The manufacturers page, can be translated quite well. The part mentioned before is only the output part of some clock generator.
It is a copy of some european high end design.
http://www.cdream5.com/htm/EX_ClockV5.htm
It is a copy of some european high end design.
http://www.cdream5.com/htm/EX_ClockV5.htm
Not DIY (irrelevant?) and no integrated BNC connectors but when it is about price and required time & effort it seems a winner. It has these:
https://www.ti.com/lit/ds/symlink/ds8921.pdf?ts=1744194419977
https://www.ti.com/lit/ds/symlink/ds8921.pdf?ts=1744194419977
Last edited:
Hi friends,
this looks like a real alternative, actually it's available from Audiophonics as well, which I prefer 😉. It does not have to be DIY, I'm pragmatic - but questionmarks arose:
1. the SPDIF input is specified with 0.5V instead of TTL
2. the second coax output is specified as AES @ 0.5V / 110Ohm
3. no second Toslink (but a 5V SPDIF output is present)
So, as it looks, we could "get there" with a few DIY modifications for my needs:
1. Look @ board layout and "intercept" at the right point (after input transformer?) to supply the TTL level TXP input signal there
2. Change the AES circuit to match the 75Ohm output to produce a second 75Ohm output
3. Use the TTL SPDIF Output with additional Toslink transmitter on the back panel
4. Move the on-board Toslink Transmitter to the back-panel location (just a few cm)
5. Connect the BNC Sockets on the system back panel with 75Ohm coax cables.
6. Optional: Remove transformers from the target system inputs.
So, all in all, this is an extremely attractive alternative to building an own board as it can do everything I need/want!
One question, though: On the picture the marked arrow for Input points outward and the arrow for Output points inward. Puzzling...
Thanks a lot for highlighting!
Regards,
Winfried
this looks like a real alternative, actually it's available from Audiophonics as well, which I prefer 😉. It does not have to be DIY, I'm pragmatic - but questionmarks arose:
1. the SPDIF input is specified with 0.5V instead of TTL
2. the second coax output is specified as AES @ 0.5V / 110Ohm
3. no second Toslink (but a 5V SPDIF output is present)
So, as it looks, we could "get there" with a few DIY modifications for my needs:
1. Look @ board layout and "intercept" at the right point (after input transformer?) to supply the TTL level TXP input signal there
2. Change the AES circuit to match the 75Ohm output to produce a second 75Ohm output
3. Use the TTL SPDIF Output with additional Toslink transmitter on the back panel
4. Move the on-board Toslink Transmitter to the back-panel location (just a few cm)
5. Connect the BNC Sockets on the system back panel with 75Ohm coax cables.
6. Optional: Remove transformers from the target system inputs.
So, all in all, this is an extremely attractive alternative to building an own board as it can do everything I need/want!
One question, though: On the picture the marked arrow for Input points outward and the arrow for Output points inward. Puzzling...
Thanks a lot for highlighting!
Regards,
Winfried
Your own design could be way smaller and more adapted to your needs but then hurry as in may things start to be expensive it seems.
As I wrote, the instructions are very basic. Anyway, as the thread starter mentioned, it can be modified quite easily. The basic, important stages are identical, so you can change it to have two identical outputs with just minor changes.
It has got two output transformers which can help with ground loops.
I wanted to DIYS a digital output first, but realized that for less than 20€ I was not able to even get a fraction of the components it uses.
The output signals are surprisingly clean, even feed by a not so good looking input signal.
You could make it even more simple, get two of them from Aliex and make a stack, still less than 25$.
It has got two output transformers which can help with ground loops.
I wanted to DIYS a digital output first, but realized that for less than 20€ I was not able to even get a fraction of the components it uses.
The output signals are surprisingly clean, even feed by a not so good looking input signal.
You could make it even more simple, get two of them from Aliex and make a stack, still less than 25$.
The output arrow could indicate the signal comes out of that solder pad, or it could indicate the signal comes out of the whole board. Thus the arrows can be interpreted ambiguously.
Similarly, for the inputs.
Thank you all for the hints and help!
I have decided to go with the board and DIY mod it, once I have it working.
The six board work items will be extended by:
7. add a small, dedicated 5V low noise regulator to the board, because a suitable supply rail (with regulator headroom) is available within 5-7cm (2-3inches) of the board
As @Turbowatch2 seems to have experience with the board, I'll go with his assessment, but will first use the coax input to test with my scope where the signal comes out again...
