TTL Converter

[luvdunhill]

I need help putting together a simple TTL converter circuit to raise three DSD / SPDIF3 signals up to TTL level. I'm tapping the lines off a TI DSD1608. Here's what I have so far. Specifically, I was looking at a single TI SN74HCU04D for all three signals. Any recommendations on the cap here? I was thinking all surface mount here, but whatever the consensus is for the caps, I'm fine with. I'm thinking of adding a regulator as well nearby, like a National LP2985 5V regulator, or similar.

Currently, these SPDIF3 lines won't leave the single box player, but perhaps down the line I'd route them to an external box via BNC cables. Either way, I'm assuming a single buffer is okay? Any other topologies that would work well?

Edited January 2, 2010 by luvdunhill

[luvdunhill]

I'd like to add that for the moment, I'm only interested in 3.3V TTL and might include a level shifter on the PCB, but it's not priority at all really... I'd use a mosfet or something simple.

[morphsci]

Well I cannot really you help you here as my knowledge of circuits is only slightly better than my knowledge of quantum physics. I will be watching this thread as I can certainly see a use for this in my future. So all you circuit gurus, get crackin'.

[luvdunhill]

well, I'm already expanded my scope to have both a TTL to coax, and coax to TTL circuits, as well as level shifting. It would be easy enough to also add a toslink connector option. I'm not sure the level shifting is really necessary, so I'm still a bit confused on that point. However, I'll probably add it anyways. I'm also thinking of paralleling more inverters together, perhaps three.

I'll post the circuit tomorrow and see if perhaps that will get me a bit more feedback

[Pars]

DIYHiFi.org • View topic - Recommend me some coax and BNC connectors

Some possibly helpful stuff here. I know I have seen the paralleled 74HC04s before.

[luvdunhill]

Some possibly helpful stuff here. I know I have seen the paralleled 74HC04s before.

cool, I'll add it to my list of links. Yeah, amb did this on the y1, and a certain someone here had a exchange with Ti at headwize and recommended the unbuffered version of the hex inverters

[luvdunhill]

Does anyone know how people usually handle the clock IN/OUT signal that's available on the rear panel of various DACs? I'm curious about how to terminate both ends. I was thinking of treating it just like SPDIF.

[luvdunhill]

Ok, here's where I am at.

Three types of input, transformer coupled coaxial, optical, and TTL. After this input, there is DC blocking cap for the TTL case and then this is fed into a buffer that can be configured to have gain. This performs the optional level shifting from TTL. Then into parallel buffers that drive the output. There is also another DC blocking cap after, that might be used depending on the final configuration. The output can be taken as TTL, transformer coupled coax, or optical. A voltage divider plus series resistor allows for proper termination into a BNC connector.

So, in my case I'll use TTL -> TTL. But this could be used for a coax -> optical, optical -> coax, ttl -> coax, ttl -> optical, etc.

Perhaps now there might be a bit more discussion?

[cetoole]

It looks like you have the TTL input before the amplification stage, though I guess it will just go to the rails and act as a buffer? Also, is it any real benefit to do the bypassing of the DC blocking caps to the extent that you are?

What are you actually doing with TTL signals?

[luvdunhill]

There doesn't necessarily need to be amplification if the feedback resistor is left off, in which case it just acts of a buffer. I guess it's just a bonus that two inverting stages cancel out. The reason I need the amplification is if I want to bring the signal up to TTL level, or go from 3.3v TTL to 5v TTL. (Worse case, I could use the voltage divider at the output to shift from 5v TTL to 3.3v TTL)

The bypassing is so I can have one TH part (Wima MKS02) and/or two NP0 caps. Probably overkill, yes.

So, my problem is that where I'm tapping the DSD1608 chip, the signal isn't very strong. Also, the Wolfson isn't 5.0v compliant, it expects 3.3v logic. So, what I'm assuming is that the DSD signal is basically needs to be 3.3v TTL logic and I either need to buffer the signal at a minimum to drive the signal a couple feet, or buffer and level shift down to 3.3v TTL. Same with PCM I'm assuming (need buffer and/or level shifting to 3.3v).

The bonus of this approach, I could ship the data through a transformer to a BNC connector then out to an external DSD DAC and then use the same little board on the other end as well. So, some flexibility.... but more than anything, it's to experiment I guess and hope I get lucky and it works!

[Dusty Chalk]

So you're redesigning a digital interface? Like Coax or XLR?

Sorry for the mundane question, I guess I'm just trying to point out that you're really NOT trying to redesign a digital interface, you just want to amplify it enough to get from point A to point B.

I don't know if that helps, though.

[luvdunhill]

So you're redesigning a digital interface? Like Coax or XLR?

Sorry for the mundane question, I guess I'm just trying to point out that you're really NOT trying to redesign a digital interface, you just want to amplify it enough to get from point A to point B.

I don't know if that helps, though.

well, galvanic isolation never hurts, right? I may want to route it to an external DAC in the future, hence all the extra stuff.