"Yes, and how many 75 ohm BNC plugs do you see on commercial gear?"
Almost never. The impedance bump that is created by a 50 ohm BNC connecter isn't a problem ... even with HDTV.
If an HDTV signal can make it through a 50 ohm BNC connector, why in the world is anyone worried about passing digital audio through an RCA connector?
Anyone want to measure the actual impedance of the "video" cables which are equipped with RCA connectors on each end?
I can assure you that those cables are nowhere near 75 ohms.
Almost never. The impedance bump that is created by a 50 ohm BNC connecter isn't a problem ... even with HDTV.
If an HDTV signal can make it through a 50 ohm BNC connector, why in the world is anyone worried about passing digital audio through an RCA connector?
Anyone want to measure the actual impedance of the "video" cables which are equipped with RCA connectors on each end?
I can assure you that those cables are nowhere near 75 ohms.
Why???
Because I have demonstrated, on many occasions, that the effects of reflections can be easily heard. Even on some of the worst systems around.
And it does matter. Probably more so on short cables.
You have to stop thinking in the frequency domain, and switch to time domain when it comes to this SPDIF stuff.
Jocko
Because I have demonstrated, on many occasions, that the effects of reflections can be easily heard. Even on some of the worst systems around.
And it does matter. Probably more so on short cables.
You have to stop thinking in the frequency domain, and switch to time domain when it comes to this SPDIF stuff.
Jocko
Video gear! Reflections cause echos which cause ghosts, which can get very visible.carlosfm said:
Yes, and how many 75 ohm BNC plugs do you see on commercial gear?
I worked in a cable company head end for a while, everything that wasn't an F connector was a 75 ohm BNC connector. And the monitor I'm typing this post on has five 75-ohm BNC jacks on the back.
75 ohm BNC's are used extensively in professional video applications--- and to some extent in cable TV , although other connectors like F's etc are more common..
A modern digital video studio is built with component bandwidths of 150 MHz ++, - and correct cabling DOES indeed matter...
Composit video inputs are bandwidth limited to 3.5-4
Mhz because of the colour carrier,- and you hardly see any diffference at all on domestic equipment
A modern digital video studio is built with component bandwidths of 150 MHz ++, - and correct cabling DOES indeed matter...
Composit video inputs are bandwidth limited to 3.5-4
Mhz because of the colour carrier,- and you hardly see any diffference at all on domestic equipment
Mikett,...
I just don't know if you really will gain anything but good beliefs...
Personally I think much of this cable-connector debates are just another way of fighting the demons in ones head...
The very best part of the arguments still has no basis in solid physics or electronic theory, and as a professional engineer, that always makes me very sceptical, - but alas-- that is also the demon fighters main argument, that we are "too limited in our minds" to undedrstand all the sensational differernces......
As for the Phono-BNC thing... a similar analogy can be found with the socalled UHF connector, common on a myriad of amateur radio and professional short wave radios... and completely useless on UHF freq's....
As the phono, this is also a connector with "non-specific" impedance, based on its construction.....but still quite OK for short wave, up to 30-50 MHz....
My personal guess, without testing on vector analysers or reflectometers, is that the phono is reasonably OK for bandwiths up to severeal tens of MHz....
If your going high speed in digital domain, trying to maintain minimum phase lag, - that's quite another story......
I seriously don't think you will gain very much, as long as you use good quality coax and connectors,--
and they don't have to cost a fortune....
I just don't know if you really will gain anything but good beliefs...
Personally I think much of this cable-connector debates are just another way of fighting the demons in ones head...
The very best part of the arguments still has no basis in solid physics or electronic theory, and as a professional engineer, that always makes me very sceptical, - but alas-- that is also the demon fighters main argument, that we are "too limited in our minds" to undedrstand all the sensational differernces......
As for the Phono-BNC thing... a similar analogy can be found with the socalled UHF connector, common on a myriad of amateur radio and professional short wave radios... and completely useless on UHF freq's....
As the phono, this is also a connector with "non-specific" impedance, based on its construction.....but still quite OK for short wave, up to 30-50 MHz....
