RG6 cable
Hi,
I have a look about RG6 cable. I'm not able to find here a RG6 cable which is 4,5 Ghz proof for the moment (see Bappe post about the nine harmonics needed in a good EBU signal = around 4 Ghz with a 6M hz fundamental).
You can find easily some ready made BNC cable wich climb to 3 Ghz max which is ok for 8 harmonics, but if the nine harmonics is needed for a better quality (why not to do the best if we can) it's not enough.
There are many specifications for a RG6 cable : RG-6 has many specifications: F6TSSV, F6TSSVcu, F6TVS, F6TVScu, F660BV, F660BVcu, F660BVF,F660BVM, F690BV, F690BVcu, F690BVF, F690BVM maximum range
My understanding is only the last F690BVM maximum range is 4,5 Ghz proof : very low loss ! Don't find any quad shelded (>= 120 db) as Bappe advised above (surmise its very usefull to protect against wifi and cellulars in our environment!) in this F690BVM specification. Maybe for Satellite TV in professional environment ?!
like for example the famous Belden 1694A: 1694A Belden 4.5GHz RG6/U Precision Video Cable for Analog and Digital Applications | tselectronic.com. But not quad shelded
I already save money with buying only 100 nf 0605 smd NPO (thanks Bappe) as DC stopper ( coils seem expensive and hasardous for enthusiasts Diyers like I am)
Am i too fanatic to want the best as it's far to be the most expensive device in our Hifi systems ?
What do you use fellows here in Europe, have you an Internet source for such high quality cable (ready made with BNC or not) ? At Mouser I find only ready made 3 Ghz max BNC cable...
Firewire would be better but it diseapper in public laptop to be more and more seen only in professionnal products. What will we do tommorow when we need more than 192 khz for audio high quality products ? Come back to one box source-DAC ?
thank you
Hi,
I have a look about RG6 cable. I'm not able to find here a RG6 cable which is 4,5 Ghz proof for the moment (see Bappe post about the nine harmonics needed in a good EBU signal = around 4 Ghz with a 6M hz fundamental).
You can find easily some ready made BNC cable wich climb to 3 Ghz max which is ok for 8 harmonics, but if the nine harmonics is needed for a better quality (why not to do the best if we can) it's not enough.
There are many specifications for a RG6 cable : RG-6 has many specifications: F6TSSV, F6TSSVcu, F6TVS, F6TVScu, F660BV, F660BVcu, F660BVF,F660BVM, F690BV, F690BVcu, F690BVF, F690BVM maximum range
My understanding is only the last F690BVM maximum range is 4,5 Ghz proof : very low loss ! Don't find any quad shelded (>= 120 db) as Bappe advised above (surmise its very usefull to protect against wifi and cellulars in our environment!) in this F690BVM specification. Maybe for Satellite TV in professional environment ?!
like for example the famous Belden 1694A: 1694A Belden 4.5GHz RG6/U Precision Video Cable for Analog and Digital Applications | tselectronic.com. But not quad shelded
I already save money with buying only 100 nf 0605 smd NPO (thanks Bappe) as DC stopper ( coils seem expensive and hasardous for enthusiasts Diyers like I am)
Am i too fanatic to want the best as it's far to be the most expensive device in our Hifi systems ?
What do you use fellows here in Europe, have you an Internet source for such high quality cable (ready made with BNC or not) ? At Mouser I find only ready made 3 Ghz max BNC cable...Firewire would be better but it diseapper in public laptop to be more and more seen only in professionnal products. What will we do tommorow when we need more than 192 khz for audio high quality products ? Come back to one box source-DAC ?
thank you
Last edited:
See post 34 which is an explanation not just an assertion.
My understanding is with 6 Mhz, the nine fundamental is 4 Ghz as I surmise you perfectly understood. RG6 seem to be video cable as well ?
I don't know... explain us why it seems so ridiculous for you ! At least you will write for others as well !
My understanding is with 6 Mhz, the nine fundamental is 4 Ghz as I surmise you perfectly understood. RG6 seem to be video cable as well ?
I don't know... explain us why it seems so ridiculous for you ! At least you will write for others as well !
This is only relevant in SMPS going into burst mode, and not relevant for COG capacitors, that are pretty stable, you are very unlikely to excite these capacitors with SPDIF to create noise.
