Hello mrsavage,
The price of the AA5381v1 has not been change from the initial group buy,
the 4 layers gold finish bare PCB cost 35€ worldwide shipping included (priority letter).
For more convenience i can provide the board with the logic array (EPM3068) soldered
and programmed for 15€ more.I only accept Paypal.
As soon Payment done, i send to your e-mail a compete design folder including all design files (schematic, bom, mechanical drawing, EPLD source code and more).
Only Gerber files are not provided.
If you are interested, send me a PM with your personal e-mail address and what you want exactly, and then i will send you a Paypal invoice.
The ADC board reach the specs that we can expect from this IC, not more not less.
But it's difficult to judge only with datasheet, and build a real board is the best way to show how it really perform his job !
I had stated that my previous ADC project based on the AK5394A ADC give better THD results and lower noise floor.You can show all measurement results in posts of this thread.
Despite that, the AA5381v1 ADC is very good and offer many function that make it very
useful for bench measurement.
You can look it's manual HERE.
Regards.
Frex
The price of the AA5381v1 has not been change from the initial group buy,
the 4 layers gold finish bare PCB cost 35€ worldwide shipping included (priority letter).
For more convenience i can provide the board with the logic array (EPM3068) soldered
and programmed for 15€ more.I only accept Paypal.
As soon Payment done, i send to your e-mail a compete design folder including all design files (schematic, bom, mechanical drawing, EPLD source code and more).
Only Gerber files are not provided.
If you are interested, send me a PM with your personal e-mail address and what you want exactly, and then i will send you a Paypal invoice.
The ADC board reach the specs that we can expect from this IC, not more not less.
But it's difficult to judge only with datasheet, and build a real board is the best way to show how it really perform his job !
I had stated that my previous ADC project based on the AK5394A ADC give better THD results and lower noise floor.You can show all measurement results in posts of this thread.
Despite that, the AA5381v1 ADC is very good and offer many function that make it very
useful for bench measurement.
You can look it's manual HERE.
Regards.
Frex
SPDIF to USB converter - options
Hi Frex,
What are the options for getting the SPDIF data back to the PC? Which devices would you recommend (sound cards, converters, ...)?
I found this little XMOS based streamer on miniDSP web:
http://www.minidsp.com/products/usb-audio-interface/usbstreamer
This seems to be a nice option to optically isolate the "SPDIF recorder".
BTW: Thanks for the oportunity to buy this nice tool.
Daniel
Hi Frex,
What are the options for getting the SPDIF data back to the PC? Which devices would you recommend (sound cards, converters, ...)?
I found this little XMOS based streamer on miniDSP web:
http://www.minidsp.com/products/usb-audio-interface/usbstreamer
This seems to be a nice option to optically isolate the "SPDIF recorder".
BTW: Thanks for the oportunity to buy this nice tool.
Daniel
Last edited:
HEllo,
I personally use an ESI Juli@ PCI sound card.
It's a low cost solution (less than 100€) and this card allow to work with the maximum
sampling rate of the ADC without issue (192kHz).
Any FFT software (like Spectralab,Audiotester, and many others) work perfectly with.
If you have the choice, prefer to use a coaxial SPDIF link instead of optical, the jitter figure is genrally better especially if the link is long.
About the miniDSP product, i don't know this one but it seem to could working fine with. Maybe a concern about the driver to be used with FFT software ?...
Regards
FRex
I personally use an ESI Juli@ PCI sound card.
It's a low cost solution (less than 100€) and this card allow to work with the maximum
sampling rate of the ADC without issue (192kHz).
Any FFT software (like Spectralab,Audiotester, and many others) work perfectly with.
If you have the choice, prefer to use a coaxial SPDIF link instead of optical, the jitter figure is genrally better especially if the link is long.
About the miniDSP product, i don't know this one but it seem to could working fine with. Maybe a concern about the driver to be used with FFT software ?...
Regards
FRex
Are the pcb, schematic etc available for the previous one [ based on AK5494A] ?
Thx-RNMarsh
Last edited:
Hello,
As you have seen, the OPA1632 has a very high input bias current.
This makes very difficult in a real world to drive it with a signal source without adding very high input offset voltage due to this bias.
In addition, the AD8599 have lower THD level and lower noise floor and the input buffer stage is more versatile.
