S/PDIF Digital output

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Peter Daniel said:
Here how it's implemented in one of the very expensive product from a famous audio manufacturer. Somebody mentioned that: "Jocko will be amused about it.. "


Do NOT use this as an example of a good design. It is never good to drive a transmission-line with a Flip-flop. Even junior digital designers know this much....

Steve N.
Empirical Audio
Manufacturer
 
carlosfm said:


I'm installing clocks on cdps and transports since 1999.The Tent XO3 is a unique thing (as features go) and it has an amazing price/performance ratio.
It is better than any LC-Audio clock, and I tried' em all.
IMHO.

Have you tried Superclock2 or 3 from Audiocom international in the UK?

I have been modding for 3-4 years myself and I like these, although I further mod them.

Steve N.
 
I was looking at just the Pi pad. Wich adapts the 100R to the 75R of the connector?

75R at the output // 137R +26,1R + 26,1R = 100,667R at the trafo, Seems right?

But seen from the other direction

100R at the trafo +26,1R + 26,1R // 137 = 72,1R at the output.

Wich seems a little low?
 
audioengr said:



1) because the internal feedback in a Flip-Flop is more sensitive to transmission-line reflections that can change it's state
2) because the drive strength of a Flip-flop is usually quite low.
3) It is not its intended use - use a driver to drive a transmission-line

Steve N.
Empirical Audio


Hi Steve

1) Can be easlly solved by taking care of proper (series) matching. I've never seen that phenomena by the way.
2) Is not an issue when using it as a driver for SPDIF (transformer coupled)
3) The backend of a flipflop consists of a quasi PP output stage. At the risk of discussing definitions, I'd call that a driver.

best
 
Guido Tent said:



Hi Steve

1) Can be easlly solved by taking care of proper (series) matching. I've never seen that phenomena by the way.
2) Is not an issue when using it as a driver for SPDIF (transformer coupled)
3) The backend of a flipflop consists of a quasi PP output stage. At the risk of discussing definitions, I'd call that a driver.

best


Suit yourself. You can design it anyway that you like.

Steve N.
 
audioengr said:
Have you tried Superclock2 or 3 from Audiocom international in the UK?

I have been modding for 3-4 years myself and I like these, although I further mod them.

Steve N.

When I bought my first clock back in 1999 the Audiocom (and also Trichord) was more expensive than the LC-Audio, and I opted for the later.
These were the days when you sent an e-mail and waited for a week (if you were lucky) for an answer, when somebody remembered to check the mailbox of the company.
No, I never tried the Audiocom clocks, I kept with LC-Audio until some years ago I tried the Tent clocks and never looked back.
LC-Audio is horribly expensive these days, and inferior to the Tent clocks IMO.
And it's a pleasure to buy from Guido, I can't say the same about LC-Audio.
 
Katapum's RX input not right

I was just looking at Katapum's posting on the RX input and the impedance looking from the bnc connector is not 75 ohms. You will get better impedance match if you get rid of the 100ohm on the primary side, replace the 75ohm with 150ohm, and replace the 300ohm on the secondary with 150ohm. The zobel cap is around 22pf if the transformer leakage is ~0.5uH.

Experts, please comment if the calculations are right.

Thanks,
rlim
 

Attachments

  • digin2_3_jh_np.jpg
    digin2_3_jh_np.jpg
    20.4 KB · Views: 1,986
Re: Katapum's RX input not right

rlim said:
I was just looking at Katapum's posting on the RX input and the impedance looking from the bnc connector is not 75 ohms. You will get better impedance match if you get rid of the 100ohm on the primary side, replace the 75ohm with 150ohm, and replace the 300ohm on the secondary with 150ohm. The zobel cap is around 22pf if the transformer leakage is ~0.5uH.

Experts, please comment if the calculations are right.

Thanks,
rlim

this looks quite similar tom what we did a while ago

http://www.tentlabs.com/Products/DIYDAC/DIYDAC.html

I'd suggest a screen in the transformer to increase common mode rejection

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