MLCC SMD ceramic capacitors X7R, COG etc. are used in mil automotive and other situations where the temperature ranges met are more severe than in commercial and especially domestic products, again this will not be a problem.
Generally the highest frequency of interest is based on the knee frequency, calculated from the signal rise time. This rise time will depend also on the chosen device driving the line.
http://www.develeast.ch/downloads/TN001_RiseTimeAndKneeFrequency_v01.pdf
http://www.develeast.ch/downloads/TN001_RiseTimeAndKneeFrequency_v01.pdf
yeap, I'm unable to explain to you why I take the calcul of the next octave to calcul here harmonics.... I think wine is really the better explanation.
I really have to beginn the water ! In fact for medical reason I had to stop water ten years ago because I maid too much Fe ! ( don't you think that my medical knowledge is the same than my electronic knowledge ?).
Some sims of SPDIF from another interesting discussion. I have some from a few years ago if I can find them discussing the silly suggestion of a longer cable. The shorter any signal travels the better, and specifying a certain cable length without other parameters such as its velocity factor is silly.
http://www.diyaudio.com/forums/digi...-wave-spdif-cable-possible-9.html#post3819255
I am on site (working with acoustic as well as electronics engineers) so don't have any time to do any more sims at the moment, I would like to sim a full interface from transmitter to receiver (again).
http://www.diyaudio.com/forums/digi...-wave-spdif-cable-possible-9.html#post3819255
I am on site (working with acoustic as well as electronics engineers) so don't have any time to do any more sims at the moment, I would like to sim a full interface from transmitter to receiver (again).
A really long cable might help if the terminations are really bad - the cable attenuation will reduce the effect of reflections. However, boring silly things like ensuring good terminations is a much better of solving that problem. Any normal commercial equipment will have sufficiently good terminations, only DIY and possibly some 'high-end' kit will have poor terminations. As a general rule, people only play with cables if that is the only thing they can play with: they invent 'problems' and then 'solve' them.
I agree.
The problem with a longer cable is that it will add other problems, such as increased jitter due to possible increased ISI etc, but would I worry at these speeds; my view is a short cable and proper termination as you say. The other thing is how much influence are the reflections going to have on the received data, nil, unless they are really really bad.
The SPDIF interface is a quite slow Manchester encoded data transfer mechanism, playing about with simulations in the past convince me that it should be quite simple to engineer a correct interface, even a basic 75R resistor.
The problem with a longer cable is that it will add other problems, such as increased jitter due to possible increased ISI etc, but would I worry at these speeds; my view is a short cable and proper termination as you say. The other thing is how much influence are the reflections going to have on the received data, nil, unless they are really really bad.
The SPDIF interface is a quite slow Manchester encoded data transfer mechanism, playing about with simulations in the past convince me that it should be quite simple to engineer a correct interface, even a basic 75R resistor.
I saw the sims : yeap : 75 ohms connector is cleaner than the 50 ohms (Did you hear a difference as Julf Suggested ? Don't understand if it was a serious question or not)
Good news :
I do not find a quaded shielded ready made RG6 cable but found the excelent Belden 1794a on Mouser or Digikey with true 75 ohms BNC connectors at an acceptable price ! Its for placebo as I understand now than the capacitance with a lower bandwidth RG6 will give the same result
I have to buy two female plugs for DACS and squeezebox... I surmise the 0.5 cm between th eplug and the pcb doesn't matter ? It will be a simple 0.5 mm silvered copper for the hot (non coaxial).
Just for theory :You talk about length : did the impedance matter if the distance is very short as 10 cm? If soldered point to point on the two pcbs without any connectors: will we hear a difference this time between these shielded coax:
10 cm 50 ohms wire
10 cm 75 ohms wire
10 cm 110 ohms wire
10 cm general purpose silver wire with a diameter of 0.5 mm2 but non shielded ? (GND connector : the same)
Does a shunt 75 ohms resistor at the receiver side will give the same result in a later test with the same cable as just above ?
