What is the clear benefit to a SAR ADC here? The downside is obvious; it is much more work to implement because you will have to handle oversampling and digital filtering yourself. If you are picking a part that is 1Msps it is not clear to me that you will get an effective sample rate any higher than a good audio converter anyway once you take into account oversampling.
The switchcraft mini's look pretty good. Same as the standard but smaller.
I don't like like these newer plastic XLR type..
Attachments
Hi David,
Heavens no! Only metal connectors.
I don't like the smaller barrel as jacks to get bumped now and again. It just seems liek a better idea to use the largest metal barrel that can be adapted down rather than adapt to a larger connector having mechanical advantage on the mini XLR. I think too, most of us will see the full size XLR connector. Most of us will be using the XLR to cables with a variety of plugs on the ends. A different cable for each plug type. That's how I approach things here. I'm just redoing the speaker dummy load box with the Speak-on jack. Different cables will terminate in the various styles. The Speak-on connector will make the job of changing cables into a pleasure rather than the PITA it currently is. The same idea carries over to the tester and will work very well I think. The larger XLR makes it easy to install a variety of wire types.
You can always go more compact and use a 180° 5 pin DIN jack! 
Look at it this way. Using the HP 339A (and most other units) means dealing with 5-way binding posts on 3/4" centers. At least what we are discussing now is completely shielded - unlike the binding posts.
Would I be correct in assuming you are already standardized on the mini XLR connector format?
-Chris
Heavens no! Only metal connectors.
I don't like the smaller barrel as jacks to get bumped now and again. It just seems liek a better idea to use the largest metal barrel that can be adapted down rather than adapt to a larger connector having mechanical advantage on the mini XLR. I think too, most of us will see the full size XLR connector. Most of us will be using the XLR to cables with a variety of plugs on the ends. A different cable for each plug type. That's how I approach things here. I'm just redoing the speaker dummy load box with the Speak-on jack. Different cables will terminate in the various styles. The Speak-on connector will make the job of changing cables into a pleasure rather than the PITA it currently is. The same idea carries over to the tester and will work very well I think. The larger XLR makes it easy to install a variety of wire types.
You can always go more compact and use a 180° 5 pin DIN jack! Look at it this way. Using the HP 339A (and most other units) means dealing with 5-way binding posts on 3/4" centers. At least what we are discussing now is completely shielded - unlike the binding posts.
Would I be correct in assuming you are already standardized on the mini XLR connector format?
-Chris
Hi David,
Heavens no! Only metal connectors.
I don't like the smaller barrel as jacks to get bumped now and again. It just seems liek a better idea to use the largest metal barrel that can be adapted down rather than adapt to a larger connector having mechanical advantage on the mini XLR. I think too, most of us will see the full size XLR connector. Most of us will be using the XLR to cables with a variety of plugs on the ends. A different cable for each plug type. That's how I approach things here. I'm just redoing the speaker dummy load box with the Speak-on jack. Different cables will terminate in the various styles. The Speak-on connector will make the job of changing cables into a pleasure rather than the PITA it currently is. The same idea carries over to the tester and will work very well I think. The larger XLR makes it easy to install a variety of wire types.
Look at it this way. Using the HP 339A (and most other units) means dealing with 5-way binding posts on 3/4" centers. At least what we are discussing now is completely shielded - unlike the binding posts.
Would I be correct in assuming you are already standardized on the mini XLR connector format?
-Chris
Nope never used the mini's. Just thought they'd be less likely to get bumped around since they wouldn't stick out as far.
I see your joking about those 5 pin din thingys. XLR also comes in 4,5,6 pin and maybe more. OKay regular XLR then.
Hi David,
But, I like the multi-pin series of XLR connectors. The DIN plugs have very close pins that are typically oxidized and no fun to solder. Then there is the issue of getting all that wire through the connector insulating cover.
My current FM stereo generator uses the 180° 5-pin DIN connectors. I have one that is fully populated too.
-Chris
But, I like the multi-pin series of XLR connectors. The DIN plugs have very close pins that are typically oxidized and no fun to solder. Then there is the issue of getting all that wire through the connector insulating cover.
My current FM stereo generator uses the 180° 5-pin DIN connectors. I have one that is fully populated too.
-Chris
What do you mean by "yourself"? There are ready to go solutions.
As for the sample rate, that was my other question, because as far as I know, at the moment, there is no USB solution that goes above 192KHz.
Mixing boards are far from the only equipment with XLR and/or TRS connectors that one encounters in real life. One also finds DACs, ADCs, equalizers, crossovers, mic preamps, etc. etc. with these connectors. Perhaps you lack the experience of working with professional audio gear. Your life may change in the future, why not be ready?
In fact balanced I/O is far less susceptible to noise and EMI pickup when installed in a real world system, and any good piece of audio gear with pretensions to high performance should have it.
Pro audio gear is just like consumer gear in that its performance level varies all over the map from sublime to ridiculous. However, it is very likely that pro gear will have balanced analog I/O and that TRS or XLR connectors will be involved. They are bulky and may be generally a little low on bling, but they are generally technically solid and reasonably priced.
