SAR ADC for high performance audio ADC project [LTC2380-24]

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.

TNT

Member
Joined 2003
Paid Member
Is this for measurement or audio?

//

Hi Frex and all,

I've kindly been guided to this thread which I find to be very interesting indeed :)

As it is I'm also designing an ADC based on the LTC2380-24 and since I'd like to use its full sampling rate (1.536 MHz is likely to work according to LT) I recently started this thread to find a solution to transferring the data from some ADC channels into the PC without drop-outs etc.:

http://www.diyaudio.com/forums/pc-based/292388-seek-programming-assistance-paid.html

Just in case it is of interest to others as well (and please feel welcome to suggest solutions if you have some ideas).

Cheers,

Jesper
 
Hello,

TNT,
Yes, i will try to make a full ADC (+DAC ?) design. As my others projects,
all will be available and i will sale only the bare PCB. But before, there are much job to do !.
So, i will post regularly the project advancement here.

Last week my second ESI juli@ soun-card won't work properly anymore,
with exactly same problem as the previous one...
The control panel crash at start-up and so i'm enable to select synch to SPDIF input.
That seem to be hardware failure, maybe corrupted data in on board EEPROM of juli@...
For now, i try to make some hardware modifications to the juli@ to allow external synch from SPDIF
always working. I'll post pictures when done.

So, without my ESI juli@ i can't test the RIF filter (DF1704) i added on the LT2380-24 ADC.
It's ready to be used, and data in/out to/from CPLD is ok...

I think also for another hardware to replace the ESI juli@, like USBstreamer or XMOS PCB,
but for now i still have some doubt about this solution...


Gentlevoice,

Thank you for the interest,
I answer you on your new thread. That avoid mixing both.

Regards.

Frex
 
Member
Joined 2004
Paid Member
Last week my second ESI juli@ soun-card won't work properly anymore,
with exactly same problem as the previous one...
The control panel crash at start-up and so i'm enable to select synch to SPDIF input.
That seem to be hardware failure, maybe corrupted data in on board EEPROM of juli@...
For now, i try to make some hardware modifications to the juli@ to allow external synch from SPDIF
always working. I'll post pictures when done.

So, without my ESI juli@ i can't test the RIF filter (DF1704) i added on the LT2380-24 ADC.
It's ready to be used, and data in/out to/from CPLD is ok...

I have worked with over 100 Juli@ cards and the only failures I have had were self induced. The hardware is quite robust if not abused. Are you using windows? Are you using the latest version of the drivers?

I have done extensive tweaks to the cards as well. Changing the opamps will get nowhere with the odd footprint but the connector between the digital and the analog section has lots of utility. I have documented most of the pins.
 
I agree with you Demian, the Juli@ is a really good card. I use it for more than ten years
without any issue before now. So, that is incredible that i burn both in two month !
I'm not alone, another user has reported exactly same issue on the ESI forum.
The computer used is a Dell 745 and Win7-64.
The computer configuration has not changed for more than year, nothing that could
explain the issue. All others thing on this computer work fine.

Today i de-solder the AK4114 and the CPLD to cut some tracks and lift some pins.
I want to try to use the AK4114 in hardware mode instead of I2C controlled by the
ViA Envy24 to force the board to work with incoming sampling rate coming from SPDIF input.

I already have tried successfully to synch the juli@ at with 192kHz SPDIF input by injecting
on the 24.576M crytal of Envy24 the 128Fs clock generated by the AK4114 (from SPDIF input).
With this oscillator injection the frequency of crytal is drifted to be equal to (very near)
injected frequency. With this mode i can use FFT without "jump" caused by loss of synch.


When i have bought the second ESI juli@, i looked also others replacement solution but
the list of equivalent product is very short ! Or you must spend much more money !
(The better replacement seem a used RME HDP9632 for 200€).

So even if i can work again with the ESI juli@, i looked seriously for a XMOS board.
That could be interesting for building a digital input sound-card supporting higher sampling rate (384/768k)
and adding isolated SPDIF / I2S inputs..
Probably a new project in perspective...

Demian, if you have any idea about my juli@ issue, i take it.
I personally think that maybe there is corrupted data in the serial EEPROM.
Maybe a dump of it to compare or re-programme a new one and replace it ?
Regards.

Frex
 
Member
Joined 2004
Paid Member
I have a Juli@, RME 9632 and Lynx L22. The RME and the Lynx are both very good. The Lynx has better measured performance on its ADC. The RME uses the same ADC as the Juli@. Depending on your software and Windows version the Juli@ can be the most seamless and just works. The others are far more pro oriented and have a lot of features and capabilities that don't help for measurements and can add opportunities for errors like mismatched sample rates.
for most of my (professional) work I use the Juli@. if its not up to the task (ultra low distortion etc.) then the other cards come into play. If just one card the Juli is adequate for most work. Use the leftover money for other tools and interfacing accessories.
 
I have a Juli@, RME 9632 and Lynx L22. The RME and the Lynx are both very good. The Lynx has better measured performance on its ADC. The RME uses the same ADC as the Juli@. Depending on your software and Windows version the Juli@ can be the most seamless and just works. The others are far more pro oriented and have a lot of features and capabilities that don't help for measurements and can add opportunities for errors like mismatched sample rates.
for most of my (professional) work I use the Juli@. if its not up to the task (ultra low distortion etc.) then the other cards come into play. If just one card the Juli is adequate for most work. Use the leftover money for other tools and interfacing accessories.

Thanks for the detailed review / opinion.
:p
 
Hello,
I contacted ESI last week to their explain the problem i encountered with my two Juli@ card.
They answered me last Friday, in asking to take some screen-shoots of the hardware id panel.
Their first idea about the issue would be EEPROM data failure, as i thought.
I sent asked screen-shoots this Monday, and now i am waiting their verdict.

Frex
 
Member
Joined 2004
Paid Member
I dont think thete is an eprom on that card. The via chip is hard coded. Its all handled in the driver. Its possible that a crystal is misbehaving. Or some windows upgrade has unintended consequences. Im really interested in what you find.

Sent from my LG-H811 using Tapatalk
 
Yes, there is an EEPROM.
You can see it just at the top of the Envy24HT DSP, this is a 24LC02 in SOIC8 package.
The EEPROM store many information from the board manufacturer to say which functionalities
of DSP are used on the circuit board.
It store also others mfg ID that allow to recognize the board from others.

And no, crystal are ok.I tested them.
So, the soundcard work, and i can change any parameters in windows audio control,
not SPIF as input clock source that can be only switched with the ESI control panel.
So, the problem is that i can't open it anymore.

There is not a Windows issue, the juli@ has been check on several computer with
same results.


Frex
 
Member
Joined 2004
Paid Member
That would be really annoying to lose. Maybe a batch of the EEproms have a life issue. Hopefully ESI has a solution that is not buying another. if worse comes to worse. I may be able to extract the prom and read it back to make copies. Its become accepted practice on a number of older HP and Tek instruments. In this case it just sucks.
 
Member
Joined 2004
Paid Member
Interesting. I found more info on this. This link to how to use the EEPRW program Prodigy 7.1 firmware install (which runs in dos.)

I suspect the next steps are to get the eeprw program and read back a working card and then write that image to a defective card. The Juli@ is different from pretty much al the other via based cards in that it supports 176.4. ESI did some special tricks to make that work. I suspect the EEPROM has specifics to support that.
 
Hello and thank you all,

There is no answer yet from ESI.
As suggested Shinja, i made some search and found on hydrogenaudio forum
a post with procedure to DUMP and reprogram the EEPROM of Juli@ or M-Audio192.
Unfortunately, the post is old and the links to download the program are broken.
So i have sent a PM to the member to ask him if he can provide me a new link.
No answer yet.
Anyway, i will wait mid July (back from holidays) and then if no answer i will ask to the generous mkc to dump his Juli@. ;)

Frex

Frex
 
You are right TNT, but i used it to receive the digital data of the LT2380-24 based ADC.
My both Juli@ not work anymore and i try to repare them with ESI help.
So don't worry the project will go ahead anyway.

I have bought a XMOS board (from DIYink) and started to design with it a small unit
able to receive and send I2S and SPDIF stream data (RJ45, coaxial and optical connectors)
at sampling rate up to 384kHz as allow XMOS CPU.
I will also add seeparate MCLK in/out (BNC connectors).
To be safe, all I/O will be of course fully isolated from USB.
That solution would be a real replacement/complement to my juli@.
Regards.

Frex
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.