On the scope I understand it is visible, are these few ohms on short distances listenable ? I suppose hot//gnd on the receiver chip pin is not exactly 75 ohms?
does the engineer for general audio equipment match the input/output of the digital devices they design for working with non 75 ohms RCA plugs and non really true 75 ohms spidf wire (because non RCA 75 ohms plugs for most of the vendors...) ? Or they just don't care with basic devices under xxxxx dollars ?
yesterday I tried 2 different 75 ohms cable on RCA plugs (25 ohms) with screw for the hot point (bad contact I surmise for these frequencies... a 450° silver solder will be better here maybe...). Same 20 cm length with a swap with the same rca connectors : sound was different. The generic TV with one hot wire was better than the double shelded low loss (54 pf/km) cable with multi wires hot point.
the source is a squeezebox Duet (light rca plug, and when switched off: 115 ohms measured between hot and gnd on it !).
just in front of the Philips spidf buffer emmiter : have 216 ohms shunter with 118 ohms (well = to 75 ohms as writed after in the thread).
http://www.diyaudio.com/forums/digital-line-level/248695-squeezebox-duett-sympatic-caps-tip-4.html
For experiences : the DAC is the Subbu with an rca plug and short 2cm silver wire (0.5 mm diam...) with very thin PFTE insulation.
In suppose this time in the real world with two different wires of 75 ohms and without connector no differences will be hearable between those two cables and/or different lengths (10 cm or 1m...) if cable is soldered slaved on the pcbs...
next try (have to buy it) will be with a newada pulse transformer at less 10$ as it seems important for the bad squeezebox digital output (don't want to go with tolinsk as I have no DACs with it). But Damned, you serious people don't maid trace for such smd pulse transformer on your PCB
surmise it has to be shielded if near on the W8804 spidf input... and near the dac chip ?Hé digital is fun !
have to read the entire squre wave digital spidf thread....
Last edited:
The shorter the cable and the slower the rise time the less does the impedance matter. As the cable will normally be significantly longer than the connectors, the cable impedance matters much more than the connector impedance. Digital audio signals are quite robust, which is why crazy DIY cables or interface 'improvements' don't do much damage.
Most changes thought to be heard are not real. Those which are real will be due to one of the following:
1. audio grounding differences, due to the cable screen resistance
2. particularly poor cables can corrupt the signal or add lots of jitter (perhaps from RF interference) - some may prefer this sound, by confusing interference for 'detail'
Most changes thought to be heard are not real. Those which are real will be due to one of the following:
1. audio grounding differences, due to the cable screen resistance
2. particularly poor cables can corrupt the signal or add lots of jitter (perhaps from RF interference) - some may prefer this sound, by confusing interference for 'detail'
3. using to fast a rise time to get a nice 'square wave'. Some erroneously think that a really square wave (1ns or faster rise time) is better because the wave looks more square. This then puts to much high frequency components into an unmatched, incorrectly terminated line with many impedance mismatches (PCB trace, connector, cable, connector, PCB trace etc.) leading to reflections and other signal integrity problems, the most important to audio is increased jitter.
Thank you,
Oh yes, the point 2. I notice it this week with the very poor cable of my post above (the 75 ohms one with multiple wires for the hot and double insulation) give harsch sound in the highs and give real physic pain to the ears. Try both less than 20 cm and 1 meter : the same. Direct to the garbadge !
Without equipment to measure, I surmise the point 1. will occur by chance for a basic diyer. I suppose than the impedance of the shielding has to be lower than the hot wire. The more shields, the more you have surface and material = the lower the impedance ?
Anyway I undersand than the multiple layers for shielding are here to fight against Wifi, satelite TV, GSM devices.
Suppose as well than material of the shields/Gnd wire(s) could be important to reduce its capacitance (silvered cooper ?) ? Is dielectric insulation absorbtion is important here for the gnd wire ? Saw than Mogami use sometimes a carbon layer to fight against it... for some 50 ohms cable for reccording devices in analog domain like mics for reccording, but i don't know if Mogami exist for 75 ohms (I saw just a very simple one simple shielded with 51 pF/m. It's true most of the time the pros in studio use big length so balanced wires ! hummm I mix too much things here and it's not important here for normal diyer (non fanatic) as good cable already exist. curiosity is a hole without end !
Well by now I will use for real life 75 ohms plugs on my digital devices and very short cables with true 75 ohms plugs. ready made cable can be found with short length on Mouser with the same Belden 1794a < to 50 cm for example.
