Hi everyone. A kind of general question about twisting and/or braiding cables.
The speaker cables at http://www.geocities.com/jonrisch/cables.htm
seem very interesting. Jon's instructions are to twist the pair of belden coaxel cables used for each speaker. I'm thinking of using a bi-amp system (dual subs up to 250hz and electrostatics above).
Since I have to run four cable for each channel, would it be better to braid the four cables together and split off the pairs on each end running from the appropriate amp and speaker, or should I just twist one pair for the sub and twist another for the esl. I ask because of the many threads describing braiding belden computer cable.
Please excuse me if this is a silly question. Thanks for the help.
Best,
Ralph
The speaker cables at http://www.geocities.com/jonrisch/cables.htm
seem very interesting. Jon's instructions are to twist the pair of belden coaxel cables used for each speaker. I'm thinking of using a bi-amp system (dual subs up to 250hz and electrostatics above).
Since I have to run four cable for each channel, would it be better to braid the four cables together and split off the pairs on each end running from the appropriate amp and speaker, or should I just twist one pair for the sub and twist another for the esl. I ask because of the many threads describing braiding belden computer cable.
Please excuse me if this is a silly question. Thanks for the help.
Best,
Ralph
Not a silly question at all.
However suggesting someone twist pairs of coax cable is pretty silly, since the shield in the coax makes this technique totally ineffective (twisting only works with unshielded wires, you can put a shield over a twisted pair, but if the individual wires are shielded then twisting does no good at all).
Coax cables are designed to provide full coverage shielding with specific characteristic impedance. Neither of these attributes matter when driving a speaker. Coax does not normally have very large diameter center conductor. Depending on the cable chosen it could negatively effect the amps dampning factor.
An amp's output impedance is measured in fractions of an ohm. The speaker input impedance itself is below 10 ohms which makes this circuit very unlikely to pick up audible induced noise. For audio frequencies the cable length would have to reach several miles before you would need to be concerned about transmission line characteristcs of the cable.
In my opinion, the best speaker cables are short and fat. If you want to invoke wire vodoo, then try the Cat5 style cables. These are cheap and shouldn't create problems (unlike some audiophile cables that can make some audiophile amps unstable and oscillate).
To answer your guestion, the subs and tops wiring are not carrying the same signal so I don't think twisting them together would be a good idea. If you go with coax then twisting won't actually do anything in which case this will not be a concern.
Phil
However suggesting someone twist pairs of coax cable is pretty silly, since the shield in the coax makes this technique totally ineffective (twisting only works with unshielded wires, you can put a shield over a twisted pair, but if the individual wires are shielded then twisting does no good at all).
Coax cables are designed to provide full coverage shielding with specific characteristic impedance. Neither of these attributes matter when driving a speaker. Coax does not normally have very large diameter center conductor. Depending on the cable chosen it could negatively effect the amps dampning factor.
An amp's output impedance is measured in fractions of an ohm. The speaker input impedance itself is below 10 ohms which makes this circuit very unlikely to pick up audible induced noise. For audio frequencies the cable length would have to reach several miles before you would need to be concerned about transmission line characteristcs of the cable.
In my opinion, the best speaker cables are short and fat. If you want to invoke wire vodoo, then try the Cat5 style cables. These are cheap and shouldn't create problems (unlike some audiophile cables that can make some audiophile amps unstable and oscillate).
To answer your guestion, the subs and tops wiring are not carrying the same signal so I don't think twisting them together would be a good idea. If you go with coax then twisting won't actually do anything in which case this will not be a concern.
Phil
Hi Phil,
The cables I'm talking about use TWO coaxels for each speaker lead. The center of each coaxel is crossed over and soldered to the braid of the other coaxel.
You can see pictures and instructions for the cable at the link at the top of the page or here:
http://www.geocities.com/jonrisch/cables.htm
I've seen this cable referred to by a couple of sites and the general opinion is very high. After you take a look at it you'll see why I was asking about braiding & twisting.
Thanks again & I look forward to hearing from you.
Ralph
The cables I'm talking about use TWO coaxels for each speaker lead. The center of each coaxel is crossed over and soldered to the braid of the other coaxel.
You can see pictures and instructions for the cable at the link at the top of the page or here:
http://www.geocities.com/jonrisch/cables.htm
I've seen this cable referred to by a couple of sites and the general opinion is very high. After you take a look at it you'll see why I was asking about braiding & twisting.