My personal guess, without testing on vector analysers or reflectometers, is that the phono is reasonably OK for bandwiths up to severeal tens of MHz....
If your going high speed in digital domain, trying to maintain minimum phase lag, - that's quite another story......
I seriously don't think you will gain very much, as long as you use good quality coax and connectors,--
and they don't have to cost a fortune....
Frank Berry said:
Hi
You can't compare these 2 worlds. They are totally different, both in terms of signal architecture , as well as when it comes to sensitivity of perception. Consider that not only data is transfered in SPFDIF, but also clock is embedded. Messing that up becomes audible very rapidly.
Jocko is right, impedance matching in an SPDIF system is very critical.
I've got one question to all of you that are so concerned about the impedance of the RCA connector. Why do you need exactly 75 Ohms connector and what do you expect if not 75 but let's say 50 or 30 or 100 Ohms? What I mean is what would you measure in all the impedance cases I mentioned above and how this would impact the SPDIF data transfer.
Any impedance mismatch causes a reflection in the signal, much the same a multipath in RF transmission. If you ran a sweep through the system you'd find lumps and dips in the response. When those get big enough you'll start losing bits. Remember that ALL digital signal transmission is an _analog_ signal representing those bits.
G²
Once upon a time RCA connectors were developed as a low cost RF connector/ you could see them used in TV tuners and some B&W video cameras. BTW baseband video is 75 ohms. Later they also got adapted for audio too, mostly at line level to some speaker interconnections. OEMs changed them mechanically from the original drawings. Now only consumer grade audio is stuck with them. The End
True, but those RCA connectors (I have them in my old CRT TV, where antenna signal is sent to two tuners) have very small shied area... and PCB-material-like insulator between centre pin and shield... so it seems someone new about impedance matching issues... they are certainly nothing like audio RCA connectors.
Critical listening tests proved that BNC’s perform better in S/PDIF signal path compared to audio RCA connectors. This was not only my impression, but everybody's who auditioned the demo. The most noticeable was low frequency definition improvement with BNC’s, and better, more precise instrument / voice placement.
Boky
Critical listening tests proved that BNC’s perform better in S/PDIF signal path compared to audio RCA connectors. This was not only my impression, but everybody's who auditioned the demo. The most noticeable was low frequency definition improvement with BNC’s, and better, more precise instrument / voice placement.
Boky
I agree with you, but I asked two times in my post: what would you measure? If the cable is impedance mismatched the reflections would be measureable. If the RCA impedance is mismatched I'm very suspicious you'd find any reflections because the RCA length is negligible compared to the overall transmition line lenght. This I recall from my RF practical and experimental experience where the freq range is in the VHF and UHF bands, i.e. more critical and far from S/PDIF freq. range.
That's not how it works. The impedance mismatch causes a reflection back down the line to the source where some will be reflected forward, particularly if there is a mismatch there too. It's not the physical length of the connector but the distance from the other end(s). So, the delay is based on the cable length, not the connector length.
As far as sound differences on SPDIF, I find it very hard to believe. There are errors or there aren't. A bit error, when it occurs, is as likely to be an MSB as an LSB. We might not notice an LSB error but EVERYONE would notice an MSB error.
G²
I think a misunderstanding is taking place here. The impedance is really defined by the space between the wires or more correctly - the ratio between the diameters of the outer and the inner parts of the wire/connector/etc. What I mean is what I've seen and measured by the means of inline power meter - when there is a small mismatch i.e. not 75 but 50 Ohms connector that is much less long than the 75 coax cable, at the beginning of the wire you wan't see big magnitude of reflected wave if you see it at all.
A small (~3dB) in line attenuator would be very effective in stopping the reflected wave
But I agree with beauty_divine that the impedance error from the connector is only going to become significant at UHF requencies. I suspect that the SPDIF transformers will have a poor return loss even at VHF
But I agree with beauty_divine that the impedance error from the connector is only going to become significant at UHF requencies. I suspect that the SPDIF transformers will have a poor return loss even at VHF
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.