT-input of the auto ranger. Not the generator output (like of the AR) that is connected to the DUT input? When talking about these things we should recognize that we are talking about 3 inputs and 2 outputs (as I see it)
Input 1; DUT
Input 2: Auto ranger
Input 3: Measuring device
Output 1: Generator (e.g. signal source to go to the DUT input)
Output 2: Auto ranger (e.g. auto ranged signal to go to the measuring device input)
And one more confusion source I spot (by me), the abbreviation AR was (by me) read as 'Audio Research', at some moment in time I have to wakeup
And now, finally (3rd edit of this message), I'm there, the original sentence 'You are talking about optimizing the AR output voltage that goes to the sound card.' is now clear to me, how difficult can it be, he (DualTriode) is talking about the AR (auto ranger) connection to the sound card (measuring device).
P.s. you may ignore this message, or you can laugh about it, but this is what happened to me, today
Last edited:
P.s. you may ignore this message, or you can laugh about it, but this is what happened to me, today
I never pass up the chance for a good laugh
But indeed, all I am concerned about is how to size the signal from the ar (no capitals) to the soundcard input. The input to the ar is irrelevant as long as it is below 120V or so.
And yes, all is RMS; I know a soundcard input level max is often specced in pk-pk (as this is related to its power supply voltage being often unipolar +5V).
But from Demian's work we know that the sweet spot for a soundcard, where it has the least distortion and not yet a lot of noise, may be at 250 to 500mV RMS.
So the max input is for reliability and the sweet spot is for max performance and that is what I am concerned about.
I know the ar is quite transparent, but to keep it up you need to hit the soundcard just right.
Edit: this is also a reaction to DT's post above.
Jan
Last edited:
I think this can be facilitated with the 'Hold' function. You can insert say 1V into the ar so it ranges to 0dB, then press Hold to keep it there. Then you can lower the signal to whatever you like but the ar wil stay at 0dB atten.
Or you lower the signal and let the ar range to its max gain of +16dB. That gives you more signal into the soundcard for better dynamic range.
Depends on what you actually want to do.
Jan
I used to be worried about seeing level into the sound card (yes, RMS) and used to hang a volt meter off the input terminals to the sound card. However, after about a day of using my setup, I became very comfortable with everything and removed the voltmeter and have not turned back. Now, I know with Pete Millett's interface, he has a RMS meter. If I were going to build an active unit, I would include the RMS meter. So long winded...nice to have, but not needed. Maybe terminals to connect an external meter if the user chooses.
Thanks Jan, that is all clear now
and please use AR not ar, it makes you(us) all sound like some sort of a pirate ar... ar... And, as an IT specialist, I'm obligated to use upper cast for each first source-letter in the new acronym or label.So you could do 'AuRa' for 'AutoRanger'
but AR works for me, I will never forget it (in this context).https://en.wikipedia.org/wiki/Camel_case
The thing is when I put up a level value it will never be exactly the same as what ARTA or whatever will show. So maybe it is best to just show the sweet spot that is selected, like 'Nom: 0.5V' .
Anyway there is both a single-ended and balanced output so you can always use the unused one for metering.
Jan
I just did a quick check with two soundcards, Juli@ and EMU 1212M The 1212M in +4 mode (Pro?) seems to want about 1V. In consumer (-10) mode its .35 volt. The Juli@ single ended input wants about .35V as well.
I don't have a typical consumer card here at the present (The Xonar DX is on its way to another project).
The steps are about 10 dB apart. It suggests 1V, .316V and .1 V may be good targets for a reference output level. I don't know what the gain/loss steps are for the AR so that's my next task. I'll take the AR set to 1V and swing the level as far as I can looking at the steps. I'll see how it handles 100V.
I don't have a typical consumer card here at the present (The Xonar DX is on its way to another project).
The steps are about 10 dB apart. It suggests 1V, .316V and .1 V may be good targets for a reference output level. I don't know what the gain/loss steps are for the AR so that's my next task. I'll take the AR set to 1V and swing the level as far as I can looking at the steps. I'll see how it handles 100V.
Hi, Jan,
Sorry if I´m missing something, but if you don´t mind, can you clarify my doubt?
If it is a dumb question, I beg your pardon in advance.
Regards,
I´m using Xonar Essence and Juli@ and both, with Virtins, have the best performance (at least with my setup) in about -10dB level of Virtins input level indicator.
Regards,
No dumb question at all! I expect to finalise and start a Kickstarter in about 6 weeks.
Jan
There's something else I'd like some thoughts on.
What are the signal levels you'd expect to input to the ar? For instance, if you want to check a 300W/8ohms power amp at full power, you'd have almost 50V RMS input signal. I would think that's about the max people would use.
This is not the same as the max input level without breaking it - in my current prototype you can easily input 100 or 120V RMS and nothing will break.
But for max input level for outputting the 'sweet spot' voltage to the soundcard, it seems to me a 50V RMS max spec would be more than enough.
Thoughts?
Jan
What are the signal levels you'd expect to input to the ar? For instance, if you want to check a 300W/8ohms power amp at full power, you'd have almost 50V RMS input signal. I would think that's about the max people would use.
This is not the same as the max input level without breaking it - in my current prototype you can easily input 100 or 120V RMS and nothing will break.
But for max input level for outputting the 'sweet spot' voltage to the soundcard, it seems to me a 50V RMS max spec would be more than enough.
Thoughts?
Jan