Regards.
FRex
As you have seen, the OPA1632 has a very high input bias current.
This makes very difficult in a real world to drive it with a signal source without adding very high input offset voltage due to this bias.
In addition, the AD8599 have lower THD level and lower noise floor and the input buffer stage is more versatile.
Regards.
FRex
@JensH: Are you familiar with Samuel Groner's work with measuring different characteristics of quite some opamps?
There's a link here:
http://www.sg-acoustics.ch/analogue_audio/ic_opamps/pdf/opamp_distortion.pdf
Regards,
Jesper
There's a link here:
http://www.sg-acoustics.ch/analogue_audio/ic_opamps/pdf/opamp_distortion.pdf
Regards,
Jesper
@ FRex
Thanks for the clarification. The OPA1632 is probably not good for an input stage. I would expect it to work fine as an ADC driver though, but I have not tried it yet.
According to the Samuel Groner document the AD8599 actually seems to have a medium to high distortion level. Have you considered the OPA1611/OP1612? IIRC it is the same as OPA211/OPA2211 (except for the DC accuracy), which is included in Samuels paper. The distortion and the noise is lower than for the AD8599.
@gentlevoice
Yes, I know that document. Very useful.
Thanks for the clarification. The OPA1632 is probably not good for an input stage. I would expect it to work fine as an ADC driver though, but I have not tried it yet.
According to the Samuel Groner document the AD8599 actually seems to have a medium to high distortion level. Have you considered the OPA1611/OP1612? IIRC it is the same as OPA211/OPA2211 (except for the DC accuracy), which is included in Samuels paper. The distortion and the noise is lower than for the AD8599.
@gentlevoice
Yes, I know that document. Very useful.
For a general purpose measurement input the current noise of the LME49990 is a bit on the high side. For doing measurements on low impedance circuits this is not a problem, but if the source impedance is more than a few hundred ohm it is not ideal. But apart from that, it is a great op-amp. And certainly lower cost than the AD797.
And I agree, an update of Samuels paper would be very useful.
And I agree, an update of Samuels paper would be very useful.
Hello all,
JensH, i don't have tried to use another OPAMP as input buffer, and I am not sure that the
OPA1611 or OPA211 was available when i had designed it.
I've seen the very good (and long) Samuel's paper about OPAMP, but IMHO i don't think that make noticeable difference.
The LM49990 seem to be very good, only Ibias is slightly high but THD figure is exceptional. For replacing easily the AD8599 i need dual OPAMP, so it can't fit.
Anyway, and has i already say in the forum, i want to build a new ADC using my feedback
of my previous AK5394v1 and AA5381v1 ADC but using more recent ADC IC.
Unfortunately, not many IC outperform the AK5394 despite it's age...
I wait a real datasheet and real world specs of ADC from ESS (ES9102) that promise very high performance, but it's for now only words...
I would be also interested by the AK5388 ADC (4 channels), but this part is hard to find...
DIY...
FRex
JensH, i don't have tried to use another OPAMP as input buffer, and I am not sure that the
OPA1611 or OPA211 was available when i had designed it.
I've seen the very good (and long) Samuel's paper about OPAMP, but IMHO i don't think that make noticeable difference.
The LM49990 seem to be very good, only Ibias is slightly high but THD figure is exceptional. For replacing easily the AD8599 i need dual OPAMP, so it can't fit.
Anyway, and has i already say in the forum, i want to build a new ADC using my feedback
of my previous AK5394v1 and AA5381v1 ADC but using more recent ADC IC.
Unfortunately, not many IC outperform the AK5394 despite it's age...
I wait a real datasheet and real world specs of ADC from ESS (ES9102) that promise very high performance, but it's for now only words...
I would be also interested by the AK5388 ADC (4 channels), but this part is hard to find...
DIY...
FRex
I also plan to make a design with the AK5394A. But the ES9102 definitely looks tempting as well (except for the price of course), so maybe we should try to make a design based on that. Perhaps with the OPA1632 or 2x LME49990 as input buffer?
First I will try to finish my USBStreamer interface though. Then the next step will be ADC and DAC.
First I will try to finish my USBStreamer interface though. Then the next step will be ADC and DAC.