Edit: The jumper soldering table makes it clear:
Things look promising, the order goes out!
Regards,
Winfried
I have decided to go with the board and DIY mod it, once I have it working.
The six board work items will be extended by:
7. add a small, dedicated 5V low noise regulator to the board, because a suitable supply rail (with regulator headroom) is available within 5-7cm (2-3inches) of the board
Well, I was not planning on a dedicated board layout design, but an experimentation board build, so the board is a much cleaner solution and: There's PLENTY of room/air in an HDP-3 😉
As @Turbowatch2 seems to have experience with the board, I'll go with his assessment, but will first use the coax input to test with my scope where the signal comes out again...
Edit: The jumper soldering table makes it clear:
Things look promising, the order goes out!
Regards,
Winfried
Last edited:
OK. I have recieved the board and have a couple of findings after analysing parts of the layout and want to share them in case somebody plans to use the board:
1. Coax and the AES3, are specified with 0.5V
2. Both DS8921M output drive circuits use identical components (and connections) driving the respective transformer primary windings
3. On the transformers' secondary sides the only difference between AES3 and Coax is the (expectedly) grounded pin on Coax side
Therefore the only things I have to modify to get "my" two Coax SPDIF Outs are:
1. ground "AES -" connection of the AES3 output
2. Close JP1 1-2 (and open 2-3) to activate the J6 TTL input (which disconnects the coax SPDIF input circuitry altogether)
3. Connect J6 Signal with CS8420 TXN and J6 GND with CS8420 DGND
4. Run coax cables from the output connectors to the BNC sockets mounted on the HDP-3 back panel
To get two TOSLINK Outs I will:
1. Remove the OPT1 connector
2. Use the TTL output connections for optical connectors mounted on the HDP-3 back panel
I'm really excited!
Winfried
1. Coax and the AES3, are specified with 0.5V
2. Both DS8921M output drive circuits use identical components (and connections) driving the respective transformer primary windings
3. On the transformers' secondary sides the only difference between AES3 and Coax is the (expectedly) grounded pin on Coax side
Therefore the only things I have to modify to get "my" two Coax SPDIF Outs are:
1. ground "AES -" connection of the AES3 output
2. Close JP1 1-2 (and open 2-3) to activate the J6 TTL input (which disconnects the coax SPDIF input circuitry altogether)
3. Connect J6 Signal with CS8420 TXN and J6 GND with CS8420 DGND
4. Run coax cables from the output connectors to the BNC sockets mounted on the HDP-3 back panel
To get two TOSLINK Outs I will:
1. Remove the OPT1 connector
2. Use the TTL output connections for optical connectors mounted on the HDP-3 back panel
I'm really excited!
Winfried
Hi folks,
in the mean time I have analysed the transformer outputs of the Toslink/SPDIF/AES Board. Now look at this:
The only difference between the koax and the AES output is that AES has a "-" pad and a GND pad (as expected), but drivers (U2 and U3) have identical(!) components in their output circuitry! So, both will have the same (specified) output voltage swing of 0.5V.
In order to have two BNC koax outputs I should therefore only have to solder the "AES -" to "GND" and connect a cable to the BNC. Am I not missing something? It somehow looks too simple ...
Regards,
Winfried
in the mean time I have analysed the transformer outputs of the Toslink/SPDIF/AES Board. Now look at this:
The only difference between the koax and the AES output is that AES has a "-" pad and a GND pad (as expected), but drivers (U2 and U3) have identical(!) components in their output circuitry! So, both will have the same (specified) output voltage swing of 0.5V.
In order to have two BNC koax outputs I should therefore only have to solder the "AES -" to "GND" and connect a cable to the BNC. Am I not missing something? It somehow looks too simple ...
Regards,
Winfried
If I remeber right, it is that simple. Best is to power the board, connect a CD players digital out and have it at idle. You should see a very clean sqare wave at the output. As the fake (probably) pulse converters are identical, you get two identical, galvanically isolated outputs.
It puzzled me that the output signals were identical and not what the board said. Which made things simpler in the end.
Keep in mind, you got a quite low frequency, digital signal, no sensitive analog audio that may be disturbed by any not so Hiend resistor or piece of wire. As some may hear, I may add.
It puzzled me that the output signals were identical and not what the board said. Which made things simpler in the end.
Keep in mind, you got a quite low frequency, digital signal, no sensitive analog audio that may be disturbed by any not so Hiend resistor or piece of wire. As some may hear, I may add.
Last edited: