|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
92.6.230.247
Hi,
About to build some UBYTE-2 speaker cables and wondered about some variations-
- I notice the UBYTE-2 removes the copper mesh around the foil on the outer conductor of the coax cable, which then neccessitates the use of trailing leads of a different construction (whereas Jon Risch's original design uses the mesh as the conductor which goes to the speaker/amp terminal. I like getting rid of the stranded mesh, but I don't like adding another run of different cable and a solder connection on both ends. Could I cross connect the foil shields across and then use the stronger centre conductors to the speakers, or even take the centre conductor up to the binding posts then loop back down to the foil on the other side, etc? If those descriptions don't make sense, I can post a diagram!
-Secondly, can anyone answer why the cables are cross-linked? As I understand it, the idea is just to use a solid and foil conductor on each positive and negative run- it would theoretically be easier to just link the outside and centre on each run, or is this a way to prevent interaction down one coax between the inner conductor and shield (I'm sure there is a reason, I just like to fully understand the design before I build it)?
- Is there a better alternative to heat shrink for the re-insulation of the foil, like foamed Teflon? Can one get hold of higher purity foil and centre conductor in, say, 6N copper, and replace them?
Lastly, how good do these actually sound? I've heard references to good reviews, but not actually found any which describe the cables' characteristics, strengths, or shortcomings...
Cheers,
James
Follow Ups:
It was a real pain to make my version of Ubyte-2 speaker cables (bi-wire set), but it was made with silver foil wrapped around stripped coax. It's been many years, and unlike many other wires, some very very expensive, these Ubyte-2 speaker cables stay in my system. I really think a cable company with resources should design a Ubyte-2 type of cable and offer to the public.
Hi Jon,
That's interesting to hear- oddly enough, the guy I spoke to from Webro was curious enough when he asked what I wanted such a specific design for that he asked me to send him a link to the TNT site, so I did. Even if they supplied a bare cable, but built to purpose (I said that a good cable would probably something along the lines of a coax built from the ground up with high purity (OCC or 6N) copper centre conductor and foil, air spaced Teflon dielectric and outer jacket, and no braid) that would make things rather easier either to DIY, or for a company to build as a reasonably priced cable.
Who knows though- they'll probably base a decision on perceived interest in potential sales- maybe if enough people enquire it could convince them?
Where did you source high purity silver foil, and what did you use as an outer jacket and lead-out wires?
I agree though, although I haven't listened to any yet, the actual idea Jon Risch thought up is very clever, inspired, really it would be curious to see how good it would sound with a purpose built cable, seeing how many people swear by one that's been modified to suit...
First, the UBYTE-2 recipe calls for a specific coaxial cable, one that has copper foil with a vestigial copper braid over the foil to make electrical contact at either end. The core surrounding the solid center wire of pure copper has hollow channels, creating a primarily air core dielectric, one with a very low amount of dielectric involvement.
The original production runs of this special coax (generally available only in Europe) had a decently thick copper foil, so when the copper braid is removed, there was copper all around the coax core, and it was thick enough to solder to, and to make good electrical connection.
Later production runs had the copper foil so thin, that it was being cut through by the copper braid, and was left in fragmented pieces that were electrically discontinuous and did not wrap around the whole circumference of the cable. Thus, the original coaxial cable used to make the UBYTE-2 is effectively no longer available.
YOU CAN NOT CONSTRUCT THE UBYTE-2 CABLE USING REGULAR ALUMINUM FOIL COAX.
Good electrical connection for audio frequencies can not be made to aluminized mylar, or even true aluminum foil.
If you try to build the UBYTE-2 with any other coax than the originally specified type, and a spool that has thick enough copper foil, IT WILL NOT ONLY SOUND TERRIBLE, BUT IT COULD FAIL AND/OR DAMAGE YOU AMP!
Some folks have taken various coaxial cables that had a foamed teflon core around a solid center wire, and removed the jacket and braid, and then laboriously wrapped copper foil from hobby shops (or copper tape from security supply companies) around the foamed teflon core, and created their own base cable for a UBYTE-2 construction. Needless to say, this is not only difficult, but the likely hood of the hobby grade or security grade copper foil being as pure as cable grade copper is virtually nil. You then also have to insulate the copper foil of each coax from the other, and this is not easy to do with quality materials.
All of this means that if you want to be able to build a cross-connected speaker cable, Belden 89259 is still the best bet, and while some folks seem convinnced that it is not the best way to approach this geometry, due to the stranded wires and strand based braided shield, it helps to keep in mind that the big bugaboo evil of stranded wires is supposed to come from skin effect, and we are talking about a relatively small center wire for Belden 89259 (22 ga.), and braid that only has a thickness when two braid weaves overlap of perhaps the equivalent of 32 ga. wire, and the fact that the coax braid is symmetrical about the cable center means the driving forces behind strand jumping are much lower than they are for wire of a gauge of 32 AWG.
Most of the signal passes through the cable braid, and a portion through the center wires.
So, as a result of these physical dimensions, and the electrical symmetry the coaxial cable possesses, the "evils" of stranded wires are not nearly as strong or pervasive as they would be for a zip cord construction, where you can really notice the problem with stranded wire when it is 12 and 14 gauge.
RE the reason for cross-connecting the center wire of one coax to the braid of the other, this creates a self-cancellation of the magnetic fields of the various conductors, which reduces the self-inductance of the overall speaker cable. As a result, the inductance of the cross-connected 89259 speaker cable is quite low, with out the usual penalty of having a very high capacitance as a counter-balance to the low inductance.
