In article <bnu38u$152ci1$>, "eNo"
wrote:
A buddy of mine just attempted the solution at
http://www.motherboardfaqs.com/article.php?11.255, and couldn't get the coax in to work (the optical IN, for electrical reasons, can't be made to work at all). Does anyone have any solution for the NF7-S to input S/PDIF, either coax or optical? The motherboard does have IN pins on the FI02 connector, so it's curious no formal/standard solution is formally offered.
First of all, are you sure that bracket has the functionality you think it has ? I've seen two styles of connector on the brackets,
1x4 three wires = SPDIF-out only
Wires are +5, GND, SPDIF-out
2x3 five wires = SPDIF-in + SPDIF-out (if only three wires are used it is still output only). Wires are +5, GND, GND, SPDIF-in, SPDIF-out
The review for the bracket on the Newegg site also mentions this -
http://secure.newegg.com/app/CustratingReview.asp?item=13-999-753
As far as I know, SPDIF bracket connectors have a fixed function. An "in" is always an "in", an "out" is always an "out". I am suspicious of the info I see in the motherboardfaqs.com link above, because the three wires should be connected to different pins, never shorted together.
NF7-S FPIO2
MIC 1 x x 2 GND
No Conn 3 x x 4 VCC
Front Right 5 x<--->x 6 Front Right
No Conn 7 x (pin 8 missing)
Front Left 9 x<--->x 10 Front Left
GND 11 x x 12 SPDIF-in
VCC 13 x x 14 SPDIF-out
If anything is going to work, it will be BLACK to 11 (GND), RED to 13 (VCC), and WHITE to 14 (SPDIF-out). That will give you SPDIF-out optically and via copper at the same time. AFAIK, no other functionality is hidden in that bracket.
Over in Asus country, nobody has been able to get SPDIF-in to work with an Asus bracket to a Realtek ALC650 chip (the ALC650 is the same chip used on the NF7-S). After looking at the docs today, I'm convinced it will be hard to get working.
Docs for the ALC650 are here:
ftp://152.104.125.40/pc/ac97/alc650/alc650_data5.zip
See the doc "ALC650 Data Sheet_1.28.pdf" page 41 in the zip file.
I see the best hope now of getting the SPDIF-in to work, by using the SN75179. The circuit would look like this, as seen on page 41 of the Realtek datasheet. The SN75179 is a line driver/receiver, and the receiver can convert low level signals into logic signals
- it functions like an amplifier for logic signals. My belief is the ALC650 needs to see a 3.3V logic level signal, not a <1 volt level signal. A TOSLINK module gives a logic level signal, as does the 75179.
_______ -------
| | 8 | |
| .--------------| | 2 ------------
| 7 | |_____|75 ohm |__ NF7-S pin 12 | O--------------| | |5% resistor| "SPDIF-in" | | 7 | ------------ |_______| | 5 | | 1 | RCA Conn | 7 | 1 | 9 | NF7-S pin 13 ------+------| | "VCC" | | |
0.1uf ___ | |
cap ___ | | Pin 3, 5, 6 are no connect. | 4 | | (Purists would ground pin 3 NF7-S pin 11 ------+------| | but it isn't essential.) "GND" | |
--------
I added the 75 ohm resistor to the output, to limit the current that can flow from the 75179, if the output voltage swings greater than 3.3 volts. To remove it from the circuit, I would have to verify the output of the 75179 first, that it doesn't swing too high. If it didn't, the resistor could be removed. The resistor also provides a measure of series damping for transmission line theory.
Assemble on a small board, that can be mounted right next to the FPIO2 header. This is essential, because SPDIF has a bandwidth of 6MHz, and the wire running from pin 2 of the 75179 to the header should be kept to 1/2" or so. The 0.1 microfarad capacitor should be mounted on top of the IC, to keep the leads short. The power leads can be a little longer if you want, as the capacitor cleans up the power at the IC. The capacitor would be a non-polarized ceramic cap - you could even substitute a small electrolytic if you had to (10uF 16V).
http://www-s.ti.com/sc/ds/sn75179b.pdf
There are a number of other RS-485/RS-422 parts that can be used in place of the 75179, but each one has some issues. Some of them run from +3.3V (and the FPIO2 header only has +5), so you would need a voltage regulator. Some are purposefully slowed down, so they cannot carry the maximum 10Mbit signal that the 75179 line receiver can handle.
Another chip that might work is the 75176 (SN75176BP 8 pin dip). The receiver and transmitter are joined together, so only one can be used at a time. The '179 transmitter and receiver are separate, and could be used at the same time (i.e. if you are doing both SPDIF in and out).
http://www-s.ti.com/sc/ds/sn75176b.pdf
The SPDIF-in circuit would look like this:
_______ -------
| | 6 | |
| .--------------| | 2 ------------
| 7 | |_____|75 ohm |__ NF7-S pin 12 | O--------------| | |5% resistor| "SPDIF-in" | | 7 | ------------ |_______| | 5 | | 1 | RCA Conn | 7 | 8 | 6 | NF7-S pin 13 ------+------| | "VCC" | | |
0.1uf ___ | |
cap ___ | | Pin 4 is no connect. | 5 | | Pin 2, 3, 5 are connected NF7-S pin 11 ------+-+----| | together and go to GND (11) "GND" | 2 | | on FPIO2. +----| | | 3 | | (If the circuit doesn't work the +----| | first time, try a 75ohm 1/8W
-------- resistor between 6 and 7 right at the chip, to terminate the line)
This web page is a good backgrounder on SPDIF, and explains the several variants, including the fact that the two pin "digital" header on the back of some DVD drives is in fact a 5V logic output. A signal like that could be passed directly to the ALC650 SPDIF-in pin, as long as the signal swing from the drive didn't exceed one diode drop more than 3.3V (about 3.9 volts would be a good max).
http://www.epanorama.net/documents/audio/spdif.html
So, at least right now, I don't see a retail solution to the problem. An SPDIF to AES/EBU converter might work, except that a properly designed converter will change the data stream, which might not make the ALC650 too happy. I think in any case, what ever you try will be an expensive experiment.
I would only construct the circuits above if you have some prior experience. If the above circuits refuse to get the LOCK bit set in the SPDIF interface, I would try adding a 75 ohm resistor across the two signals coming from the RCA jack to the differential receiver. The Realtek schematic doesn't show the resistor, but a terminator at the end of the line might help, as it is shown in the epanorama webpage circuits.
RCA jack - Radio Shack 274-852 (has a nut for bulkhead mounting?) SN75176BP - Digikey.com 296-1739-5-ND 8 pin DIP plastic package
0.1uF cap - Digikey.com 1109PHCT-ND 0.1uF 50V X7R Axial 10 for $3.44 75 ohm R - Digikey.com P75BACT-ND 75ohm 1/4W 5% (smallest I could find) Pins - Digikey.com A28351-ND 1x2 receptacle assembly 0.1" ctr Need (2) for 4 pins. Buy some spares. Perfboard - Radio Shack 276-1395 (cut to size)
I don't have any SPDIF gear, so the ideas above are just unproven theories

Sooner or later, someone is going to get that ALC650 working.
Paul