If hand made, somebody told me the better is to solder with good 4% silver and very hot >450° point of fusion. Better than the mechanical encapsulation between the wire and the plug without soldering ?! I keep into consideration : the more shielding layers, the better... if possible to find it (find also Sommer cable sold by meter in this part of the world here: Câbles Numérique Audio & Vidéo - Audiophonics - Câbles, Amplificateurs et Accessoires DIY ). impossible here in Europe to have for simple diyers good brand like Belden America or Mitsubishi Japan with length < 152 meters ! Most of cables in Europe are maid into licence in the low cost Romania but special exotic ones for aeronautic and industries.
Only for pleasure and fun this time, I will test very short cable and non 75 ohms and non coax. As I understand than the hot point diameter is 0.5 mm2, i will try more and less without insulation but air for the hot point (with the insulated gnd wire next to it... I talk about distance below 5 or 10 cm...). The closer pcbs, the better (but EMC compatibility problems?!).
At the end an aluminium or Fe Box will be usefull to protect from the outside world my digital DIYs... i used before wood for practical and economic reasons.
You pro of Digital have a hard & difficult job (I mean viewed from my chair !).... need many knowledge, many equipments for measures !
Thank you again to all for sharing and your numerous inputs above in relation to my exotic questions. Hop ... I swap to practice mode...
Oh yes, the point 2. I notice it this week with the very poor cable of my post above (the 75 ohms one with multiple wires for the hot and double insulation) give harsch sound in the highs and give real physic pain to the ears. Try both less than 20 cm and 1 meter : the same. Direct to the garbadge !
Without equipment to measure, I surmise the point 1. will occur by chance for a basic diyer. I suppose than the impedance of the shielding has to be lower than the hot wire. The more shields, the more you have surface and material = the lower the impedance ?
Anyway I undersand than the multiple layers for shielding are here to fight against Wifi, satelite TV, GSM devices.
Suppose as well than material of the shields/Gnd wire(s) could be important to reduce its capacitance (silvered cooper ?) ? Is dielectric insulation absorbtion is important here for the gnd wire ? Saw than Mogami use sometimes a carbon layer to fight against it... for some 50 ohms cable for reccording devices in analog domain like mics for reccording, but i don't know if Mogami exist for 75 ohms (I saw just a very simple one simple shielded with 51 pF/m. It's true most of the time the pros in studio use big length so balanced wires ! hummm I mix too much things here and it's not important here for normal diyer (non fanatic) as good cable already exist. curiosity is a hole without end !
Well by now I will use for real life 75 ohms plugs on my digital devices and very short cables with true 75 ohms plugs. ready made cable can be found with short length on Mouser with the same Belden 1794a < to 50 cm for example.
If hand made, somebody told me the better is to solder with good 4% silver and very hot >450° point of fusion. Better than the mechanical encapsulation between the wire and the plug without soldering ?! I keep into consideration : the more shielding layers, the better... if possible to find it (find also Sommer cable sold by meter in this part of the world here: Câbles Numérique Audio & Vidéo - Audiophonics - Câbles, Amplificateurs et Accessoires DIY ). impossible here in Europe to have for simple diyers good brand like Belden America or Mitsubishi Japan with length < 152 meters ! Most of cables in Europe are maid into licence in the low cost Romania but special exotic ones for aeronautic and industries.
Only for pleasure and fun this time, I will test very short cable and non 75 ohms and non coax. As I understand than the hot point diameter is 0.5 mm2, i will try more and less without insulation but air for the hot point (with the insulated gnd wire next to it... I talk about distance below 5 or 10 cm...). The closer pcbs, the better (but EMC compatibility problems?!).
At the end an aluminium or Fe Box will be usefull to protect from the outside world my digital DIYs... i used before wood for practical and economic reasons.
You pro of Digital have a hard & difficult job (I mean viewed from my chair !).... need many knowledge, many equipments for measures !
Thank you again to all for sharing and your numerous inputs above in relation to my exotic questions. Hop ... I swap to practice mode...
As we keep saying, all you need for digital audio is a common-or-garden basic commercial quality 75 ohm coax cable. For very short runs you might not even need that and can get away with any old bit of wire, such as a DIY cable or a 'high-end' cable. No need for multiple screens, multiple inners, silver conductors, silver solder, fancy connectors etc. etc. etc.
I know this is boring and disappointing for those who like to play with cables; the truth often is boring!
I know this is boring and disappointing for those who like to play with cables; the truth often is boring!
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.