Thanks again & I look forward to hearing from you.
Ralph
Hi,
The idea is not novel but that's not important.
Without looking at the weblink, I guess the braid serves to provide a low impedance for high frequencies, whereas to central conductor (often solid core) would serve the lower regions?
Ah! If only something was telling the electrons which way to go??
Cheers,
The idea is not novel but that's not important.
Without looking at the weblink, I guess the braid serves to provide a low impedance for high frequencies, whereas to central conductor (often solid core) would serve the lower regions?
Ah! If only something was telling the electrons which way to go??
Cheers,
I had this wonderful arrangement with the spouse. She got to do whatever Martha Stewart, Trading Spaces thingie appealed to her throughout the house, except in the den, where I had the final say, that is after exhausting about 10,000 "Oh, by the way" and "You know I was thinking" and "Are you really sure you want to do that". All of which were accompanied by the appropriate facial expression.
Then I got smart. I bought a gallon of flat black paint home and presented it as the ONLY other option.
Now besides the DIY lock she installed on the outside of my den's door, we have peace again, well kinda'.
Then I got smart. I bought a gallon of flat black paint home and presented it as the ONLY other option.
Now besides the DIY lock she installed on the outside of my den's door, we have peace again, well kinda'.
ralph-bway said:
I looked at Jon's explaination page: http://www.geocities.com/jonrisch/s1.htm. It is easy to to take comments out of context from someone's work and use them to make cheap shots, but I am not going to do that. I agree in general with his objectives. I do wish he had gone into detail about the advantages that his double coax has over regular wire.
I think this is mostly an attempt to reduce cable inductance, or perhaps it is simply to provide a more stable physical geometry. It's hard to imagine either of these mechanisms being an issue if you use properly constructed wire.
From Jon's website:
Ignoring these issues (minor in my opinion), these speaker wires are effectively two hollow braided straps with insulating cores and insulating jackets. Since the hollow insulating core will force the braid to maintain a round cross section they will have a very slightly lower impedance at high frequencies than a flat braided conductor of the same size does.
Given the amount of copper in these cables, they should make exellent speaker wires. I seriously doubt if you will notice any difference in sound quality if you don't bother hooking up the inner conductor of the coax. Using the configuration Jon recommends shouldn't cause any problems so use it if you wish. Since the outer braid is the current carrying conductor then twisting them will have it's usual effect on reducing radiated EMI and susceptibility to induced EMI.
Go for it! And don't braid the subs and mains cables together.
Phil
Twisted Pair Cable Works Well, Really Well.
The photo shows 10-pair.
20-pair works better but twice as much terminating time.
Both are available free from office floor demolition dumpsters, or very cheap from metals recyclers - you pay for the scrap copper weight only.
New 20-pair is around AUS$2.00 per metre from electrical wholesalers.
If you are happy to terminate 20-pair you get the best value for money cable around, and very good sounding at any price.
I have been using it for years and I can't be bothered to try anything else anymore.
Works really well on high power, PA, and shelf systems.
Eric.
The photo shows 10-pair.
20-pair works better but twice as much terminating time.
Both are available free from office floor demolition dumpsters, or very cheap from metals recyclers - you pay for the scrap copper weight only.
New 20-pair is around AUS$2.00 per metre from electrical wholesalers.
If you are happy to terminate 20-pair you get the best value for money cable around, and very good sounding at any price.
I have been using it for years and I can't be bothered to try anything else anymore.
Works really well on high power, PA, and shelf systems.
Eric.
Gravedigga!
I'm in a similar situation as OP with M-L SL3 (10in x 45in ESL and 10in sealed).
I would like to know if the twisted pair (CAT5?) cabling (not the Jon Risch) described would be especially low-inductance and thus extra suitable for ESL. (This is assuming one separates all the halves of the twisted pairs into plus and minus. I've seen other DIY cables that kept the pairs together, but that doesn't minimize inductance, right?)
I can bi-wire, so should I use a different type of cable for the woofer? Roger Sanders alludes to some special cable he offers for ESL, but I don't know if it can be DIYed.
Should I do both? Can I use any coax? I don't want them stiff. I'm more inclined to find some CAT5 or similar twisted pair construction.
I'm in a similar situation as OP with M-L SL3 (10in x 45in ESL and 10in sealed).
