English is not my native language, somehow one can read things I did not intend to write lol. By electrical I only meant coaxial spdiff, not optical Toslink spdiff. Those optical input/outputs are also everywhere I look, I dont know where those are used for and who uses them? Some JApanese brands even had exclusively Toslink spdiff. My Primare has them too and Cambridge Audio that sugested they had thought about digitall output using BNC on some models, uses digital RCA and Toslink too, so the BNC connector could just be marketing too, dont you think?
The assumption I read here is that if you can't hear a difference between digital RCA interconnects, the spdiff circuitry is OK and you are mentally sane, and if you can hear a difference something must be wrong, ill designed or imagined. All interconnects I used are shorter than the ft specified above, much shorter, making impedance pretty irrelevant if I read the above correctly. Well my father did the interconnect switching today and I could tell blindly which was the cable I used by low frequencies again, and so did he, he btw thought the vandenhull had more highs, I wouldnt know, because I only focussed on lows jeje. So back at ill designed input/output and differences that should not be there with 50-100 cm coaxial interconnects but are. I have noticed black square pulse transformers both on primare output and on dacamp input, does that make sense??
The assumption I read here is that if you can't hear a difference between digital RCA interconnects, the spdiff circuitry is OK and you are mentally sane, and if you can hear a difference something must be wrong, ill designed or imagined. All interconnects I used are shorter than the ft specified above, much shorter, making impedance pretty irrelevant if I read the above correctly. Well my father did the interconnect switching today and I could tell blindly which was the cable I used by low frequencies again, and so did he, he btw thought the vandenhull had more highs, I wouldnt know, because I only focussed on lows jeje. So back at ill designed input/output and differences that should not be there with 50-100 cm coaxial interconnects but are. I have noticed black square pulse transformers both on primare output and on dacamp input, does that make sense??
For a non-critical purpose like SPDIF you can get away with using 50 ohm BNC with 75 ohm cable.TheGimp said:
Given the signal was due to be converted internally to digital anyway that either means:irribeo said:
1. their particular SPDIF receiver was known to be poor - perhaps bad PCB design?
(or sample rate conversion was poor)
2. they are claiming that their internal ADC is superior to all other known ADCs anywhere on the planet.
Which do you think is most likely?
Maybe you have a ground loop and the digital coax outer is part of your audio signal ground and so carrying analogue return currents? If so, coax outer resistance could affect the behaviour of your analogue circuits.
Last edited:
Ampboard which is digital only (powerdac) is connected to Primare which only uses 2 pins of 3 in IEC power plug no earth. Ampboard is fed by +v and 0 of a single voltage conex SMPS which does have three wires to mains wallsocket. I had some problems with ampboard not functioning properly, like frozen, not responding to commands only functioning again after somekind of reset procedure. Should I add earth on ampboard somewhere you think?
My ADC and DAC (usually representing 30MHz cell phone baseband) are 12
or more pairs of LVDS with a Samtec connector at 50 ohms (100 per pair).
With intent or not, LVDS puts a DC bias on the dielectric at all times...
When I work with Audio, its an AP2700 with BNC connectors for all FCLK,
MCLK,DATA, all that junk. I have no idea if its 75 or 50ohm, and I never
measured to see if there was a DC bias on those digital signals. Work has
a way of erasing my natural curiosity for such trivia until its relevant to
something not functioning as expected...
I had a problem with a TOSLINK optical digital interconnect recently in
one of my own circuits. Didn't realize the receiver I had in hand would
not couple to DC, due to the self-leveling feature. My bad for reading
a clone part spec sheet without double-checking the original Toshiba.
Lesson: When using TOSLINK to watch the armature pass through a
coilgun, be sure you have the music playing first...
or more pairs of LVDS with a Samtec connector at 50 ohms (100 per pair).
With intent or not, LVDS puts a DC bias on the dielectric at all times...
