I would like to try and hook my transport up to my DAC using AES/EBU. The Dac only has co ax inputs... No transformer on dac input, just a 75 ohm resistor.
The transport does have a txf coupled AES/EBU output.
Will the attached conversion cct sketch do the trick? I think fig 1 would give me the right impedance to hook onto the co ax input of my DAC without loosing the AES/EBU benefit....
Any advice welcome!
Thanks
The transport does have a txf coupled AES/EBU output.
Will the attached conversion cct sketch do the trick? I think fig 1 would give me the right impedance to hook onto the co ax input of my DAC without loosing the AES/EBU benefit....
Any advice welcome!
Thanks
See this very good site: http://www.scientificonversion.com
They provide excellent transfo for this application. I'm planning in using this part myself in my DCA:
SC939-06: Transfo, Pulse 1.2:1:1.2, 0.20mH
They provide excellent transfo for this application. I'm planning in using this part myself in my DCA:
SC939-06: Transfo, Pulse 1.2:1:1.2, 0.20mH
More ideas...
http://www.epanorama.net/documents/audio/spdif.html (around the middle of the page)
/U.
http://www.epanorama.net/documents/audio/spdif.html (around the middle of the page)
/U.
Thanks guys, good links
Thanks a lot guys, I guess what I need to do is actually get the part # of the Vitec transformer I have and see what it's turns ratios are, from there I hopefully can pick a suitable circuit.....
This is all to use with my Theta Data Basic II and DS Pro Basic II.
I have a transformer on a reclock board from them that I am not using at the moment. Once I make up my mind between AES?EBU and co ax I will implement the output reclock board as required. It is configured for co ax as standard.
I could not see prices on Scientific Conversion site, does anyone know what these things cost? Only problem is that the bank transfer will possibly cost more then the TX. Using the Vitec will be first choice.....
Guillaume
Thanks a lot guys, I guess what I need to do is actually get the part # of the Vitec transformer I have and see what it's turns ratios are, from there I hopefully can pick a suitable circuit.....
This is all to use with my Theta Data Basic II and DS Pro Basic II.
I have a transformer on a reclock board from them that I am not using at the moment. Once I make up my mind between AES?EBU and co ax I will implement the output reclock board as required. It is configured for co ax as standard.
I could not see prices on Scientific Conversion site, does anyone know what these things cost? Only problem is that the bank transfer will possibly cost more then the TX. Using the Vitec will be first choice.....
Guillaume
Another link
This technical note from Rane is probably the most informative and I think it has given me the information I need....
http://www.rane.com/note149.html#foot
This technical note from Rane is probably the most informative and I think it has given me the information I need....
http://www.rane.com/note149.html#foot
I don't like any of the Scientific Conversions transformers. Not familiar with the other one, but try to find out the leakage inductance and/or interwinding capacitance.
As for your sketches.............worry about making the input look like 110 ohms, or whatever impedance that you cable has. Then play games with resistors..............
Jocko
As for your sketches.............worry about making the input look like 110 ohms, or whatever impedance that you cable has. Then play games with resistors..............
Jocko
I will do that Jocko
I am not likely to use my original idea at all. The Rane note gives me the info needed to do this properly. Worst case I may just buyt the Neutrik transformer they mention. At least they have an agent here in SA. I guess we have the sun and beautiful wildlife, but it really s#cks that we can not just get some of these sexy components you guys buy of the shelf.... Paypal does not even accept South African registered credit cards.....
Later
I am not likely to use my original idea at all. The Rane note gives me the info needed to do this properly. Worst case I may just buyt the Neutrik transformer they mention. At least they have an agent here in SA. I guess we have the sun and beautiful wildlife, but it really s#cks that we can not just get some of these sexy components you guys buy of the shelf.... Paypal does not even accept South African registered credit cards.....
Later
Mouser sells a cheap Chinese knock-off of a Schott transformer that I used to use. It is ok............has minor flaws, but is cheap, and easy to get.
Digi-key now sells something that may be Schott, built under a different name. More expensive..........seems to measure as good as the Shotts.
Of course.......if you do use a transformer..........get ready to make a Zobel. Using one without a network makes for rotten sound.
Another reason that I refuse to use SC is that they do not publish leakage inductance. You need that in order to design the network.
Jocko
Digi-key now sells something that may be Schott, built under a different name. More expensive..........seems to measure as good as the Shotts.
Of course.......if you do use a transformer..........get ready to make a Zobel. Using one without a network makes for rotten sound.
Another reason that I refuse to use SC is that they do not publish leakage inductance. You need that in order to design the network.
Jocko
Cheapskate method?
How close can you get with an H-Pad?
Is a transformer really necessary for the impendence matching or is it there to provide electrical isolation too?
I think the source side on the AES is around 110 ohms...
Assuming that both the source and receiver will work well with current matched loads, I plugged 110 ohm source and 75 ohm destination into the calculator located here:
http://www.ptme.com/et/audio/reference/calculators/db/hpad.htm
and guessed at 15 db of attenuation.
It returned these results that are not exactly common 5% resistor values.
Assuming that these ideal resistors are cheap and available, would this pad work or will it distort the signal to the point of adding jitter?
How close can you get with an H-Pad?
Is a transformer really necessary for the impendence matching or is it there to provide electrical isolation too?
I think the source side on the AES is around 110 ohms...
Assuming that both the source and receiver will work well with current matched loads, I plugged 110 ohm source and 75 ohm destination into the calculator located here:
http://www.ptme.com/et/audio/reference/calculators/db/hpad.htm
and guessed at 15 db of attenuation.
It returned these results that are not exactly common 5% resistor values.
Assuming that these ideal resistors are cheap and available, would this pad work or will it distort the signal to the point of adding jitter?
Attachments
I can measure the SC937's leakage.Jocko Homo said:
A zobel on an AES transformer, interesting... any pointers/resources on how to go about designing one of these? and how wideband can you make these? (the application has to accept 32-192 KHz sampling rates)
Also, any pointers on how to explain the part change to my boss? "yep, i'll be adding a resistor and a capacitor to this here transformer circuit, if i don't do it, it'll sound rotten"
If it helps, his response will either be "can I have some of the drugs you're on?", "what? you're fired" or "if the bits arrive OK and the clock recovery and SRC attenuate whatever jitter gets added, how in the freakin' blue jesus are you gonna hear a difference?"
Have you seen:
http://www.gpsys.com/products/accessories.htm
or
http://www.canare.com/index.cfm?objectid=DCBBA8D3-3048-7098-AF5E76E3D1E1CA3D
Ive used both for their intended and other purposes, such as this one.
Both are packaged transformers for converting AES 110 to 75 ohm transport. I don’t have any kicking around so I cant take one apart and tell you what special sauce or OEM's they are using.
While they do match the impedance there is still the level difference that needs to be dealt with.
These guys make sell or sorts of solutions for just about anyone’s budget, just need to use the right key words to find it on their site:
http://www.markertek.com/SearchProduct.asp?item=SPDIF-AES-6&off=19
http://www.gpsys.com/products/accessories.htm
or
http://www.canare.com/index.cfm?objectid=DCBBA8D3-3048-7098-AF5E76E3D1E1CA3D
Ive used both for their intended and other purposes, such as this one.
Both are packaged transformers for converting AES 110 to 75 ohm transport. I don’t have any kicking around so I cant take one apart and tell you what special sauce or OEM's they are using.
While they do match the impedance there is still the level difference that needs to be dealt with.
These guys make sell or sorts of solutions for just about anyone’s budget, just need to use the right key words to find it on their site:
http://www.markertek.com/SearchProduct.asp?item=SPDIF-AES-6&off=19
If you have gone to all the trouble of building in a SRC, then the difference may be small. But then, if you have gone to all that trouble, I can't imagine the extra dime breaking the budget. Even in Canadian money.......!
Unless you making a million, in which case it better not cost a dime.
How to design:
Easy.....you know the leakage inductance, figure out what value cap will cancel out that value. Assuming that you are talking 75R, a good transformer will need around 18-22 pF. I have some that only needed 15 pF. A mediocre one will take around 33 pF, and a rotten one needs 47 pF.
How to explain:
Talk him into buiying a TDR. Yeah, I bet that won't fly........
So, get a decent 'scope, and make one. You will need a SPDIF source, a BNC "Tee", and your circuit.
