8 Channels ADAT converter

[Skylar]

The ADAT I/O on the 1616 & 1616m can be connected together easily.

If you have two audiodock units, you will need to have two computers to host each one, as they do not operate standalone.


Are you just wanting to expand the analog I/O on your 1616m?
If so, your best option is still going to be a standalone A/D D/A converter.


For further E-MU questions (from anyone!), just send me an email or post a new topic.

 

[skipwave]

Hey Skylar,

Did you pursue this any further?

I picked up a OctopreADAT card for cheap and with the help of your info going to run it standalone.

The one question which I can't suss out from the pics/pinout is whether the card has balanced line receivers ahead of the a-d or are the balanced inputs direct to converters?

Thanks!

 

[netnoggin]

I'm going down the same road, my board just arrived yesterday. When I get home from work I'll look closer, but I did see NJM5532 dual op-amps positioned at each A/D chip. There wouldn't be too many other reasons for them to be there. Get your magnifying glass out and you can probably detect if that's what they are for.

http://ko-online.com.ua/files/njm5532.pdf

But I'll check too this evening.

NN

 

[netnoggin]

Confirmed at least that one of the outputs of the opamp chip is connected to one of the inputs on the A/D chip. Because the board is multilayered, I can't tell much past that. But it seems to verify that the opamp chip is being used for signal conditioning or buffering or both.

What I would really like to know is if there is anything on the front of the opamp, before it goes out to the connector. I just really can't say at this point. I'm not a fan of sending chip inputs to the outside world without something in between. depending on the levels, I may at least put some low value resistance in series with the input connectors I am installing. We'll see. The only thing I know about the levels pre-testing comes from the Octopre manual:

The signal which is converted to digital is taken just before
the analogue output where 0dBfs (the maximum level
before digital overload) is equivalent to a 21dBu level at
the analogue output.

NN

 

[skipwave]

Thanks, N!

Did you try metering the resistance from the input connector pad to the opamp input pin? It will likely be a bit wacky thanks to parallel components, but should show that some supporting components complete the input circuit.

 

[pyjaman]

Hello,
interesting thread...

another candidate? :
http://cgi.ebay.fr/RME-OCTAMIC-ADC-ANALOG-TO-DIGITAL-CONVERTER-CARD-NEW_W0QQitemZ370184752995QQcmdZViewItemQQptZLH_DefaultDomain_0?hash=item5630bb0f63
a little documentation in the octamic manual:
http://www.rme-audio.de/download/octamic_e.pdf

What do you think?

Laurent.

 

[netnoggin]

Looks like it has possibilities, although the price gives me pause. I paid less than half of that for the Focusrite board, and the specs look similar. Also, Skylar has already done the legwork on figuring out the interface on the Focusrite board, which is hugely appreciated!

NN

 

[netnoggin]

For those looking for the Focusrite board, there's no shortage of people selling it....

http://www.google.com/products?hl=en&as_q=focusrite+adat+option&as_epq=&as_oq=&as_eq=LE&num=10&scoring=r&as_occt=any&price1=&price2=&show=dd&safe=off

NN

 

[hobiesound]

hi people,

focusrite is stopping production of the octopre ad card btw.

grtz,

thomas

 

[netnoggin]

Any idea why?

Maybe too many people are outboarding them  ;D

It is awful easy with this card. One thing I did run into - don't underestimate the current requirements if you build a power supply for this. Notice the 3.3 volt, 3 amp regulator on the board? I'm thinking they didn't use a 3 amp instead of a 1 amp just because they had them laying around. I believe this is fed from the 5 volt line, as my regulator supplying it was working hard.

More to come, haven't quite got it together except for a smoke test (none  )

NN

 

[netnoggin]

Thanks, N!

Did you try metering the resistance from the input connector pad to the opamp input pin? It will likely be a bit wacky thanks to parallel components, but should show that some supporting components complete the input circuit.

I did finally scope it out some more. the input from the ribbon cable goes to those 10uf electrolytics, then to a resistor (don't remember the value), then to the input of the 5532. I'm happy, although purists would probably fuss about the garden-variety caps in the chain.

NN

 

[skipwave]

Thanks for the update. I suspected all those 10uF were input coupling. We could upgrade them to please the snobs, or not.  ;D

 

[Skylar]

Just noticed the question.

Looking at the AKM5383 data sheet, it has balanced inputs.
So the 5532 opamps must be buffers rather than balancing receivers.

I haven't done any more with this.
Just tested it to make sure it worked, ran RMAA for noise specs, and did a listening test for sound.
Everything worked great for me, but I think I'm going to sell the Octopre, as I really need A/D and D/A (not just D/A).

 

[skipwave]

I really need A/D and D/A (not just D/A).

Me too. Which would make the Octopre-LE ADAT card a better choice, since it does A/D/A.

We have one at the studio. I can open it up and check the headers.

 

[netnoggin]

A few notes from my project are attached.

 

[skipwave]

Thanks for sharing your notes!

I'm recycling a case as well, from a Furman headphone amp with a dead power transformer.

Same story here; The hardest part has been sourcing an appropriate power transformer locally. I'm thinking of just ordering from Antek and waiting the extra few days for delivery.

Could you share the part number on the headers you used to mate with the ribbon cable connectors?

Also, what voltage are you feeding the 5v reg with?

 

[netnoggin]

Those were in bubble packs down at the local parts house, with an inch of dust on them. Probably didn't think they were ever going to sell them. Anyway, one of them was right angled (GC Electronics 41-973 20-pin PCB Header connector, 100x100 centers) and the other was straight, but I don't have the number on the straight version. I think it was only one off, like 41-972 or something.

This looks like a better one: http://rocky.digikey.com/WebLib/CW%20Industries/Web%20Data/CWN-3,4.pdf

Digikey listing: http://parts.digikey.com/1/parts/327397-conn-hdr-20-pin-gold-vert-pcb-cwn-350-20-0000.html

 

[netnoggin]

On the 5 volt question - I am feeding the regulator from the same source as the +16V regulator - the positive output of the bridge rectifier. The transformers have 15V secondary's, so in series that's 30V CT. Can't remember the resulting voltage at the filter cap, seems like it was about + and -21 volts or so - well below the limits of the 3-pin regulators. Of course, that's a healthy differential across the 5 volt reg, so there is definitely some power being dissipated. And I suppose a cleaner, more elegant solution would be to either use a separate transformer or a separate secondary on the transformer. But neither was available. Bottom line is it's working fine.

 

[skipwave]

That's what matters! People love to compliment fancy (read expensive) front panels and vintage knobs, but not nearly enough due is given to the clever solutions that get it done.

Your build looks great to me. 8)

 

[netnoggin]

Skylar (or anybody else that has an Octopre), I just wanted to confirm something. When you first power it on, what are the default settings? 24 bit and 44.1?

On the Octopre front panel, the bit-depth LEDs are arranged in descending order (24, 20, 16), but the sampling freq is ascending (44.1, 48, etc.). I thought I had the bit-depth LED row reversed, but now I'm not so sure. Upon powering up, both rows start to the left. Also, if I in fact had them reversed, you would think pushing the button would increment them right to left instead of left to right, which it does not do. Both rows of LEDs increment left to right.