When I work with Audio, its an AP2700 with BNC connectors for all FCLK,
MCLK,DATA, all that junk. I have no idea if its 75 or 50ohm, and I never
measured to see if there was a DC bias on those digital signals. Work has
a way of erasing my natural curiosity for such trivia until its relevant to
something not functioning as expected...
I had a problem with a TOSLINK optical digital interconnect recently in
one of my own circuits. Didn't realize the receiver I had in hand would
not couple to DC, due to the self-leveling feature. My bad for reading
a clone part spec sheet without double-checking the original Toshiba.
Lesson: When using TOSLINK to watch the armature pass through a
coilgun, be sure you have the music playing first...
Last edited:
You don't need to worry about digital cables, neither input transformers. Not that it doesn't matter, quite contrary, but you have bigger problems to overcome.
Your schematic looks like POC made by Cambridge. Let's analyse CN3 digital input:
C17 capacitor, 100pF is parallel to R18. This reactive termination (should be pure resistive) will give you this horribile spike down with additional reflections:
After you remove C17, you will get this:
Then we have input buffer (U23D), made from hex inverter. R20, 100R on input (Ri) and R21, 10K as feedback resistor (Rf).
Let's calculate input impedance for this buffer:
Ri + (Rf/Aol+1)
100R + (10K/20+1)
100R + 476R
=576R
(I am using approximation of Aol=20)
Now we have input termination resistor R18, 75R in parallel with 576R. Since resistors are in parallel, input impedance is 66R. This 66R (instead 75R, as it should be), is clearly visible when DAC is switched ON:
If you want to improve sound, then please remove C17, change R20 to 4K7 and R21 to 68K.
Change input RCA for 75R BNC and connect 6dB attenuator (or 10dB) between your digital cable and DAC input. You will enjoy even more.
After all those changes you should start worry about cables.
Even with those changes and input buffer and additional attenuator, different digital cables will still sound different
Last edited:
Interesting. So if I understand right you imply that thousands of bits are getting lost in the transmition line all the time degrading the audio signal yet in spite of the supposed loss of bits I can tell about no sound artifacts of any kind.
As I commented earlier it made no difference between audio cable or 50 ohm coaxial, and I was expecting to hear something since both are quite different and some people are fussing quite a bit about wire requirements.
I guess that the ultimate test is to (digitally) send pure tones and recover the bit stream right at the DAC chip input and see if different wiring actually alter the bit stream.
I have no near enough equipment at hand to do that test but undestand that some members here do and might want to carry out the test.
As I commented earlier it made no difference between audio cable or 50 ohm coaxial, and I was expecting to hear something since both are quite different and some people are fussing quite a bit about wire requirements.
I guess that the ultimate test is to (digitally) send pure tones and recover the bit stream right at the DAC chip input and see if different wiring actually alter the bit stream.
I have no near enough equipment at hand to do that test but undestand that some members here do and might want to carry out the test.
Last edited:
sextaafondo said:
No, no "getting lost", just wandering left-right in time. Google for "jitter" to learn basic terms and effects.
Gross jitter introduced by wrong termination will mask difference between cables and connectors, like I told you. First get rid of bigger problems, then worry about cables and transformers.
Right, I get the jitter point. Then Back to a question I posted above, would it make any sense to enter the 740C through its optical ports instead? Since jitter can't be overcame completely at least this can eliminate transmision line jitter.
The problem here is I'd need to use a converter with toslink output at the PC. Does it make any sense going this way?
I prefer this option as going into the 740C and start modding, at least at this stage that there are still options.
TBH I have to stress that I don't hear anything awful from the sound right now, quite the oposite (in spite of the surely present jitter) but would like inputs whether toslink is expected to be an improvement or not.
Converter to toslink output for PC sound card : Bi Directional Optical s PDIF Coaxial Audio Converter | eBay
The problem here is I'd need to use a converter with toslink output at the PC. Does it make any sense going this way?
I prefer this option as going into the 740C and start modding, at least at this stage that there are still options.