BTW.......if you do hook one up to a TDR......make sure that you show him all the **** the Crystal RX chip dumps back unto the line. He will really hate you then.
BTW......is he related to Eugene Krabs? "Argh!!!!!! Me money!"
Edit: I can't believe that SY won't let me say C__P!
Jocko
Unless you making a million, in which case it better not cost a dime.
How to design:
Easy.....you know the leakage inductance, figure out what value cap will cancel out that value. Assuming that you are talking 75R, a good transformer will need around 18-22 pF. I have some that only needed 15 pF. A mediocre one will take around 33 pF, and a rotten one needs 47 pF.
How to explain:
Talk him into buiying a TDR. Yeah, I bet that won't fly........
So, get a decent 'scope, and make one. You will need a SPDIF source, a BNC "Tee", and your circuit.
BTW.......if you do hook one up to a TDR......make sure that you show him all the **** the Crystal RX chip dumps back unto the line. He will really hate you then.
BTW......is he related to Eugene Krabs? "Argh!!!!!! Me money!"
Edit: I can't believe that SY won't let me say C__P!
Jocko
Both seem important if driving a 50 meter cable. These are the kinds of things AES is intended to address.
I was assuming the application was to have the transport and DAC side by side, less than a meter of interconnect. If that ground loops, there’s likely something wrong with the chassis design. (Not that it doesn’t happen)
I suggested the pad with the thought that less could be more for this aplication. No trasnformer inductence, no need to zobel it. Perhaps even less jitter...
The transport and DAC will be close to each other but in several comparisons in the past I have preferred AES3 on some Transport and DAC combinations. I now have an identical length of BNC terminated Apogee Wyde Eye D/A cable as well as the AES3 version of this cable. My DAC does not have the AES3 input. Next week Friday I will go to my friend (he has the same transport as me) but a Proceed DAP DAC I think with AES3 input. He uses an expensive analysis plus AES I/C so that has never been to fair a comparison with anything I used until now.
We listened to the AES3 Apogee vs his silver Analysis Plus last week and although better, if he had to do it again he would probably not spend the extra on the A+ (It is 30 x the price) as the Apogee is a damn fine cable. It eats the Madrigal AES3 cable we also had for comparison.....
We will AB the sound of AES3 vs Co Ax on his rig using the Apogee cables. Then I will really know if I want to pursue AES3 in my rig.....
I do not wish to do a sub optimal conversion that ultimately looses the baby with the bath water and turns inferior to co ax.
I am sure the H pad would work and lock onto the stream etc but I am trying to achieve the best can with my limited ability ......
We listened to the AES3 Apogee vs his silver Analysis Plus last week and although better, if he had to do it again he would probably not spend the extra on the A+ (It is 30 x the price) as the Apogee is a damn fine cable. It eats the Madrigal AES3 cable we also had for comparison.....
We will AB the sound of AES3 vs Co Ax on his rig using the Apogee cables. Then I will really know if I want to pursue AES3 in my rig.....
I do not wish to do a sub optimal conversion that ultimately looses the baby with the bath water and turns inferior to co ax.
I am sure the H pad would work and lock onto the stream etc but I am trying to achieve the best can with my limited ability ......
Transformers make impedance matching harder, which is why I do it............
Ground loop elimination......yeah, helps some...........
Getting a differential signal...........there is the key. You already have one. You can buy a lot of resistors instead of a transformer. Your life in this case will be simpler. Even though the two units will be side by side, don't make the cable short.
Hmmm...........just how are you going to make your cable?? XLR on one end...........BNC/RCA on the other? Or are you going to gouge a big hole in the RX unit to stick a XLR on it??? XLRs actually do measure 110 ohms, but they look like crud on a TDR. Too much fringing with all the exposed part of the connector.
Jocko
Ground loop elimination......yeah, helps some...........
Getting a differential signal...........there is the key. You already have one. You can buy a lot of resistors instead of a transformer. Your life in this case will be simpler. Even though the two units will be side by side, don't make the cable short.
Hmmm...........just how are you going to make your cable?? XLR on one end...........BNC/RCA on the other? Or are you going to gouge a big hole in the RX unit to stick a XLR on it??? XLRs actually do measure 110 ohms, but they look like crud on a TDR. Too much fringing with all the exposed part of the connector.