It is very advantageous, to say the least, to be able to hook equipment up, prior to testing it.
OTOH, if you are going to build test equipment to impress visiting firemen and not for the purpose of testing audio gear, then please use what ever connectors you find most aesthetically pleasing.
The usual convention is to use 3 pin XLRs for audio, and 5 pin or more XLRs for things like power and control.
The appearance issue was raised in post #523.
Sure, the mini XLRs look sweet, but out in the real world, almost nothing mates with them.
Do try to leave emotions out of this.
Not sure what you have in mind here, but there certainly are many USB DACs that can go above 192 kHz.
There are many DAC's and USB to I2S out converters available. But there are not many options when it comes to I2S input. I have looked at a couple of options. One is the USBStreamer + some interface. The other is the SA9227. Both will be able to support 384 kHz. And the USBStreamer could even be used for 768 kHz.
Another challenge is the converters, primarily the ADC. The ADC that I use now (AK5394A) does not support sample rates higher than 216 kHz, so in reality 192 kHz, to fit into standard sample rates. The AK5397 does support higher sample rates. But in this case the distortion is not quite as good as for the AK5394A. And the frequency response at the high sample rates is far from flat. Not ideal for test equipment.
So for now I concentrate on up to 192 kHz sample rate. Higher sample rates is something I will look at later, as a possible upgrade. In this case I probably need to go to a different ADC architecture, using a fast ADC and external filtering. Perhaps a dual solution with two different ADC's.
Another challenge is the converters, primarily the ADC. The ADC that I use now (AK5394A) does not support sample rates higher than 216 kHz, so in reality 192 kHz, to fit into standard sample rates. The AK5397 does support higher sample rates. But in this case the distortion is not quite as good as for the AK5394A. And the frequency response at the high sample rates is far from flat. Not ideal for test equipment.
So for now I concentrate on up to 192 kHz sample rate. Higher sample rates is something I will look at later, as a possible upgrade. In this case I probably need to go to a different ADC architecture, using a fast ADC and external filtering. Perhaps a dual solution with two different ADC's.
Hi JensH,
Sampling at 192 KHz at those distortion levels is perfectly fine. Everyone has to understand that you've been down this road already. You have made the tough choices already.
The dual solution is how the test and measurement companies do it, but there are other factors involved in trying to sample alternatively. I've been told that it isn't that straight forward. Sticking with this solution is just fine.
-Chris
Sampling at 192 KHz at those distortion levels is perfectly fine. Everyone has to understand that you've been down this road already. You have made the tough choices already.
The dual solution is how the test and measurement companies do it, but there are other factors involved in trying to sample alternatively. I've been told that it isn't that straight forward. Sticking with this solution is just fine.
-Chris
Not sure what you have in mind either, since we are talking about I2S to USB and not the other way around.
Also let's not forget that interfacing includes the software on the pc/mac side, too. What happens there?
I am not against 192KHz, just wondering since it was mentioned in the early posts looking into higher sample rates.
But we are here for discussion, right? I don't believe that my question creates a problem, or insults anybody.
And since we are discussing about it, is there any schematic we could look at, to make it more meaningful? Or is it going to be a commercial product?
Hi Nikitas,
You are correct, which is why those posts were not deleted. I'm just pointing out that JensH has it mostly designed already. I am plenty certain that if we wait, there will always be a new chip around the corner that is a little better. But higher sampling rates are not going to buy you much. Case in point, the Keysight U8903 goes up to about 100 KHz. They have all the money and labs possible to create a marvel, yet they are stuck in the same place as JensH is.
I've been watching the new instruments, so I have an idea of what the state of the art is at the moment. What Jens is designing will probably be limited more by analog stages instead of the capability of the A to D stage he is designing. If you have ever worked on Agilent or Keysight equipment (HP too), you will see that they begin with a chassis within a chassis. Then each subsection has a shielded box to live in. All of that gets very expensive. Just putting this into perspective for you.
-Chris
You are correct, which is why those posts were not deleted. I'm just pointing out that JensH has it mostly designed already. I am plenty certain that if we wait, there will always be a new chip around the corner that is a little better. But higher sampling rates are not going to buy you much. Case in point, the Keysight U8903 goes up to about 100 KHz. They have all the money and labs possible to create a marvel, yet they are stuck in the same place as JensH is.
I've been watching the new instruments, so I have an idea of what the state of the art is at the moment. What Jens is designing will probably be limited more by analog stages instead of the capability of the A to D stage he is designing. If you have ever worked on Agilent or Keysight equipment (HP too), you will see that they begin with a chassis within a chassis. Then each subsection has a shielded box to live in. All of that gets very expensive. Just putting this into perspective for you.
-Chris
And different people design different parts of instrument... digital circuitry expert, analog circuitry expert, display... This is a one man band so if each one of us asks for one additional feature than project goes nowhere. This very interesting project seems like to be in its mature state and lets hope to see it in its final stage.