TBH I have to stress that I don't hear anything awful from the sound right now, quite the oposite (in spite of the surely present jitter) but would like inputs whether toslink is expected to be an improvement or not.
Converter to toslink output for PC sound card : Bi Directional Optical s PDIF Coaxial Audio Converter | eBay
No ground loops with optical connection I would think But I feel that would be like converting something little imperfect using questionable quality converter into something more imperfect to begin with. And then you might be buying glass cables, monocrystal less jitter lol, because the plastic fibre really stinks?? Jeje
But I feel that would be like converting something little imperfect using questionable quality converter into something more imperfect to begin with. And then you might be buying glass cables, monocrystal less jitter lol, because the plastic fibre really stinks?? JejeAnybody around that can perform jitter tolerance tests? Otherwise I think we are DOOMED and never will find the light at end of the tunnel lol
Jitter and Wander Testing for Fiber Optic Systems | Fiber Optic Training & Tutorials - FAQ, Tips & News
Jitter and Wander Testing for Fiber Optic Systems | Fiber Optic Training & Tutorials - FAQ, Tips & News
Extreme_Boky I was going to PM this to you but it seems you dont accept PMs hopefully it's not because of the opponents of the subject of this post........
Do you have a source for that picture of you posted on the second page of the thread "What difference does the quality of a digital interconnect make?" of the Silver Ribbon drawn between thicker silver wire center pins? or is that just a pic of the cross section of some interconnect product? I've been looking at silver plated occ copper but I think just reading your posts you've made me a convert to pure silver with internal hookup wire in superior geometry ribbon format........ TY you took me up a level more than I expected in studying wire. I learned so much today. I found a page which I think is the ultimate goal of electric design.
6moons audio reviews: PSC Audio AG Monolith
Monocrystal monolith design. MonoCrystal PCB. Silicon Chip is monocrystal no? OCC monocrystal ribbon wire....with shielding? All audio ribbons non trace direct to silver chip lead. Sata cable ribbon 2 sets differential pair shielded to ground pins. Burson discrete component design? I'm stilll a noob, but I'm getting the lay of the land.....
P.S. Computer weenie digital infallibility(in latin means not deceive aka papal infallibility) is a disease distracting from quantum physics common sense(information entropy). Every detail in the universe matters and they all influence each other. I have been fighting computer weenies in ##hardware chan freenode net on irc for years. It started with metallurgical discussions on wires and ended up in DVI HDMI infallibility ban for me. Which I can now prove DVI and HDMI are not not infallible as proven by 2560x1440 120hz Korean monitor signal artifacting tested across multiple DVI cables because 660mhz pixel clock is out of avg cable tested specs. 600MB/sx8 sata 3 cable? Cat6 600mhz Belden cable? Digital is not infallible and all cables are subject to analog limits. If it works better in extreme situations(shielding negated) its just a better wire overall because of superior geometry(even of the crystal grains, atoms)........
Switching from a Sata 2 unshielded cable I was having problems with Sata 3 shielded(now trying to source silver multicore cable wire with decent individual shields I picked up from another thread on this forum) noticed improvement in OS response(this is impossible to the believer of digital infallibility it either works or it doesnt there is no intermittence bug and interference doesn't exist... digital eliminates it because its a superior workaround to analog failures in telecommunications(cardas)).
Do you have a source for that picture of you posted on the second page of the thread "What difference does the quality of a digital interconnect make?" of the Silver Ribbon drawn between thicker silver wire center pins? or is that just a pic of the cross section of some interconnect product? I've been looking at silver plated occ copper but I think just reading your posts you've made me a convert to pure silver with internal hookup wire in superior geometry ribbon format........ TY you took me up a level more than I expected in studying wire. I learned so much today. I found a page which I think is the ultimate goal of electric design.