Frex:
The AK5394A is still their best effort. However they have an app note for paralleling several together to lower the noise (or use the AK5388). I have tried some of the others but nothing nearly as good as the AKM parts. AKM lists a new, even better, ADC but the designer told me that "sometime next year" is the best timeline they can offer.
I can probably get some sample parts for you if you want to purse such a project.
I have not been able to get the ESS or the ARDA ADC's so I can't really comment on them. The AKM performance was probably 10 dB lower distortion than the PCM4222 for example. This measured with the demo board stock. I think I can improve on that without too much trouble. The AKM chip is also cheaper and more available.
The AK5394A is still their best effort. However they have an app note for paralleling several together to lower the noise (or use the AK5388). I have tried some of the others but nothing nearly as good as the AKM parts. AKM lists a new, even better, ADC but the designer told me that "sometime next year" is the best timeline they can offer.
I can probably get some sample parts for you if you want to purse such a project.
I have not been able to get the ESS or the ARDA ADC's so I can't really comment on them. The AKM performance was probably 10 dB lower distortion than the PCM4222 for example. This measured with the demo board stock. I think I can improve on that without too much trouble. The AKM chip is also cheaper and more available.
Hello 1audio,
You seem to get the same conclusion of me. The AK5394 still today the best ADC.
As of you, i've read the paper about paralleling many ADC in digital domain to lowered the noise floor (about 3dB better each time you double ADC channels).
It's a interesting approach, but the inherent non-linearity of the ADC is still present
and so THD is not improved (the THD+N is !).
Nevertheless i wanted design an ADC board using this IC, but it's unavailability for me in EU was discouraged me.
So, if a new parts will appear in next year, i probaly can wait that !
And of course, if you are able to provide sample parts, this would be a very convenient way for me to build it .
Regards.
FRex
You seem to get the same conclusion of me. The AK5394 still today the best ADC.
As of you, i've read the paper about paralleling many ADC in digital domain to lowered the noise floor (about 3dB better each time you double ADC channels).
It's a interesting approach, but the inherent non-linearity of the ADC is still present
and so THD is not improved (the THD+N is !).
Nevertheless i wanted design an ADC board using this IC, but it's unavailability for me in EU was discouraged me.
So, if a new parts will appear in next year, i probaly can wait that !
And of course, if you are able to provide sample parts, this would be a very convenient way for me to build it .
Regards.
FRex
I can help you get the chips. Either is possible. Small trade for getting a PCB done.
There are some potential issues. For measurements the circuit may not conform to audiophile standards. I will check to see if the existing input circuit can be improved on. I'm getting really low distortion now with it as the attached image shows.
Its possible that summing two chips in anti-phase will cancel some of the distortion products. it would need very precision balancing for that to actually work. Makes the project more interesting.
What would you use to interface back upstream? SPDIF/AES or USB? Both are options with really good reasons to use them. USB is actually harder with the grounding problems. The cleanest and possibly most flexible would be a USB to SPDIF interface with a clock sync output and an SPDIF interface AD-DA that would support the clock sync or run standalone.
There are some potential issues. For measurements the circuit may not conform to audiophile standards. I will check to see if the existing input circuit can be improved on. I'm getting really low distortion now with it as the attached image shows.
Its possible that summing two chips in anti-phase will cancel some of the distortion products. it would need very precision balancing for that to actually work. Makes the project more interesting.
What would you use to interface back upstream? SPDIF/AES or USB? Both are options with really good reasons to use them. USB is actually harder with the grounding problems. The cleanest and possibly most flexible would be a USB to SPDIF interface with a clock sync output and an SPDIF interface AD-DA that would support the clock sync or run standalone.
Attachments
USB is good if you can tackle the possible grounding and driver issues.
You would have to isolate the USB interface as you mentioned, best way might be through isolating I2S instead of using the AD USB isolator which limits you to 11 Mbps.
Second, you have to be sure that you have an ASIO driver or confirm that the WDM or WASAPI driver (if on Windows) does not tamper with the data.
You would have to isolate the USB interface as you mentioned, best way might be through isolating I2S instead of using the AD USB isolator which limits you to 11 Mbps.
Second, you have to be sure that you have an ASIO driver or confirm that the WDM or WASAPI driver (if on Windows) does not tamper with the data.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.