The only speaker cables with lower inductance tend to have capacitance's that are five to ten times that of the cross-connected 89259.
If you do not cross-connect the two coaxes, then the inductance is not that low, and the capacitance does not go down very much, leaving you with a much lower overall performance.
Jon Risch
Found this-
Belden branded with foil shield and what I think is air spaced PE-
http://cpc.farnell.com/jsp/level5/module.jsp?moduleId=cpc/517111.xml
http://cpc.farnell.com/1/1/59301-coaxial-ct100-ct100c0-00100-belden.html
I'll phone them to see if I can get a sample- don't fancy buying a whole reel to find out the foil is crap or it's not air spaced PE!
I don't see any copper foil, only copper braid.
If you were originally going for the foil and solid wire version, this may not be quite as good, but only actually making them will let you hear if there is a difference in your system.
Jon Risch
Hi Jon
Thanks very much for all that info, that's really interesting and I'm especially happy to have an explanation of why, with very high quality cable designed in the right way, there is no problem with stranding.
Regarding the coaxial cable I found- I checked with the belden catalogue and there is foil shielding as well as braid on that model. I also found a company here called Webro, who are considered as good, apparently. Here's the thing though- the CT-100 cable which are all the ones I found with the similar construction to the UBYTE specced cable, is being phased out- it seems that there are only old stock left of decent manufacturers like Belden and Webro. I spoke to one of the managers at Webro and got told that the CT-100 cable, which used to be the most popular, and which called for exactly the UBYTE-2 design of a pure copper conductor and foil, with air spaced PE dielectric, has now been essentially abandoned in favour of foamed PE, which the satellite industry consider better. So it may be the last chance to make a UBYTE-2 by buying the last reels from good manufacturers. Maplins has never really sold the best, it's just convenient for hobbyists and generally I actually avoid it.
In any case, I've found a local installer with an old reel, and he's letting me strip some back to see if the foil's OK. To be honest, now you've explained stranding better, I'm more tempted to try your original design in the first instance, as it's much easier, and I prefer it in some ways, as I mentioned in the OP!
Anyway, thanks so much again, I really do appreciate your help and support- most kind!
Best,
James
Hi Jon,
Thanks for the reply! OK, well that's annoying, as I already bought the cable specified on the TNT website from my local Maplins (I live in the UK) the stock number was the same. The foil looks decent enough to the eye at least- is there a specified thickness I could check with a gauge, if the production runs are so hit and miss? I must say, it's a good job I happened to post on here about it if there are issues which could cause potential damage- one would think the guy on the TNT site would be responsible enough to update his site and inform people the cable is no longer suitable.
Anyway, OK I hear what you're saying about the stranded construction, but it was one of the main reasons I was interested in a UBYTE-2 build, to be honest. Although this seems like a good opportunity to ask, since we're on the subject of stranding ;) because I've heard so much negative about it and yet some of the best speaker cables seem to use it, e.g., Kimber. I think they use a technique called 'varistrand' where each small gauge conductor has strands composed of different diameters to 'spread' skin effect over different frequencies, or at least that's the theory. What I've never understood, and it also applies to the braid example in the Belden you describe, is won't all the little thin wires, which aren't individually insulated, just short together and act as one bigger conductor, and will you get tiny distortions as the wires short? Maybe you don't, I'm just really curious for an explanation! I've noticed that even Kimber have started using solid core on their highest range interconnects and speaker wires now.
I suppose the other alternative is ro use the cable with the foil AND copper mesh in place?
Coming back to the foil only construction type for a moment, has anyone found alternative foil based coax? There are so many on the market here using pure copper foils and foamed Teflon dielectrics. Regarding self winding the cable in copper, the UK site hificollective.com does sell copper foil in various sizes for audio use (Mundorf brand, if memory serves), so that's a possibility I suppose, although that seems like it would be quite a job, would anyone who's been there and done it care to comment?
Best regards
RE 1st paragraph, if the foil is not all fragmented and does not look damaged as you strip back the jacket, then it may be that you got a "good" batch of the coax, and can proceed as directed by the TNT instructions. It was always very thin, but the original batches were substantial enough to be worked with without falling apart. If nothing else, you'll find out once yu start the full assembly and construction.
RE the second paragraph, the stranded wire issue has always been one of degree, rather than an absolute go/no go situation. The best example of stranded wire sounding "bad" is 12 ga. zip cord speaker wire; of course, neither solid or stranded 12 ga. sounds that good to begin with, so it's kind of a moot point to use solid 12 ga. As the wire size goes down, the stranding becomes less and less of an issue, and once you reach a certain point, it just doesn't matter any more. The big argument is then, at what size does this occur?
Kimber uses very pure conductor materials, which tend to reduce any strand jumping problems, as it is the not so solid contact between the strands, and oxidation and other contaminant effects that cause the stranding to become a sonic issue in the first place. If all the strands all made perfect electrical contact, no penalty to using stranded wires.
RE your last paragraph, beware the superficial specifications of various coaxial cables, the center wire may be copper-covered steel, or silver plated, or some other undesirable situation, if there really are that many copper foil coaxes available in Britain/Europe that also have excellent dielectrics, and are of an all pure copper construction, this is news to me.
Jon Risch
Post a Followup:
FAQ |
Post a Message! |
Forgot Password? |
|
||||||||||||||
|
This post is made possible by the generous support of people like you and our sponsors: