Howdy, all --
The data sheet doesn't exactly answer this question. Is it possible for the DIT4192 or DIT4096 to drive both a TOTX-type fiber transmitter and a coax through the usual transformer at the same time? Or is it best to put a gate or a bus switch between the transmitter and the two types of outputs?
Thanks!
Hi, Andy,
We usually switch between these, but let me have someone look into your question.
-d2
-----
Don Dapkus
Audio Applications Engineering Manager
Dallas, TX USA
Andy,
The PCM1808 EVM User's Guide (pg 16) has an example of both outputs being used: http://www.ti.com/lit/ug/sleu078/sleu078.pdf .
Patrick Galvin
Patrick Galvin Andy, The PCM1808 EVM User's Guide (pg 16) has an example of both outputs being used: http://www.ti.com/lit/ug/sleu078/sleu078.pdf .
Oh, I see.. the coax output is taken at TX- and if you look at the transformer schematic, you notice that it inverts the signal polarity. The dotted pin goes to ground and the undotted side of that coil goes to TX-. I notice that the data sheet for the DIT4096 connects both TX+ and TX- to the transformer, and as such signal swing across it requires the 2:1 transformer to attenuate at the output. I assume that the output is basically a high-current complementary pair and using one of the outputs only isn't a problem (otherwise the TOTX driver wouldn't work).
If only TI would choose parts for eval platforms where one can get a data sheet in English! The only data sheet for the DA-02 is in Japanese. It would be nice to know whether it's a 2:1 or 1:1 or whatever transformer. Oh, well, the eval is several years old.
-a
Sorry about the transformer. This EVM was actually designed by the Burr-Brown guys in Japan, hence why they used Japanese components! :-)
Don Dapkus Andy, Sorry about the transformer. This EVM was actually designed by the Burr-Brown guys in Japan, hence why they used Japanese components! :-) -d2
If it was designed by the Burr-Brown guys in Tucson, I could drive over to them by the airport on my lunch hour.
Oh, wait ...
Anyways, seriously, is an English data sheet available? I know I can use a Scientific Conversion or whatever part as a replacement, but it helps to know the turns ratio ...
Too many people knew where the old BB plant was in Tucson. We had to relocate them all to an undisclosed location... Even myself have never been to this new site... I think it might be buried underground in the desert...
Why don't you just use the Scientific Conversion one like show in fig 13, but flip the polarity. The guy who runs Scientific Conversion is a hoot! If you've got a free hour or two, give him a call, and he'll talk you ear off about transformers...
Don Dapkus Andy, Too many people knew where the old BB plant was in Tucson. We had to relocate them all to an undisclosed location... Even myself have never been to this new site... I think it might be buried underground in the desert...
It's actually pretty sad, driving by the old BB plant. Oh, well. I hope the staff are happy in their underground bunker! Summer's coming. Being underground isn't so bad.
Using a Scientific Conversion part is my plan. I'm just not sure which one to choose. The "usual" configuration is to use the TX+ and TX- lines to drive the transformer primary differentially, and in the "use both optical and coax" design, the transformer primary is driven single ended. My assumption is that with the differential drive, you need to attenuate in the transformer to keep the coax amplitude within S/PDIF spec (a volt, I think). In single-ended drive mode, you don't (?) need the attenuation. So maybe it's the 1:1 transformer in the single-ended mode. What's the worst that could happen!
I'm going to get samples of the SC parts. The footprints for 2:1 and 1:! parts are the same, so for first article builds, it's easy enough to swap them out.
Call the dude. He will set you straight. He knows this stuff in and out!
Don Dapkus is on vacation until 4/16 Call the dude. He will set you straight. He knows this stuff in and out!
Will do! I just have to clear out a block of time ...
-A