Jocko
Hi Jocko
Great to have your input (excuse pun) on this matter as you certainly know your way around this stuff.
The Idea was to have a great big hole in the DAC that I will fit an XLR into.... The DAC is already in a DIY box so it does not matter but I do have access to punches so I could do a neat job. The input connectors are linked to the PCB with cables (now also Apogee).
My intention was to have an XLR input socket on the DAC, Apogee AES/EBU cable from the socket to the transformer and attenuator which I would locate as close to the 75 ohm input on the PCB as possible so I can have leads of a few mm going from the attenuator pad to the 75 ohm input and keep the differential signal for as long as possible.
Resistors only would make life a lot easier but will it work? The digital input earths (3 electronically selected ones) are all common (but not DC coupled to chassis earth, not sure how).
Will I still have CMNR? There are no transformers on the DAC input at all.
Great to have your input (excuse pun) on this matter as you certainly know your way around this stuff.
The Idea was to have a great big hole in the DAC that I will fit an XLR into.... The DAC is already in a DIY box so it does not matter but I do have access to punches so I could do a neat job. The input connectors are linked to the PCB with cables (now also Apogee).
My intention was to have an XLR input socket on the DAC, Apogee AES/EBU cable from the socket to the transformer and attenuator which I would locate as close to the 75 ohm input on the PCB as possible so I can have leads of a few mm going from the attenuator pad to the 75 ohm input and keep the differential signal for as long as possible.
Resistors only would make life a lot easier but will it work? The digital input earths (3 electronically selected ones) are all common (but not DC coupled to chassis earth, not sure how).
Will I still have CMNR? There are no transformers on the DAC input at all.
CMNR, Can an H-pad fit in a Parakeet coffin?
XLR’s were a great alternative to Amphenal connectors and fanstock clips. I have a jig for soldering rows of them… With any luck it (like me) will keep gathering dust.
Are they still 110 ohms after stuffing the shell full of resistors and heat shrink for an H-Pad? I think I would sooner put a BNC on the transport chassis than an XLR on the DAC.
How bout DB-9 connectors do they maintain impendence?
What ever happened to SDIF-2? I liked life when the clock was out of band. It sounded good and let lots of channels sync up too.
I love the name “wide eye” for the Apogee product. Very clever, almost makes me chuckle. I have a copy of an email to a friend form Bob Clearmountain where he endorses its use for word clock distribution. Of course I understand his wife is one of the company principles. Don’t we know someone else who’s wife is a big player with them?
I think it would be interesting to try a single pair striped from a piece of Cat-5 UTP. I understand that some guys have been trying to use it for data instead of speaker cable.
I seem to recall some protocol differences between AES3 and PSEDIF-2 but cant for the life of me remember if that affects timing. I don’t think I’ll get near a book in the next week to look it up.
Still trying to figure out why I was playing on the Internet at 2:30 am. Thanks for suffering my comments.
XLR’s were a great alternative to Amphenal connectors and fanstock clips. I have a jig for soldering rows of them… With any luck it (like me) will keep gathering dust.
Are they still 110 ohms after stuffing the shell full of resistors and heat shrink for an H-Pad? I think I would sooner put a BNC on the transport chassis than an XLR on the DAC.
How bout DB-9 connectors do they maintain impendence?
What ever happened to SDIF-2? I liked life when the clock was out of band. It sounded good and let lots of channels sync up too.
I love the name “wide eye” for the Apogee product. Very clever, almost makes me chuckle. I have a copy of an email to a friend form Bob Clearmountain where he endorses its use for word clock distribution. Of course I understand his wife is one of the company principles. Don’t we know someone else who’s wife is a big player with them?
I think it would be interesting to try a single pair striped from a piece of Cat-5 UTP. I understand that some guys have been trying to use it for data instead of speaker cable.
I seem to recall some protocol differences between AES3 and PSEDIF-2 but cant for the life of me remember if that affects timing. I don’t think I’ll get near a book in the next week to look it up.
Still trying to figure out why I was playing on the Internet at 2:30 am. Thanks for suffering my comments.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.