6moons audio reviews: PSC Audio AG Monolith
Monocrystal monolith design. MonoCrystal PCB. Silicon Chip is monocrystal no? OCC monocrystal ribbon wire....with shielding? All audio ribbons non trace direct to silver chip lead. Sata cable ribbon 2 sets differential pair shielded to ground pins. Burson discrete component design? I'm stilll a noob, but I'm getting the lay of the land.....
P.S. Computer weenie digital infallibility(in latin means not deceive aka papal infallibility) is a disease distracting from quantum physics common sense(information entropy). Every detail in the universe matters and they all influence each other. I have been fighting computer weenies in ##hardware chan freenode net on irc for years. It started with metallurgical discussions on wires and ended up in DVI HDMI infallibility ban for me. Which I can now prove DVI and HDMI are not not infallible as proven by 2560x1440 120hz Korean monitor signal artifacting tested across multiple DVI cables because 660mhz pixel clock is out of avg cable tested specs. 600MB/sx8 sata 3 cable? Cat6 600mhz Belden cable? Digital is not infallible and all cables are subject to analog limits. If it works better in extreme situations(shielding negated) its just a better wire overall because of superior geometry(even of the crystal grains, atoms)........
Switching from a Sata 2 unshielded cable I was having problems with Sata 3 shielded(now trying to source silver multicore cable wire with decent individual shields I picked up from another thread on this forum) noticed improvement in OS response(this is impossible to the believer of digital infallibility it either works or it doesnt there is no intermittence bug and interference doesn't exist... digital eliminates it because its a superior workaround to analog failures in telecommunications(cardas)).
Resistance does not matter here I would guess. I once got some other vandenhull interlinks to try-out, totally carbon, single-molecules structure lol, very high resistance compared to copper and silver,. must have them somewhere, I liked the totally carbon idea Succesors had metal in shield, to prevent poweramp failures lol not even half as fascinating lol. I will search to try them as digital RCA jeje. On analog RCA they did not sound bad which was amazing considering the resistance.
Succesors had metal in shield, to prevent poweramp failures lol not even half as fascinating lol. I will search to try them as digital RCA jeje. On analog RCA they did not sound bad which was amazing considering the resistance.
I use HD-SDI video signals every day. There the cable can make or break the signal, tho short runs aren't much trouble. These are a much higher bitrate than SPDIF. The nice thing about the digital signal is that if it gets messy after about 100 meter, I can "reclock" it. Basically a driver that decodes and reforms the signal before sending it out again for another 100 meters in great shape. But the signal can only be reclocked so many times before jitter builds up to an unacceptable level, a level so high the receiver can't deal with it. Nothing is perfect, but digital transmission sure makes HD video a lot easier.
It seems to me that the bitrates and bandwidth used by standard digital audio are so low, and the cables so short, that you'd have to really try hard to get it wrong. Of course I do keep one very thin, very cheap toslink cable that sounds dreadful, just as a reminder.
It seems to me that the bitrates and bandwidth used by standard digital audio are so low, and the cables so short, that you'd have to really try hard to get it wrong. Of course I do keep one very thin, very cheap toslink cable that sounds dreadful, just as a reminder.
So the crystal is the overriding factor but resistance must be quantum information loss and therefore audio information loss as well...... funny Van den Hul came up in an interview I was reading in connection with OCC.
An Interview with Steven Huang of Audio Sensibility | Dagogo | A Unique Audiophile Experience
Which also links to this very forum(advancements conceived here) of unorthodox and simplistic variety..... Groundside Electrons post?
Van Den Hul - Carbon and Hybrid Technology | technology
After reading the second half of this page, I am convinced the answer is close, but I am just too ignorant to see it.....
An Interview with Steven Huang of Audio Sensibility | Dagogo | A Unique Audiophile Experience
Which also links to this very forum(advancements conceived here) of unorthodox and simplistic variety..... Groundside Electrons post?
Van Den Hul - Carbon and Hybrid Technology | technology
After reading the second half of this page, I am convinced the answer is close, but I am just too ignorant to see it.....
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.