I would like to know if the twisted pair (CAT5?) cabling (not the Jon Risch) described would be especially low-inductance and thus extra suitable for ESL. (This is assuming one separates all the halves of the twisted pairs into plus and minus. I've seen other DIY cables that kept the pairs together, but that doesn't minimize inductance, right?)
I can bi-wire, so should I use a different type of cable for the woofer? Roger Sanders alludes to some special cable he offers for ESL, but I don't know if it can be DIYed.
Should I do both? Can I use any coax? I don't want them stiff. I'm more inclined to find some CAT5 or similar twisted pair construction.
I was gifted a roll of CAT3, which is 4 twisted pairs of solid copper 24AWG in a vinyl outer housing. I was thinking of separating the pairs into plus and minus at each end and braiding together 4 of these runs for a total of 16x 24AWG= 12AWG equivalent.
If four runs of this CAT3 are braided together, would this make low-inductance speaker wire suitable for my ESL?
Is there something else I should try?
If four runs of this CAT3 are braided together, would this make low-inductance speaker wire suitable for my ESL?
Is there something else I should try?
Thanks for the responses.
Right, SL3 is 1.5 ohms at 20kHz and nominally 4 ohms. XO is at 250Hz.
Running only one run of CAT3 gives me only 18AWG, which I intuitively thought would be too thin, even if bi-wiring. Although I don't believe everything I read, the M-L SL3 user's manual also says "Under no circumstances should a wire of gauge higher (thinner) than #16 be used." But I'm currently running stranded 18AWG and not bi-amping, and I don't think there's a problem (Yet? I've only had these for 2 weeks).
So to keep things neat and since I have so much CAT3, I figured I would braid 4x CAT3 (looks better than 3x), giving me 12AWG, and possibly also lowering the inductance compared with one run of CAT3 (I have no way of measuring inductance to confirm) and giving me the option to NOT bi-wire if desired. My cables will be 7-8 feet long.
I'm also trying to avoid much of the work involved with separating strands and endless braiding as described here:
VenHaus Audio DIY CAT5 speaker wire link
Right, SL3 is 1.5 ohms at 20kHz and nominally 4 ohms. XO is at 250Hz.
Running only one run of CAT3 gives me only 18AWG, which I intuitively thought would be too thin, even if bi-wiring. Although I don't believe everything I read, the M-L SL3 user's manual also says "Under no circumstances should a wire of gauge higher (thinner) than #16 be used." But I'm currently running stranded 18AWG and not bi-amping, and I don't think there's a problem (Yet? I've only had these for 2 weeks).
So to keep things neat and since I have so much CAT3, I figured I would braid 4x CAT3 (looks better than 3x), giving me 12AWG, and possibly also lowering the inductance compared with one run of CAT3 (I have no way of measuring inductance to confirm) and giving me the option to NOT bi-wire if desired. My cables will be 7-8 feet long.
I'm also trying to avoid much of the work involved with separating strands and endless braiding as described here:
VenHaus Audio DIY CAT5 speaker wire link
Re: Gravedigga!
There is a drawback. Connecting wires in parallel leads to an increase in capacitance. Zctot = Zc1 + Zc2 + .. + Zcn. The same is true, if you twist wires together. By twisting them you bring them nearer to each other, thus raising capacitance even more. To bring capacitance down, you need to make the space between them as big as possible, their diameter as small as possible and their length as short as possible. The contradicting demands we all love so much, because they force us (and cable makers) to make compromises.
In theory bi-amping with stranded wire and big diameter for the lows and single wire with small diameter for the highs is the perfect solution. In practice there are a few questions with no simple answer. Where is the frontier between lows and highs? And does it coincide with the speakers crossover frequency for bi-wiring? What about the mids? And what about the interaction between cable impedance, speakers and amplifier?
On the other hand, there are inductors in the path before a bass driver with several mH of inductance (and several percent of tolerance!). Do a few µH of cable inductance really matter? And there are capacitors in the path before a tweeter with several µF of capacitance (and also several percent of tolerance). Do a few pF of cable capacity matter? In either case the cable impedances will be hundred or more times lower than the tolerances from the left speaker to the right. Interesting read here with surprising conclusion.
You will have to experiment to find the solution that works best for you and your speakers (and amp).
The inductance of a wire depends on its length. So the best way to keep inductance low in the first place, is to keep the wires short. Connecting wires in parallel leads to a reduction of inductance. 1/Zltot = 1/Zl1 + 1/Zl2 + .. + 1/Zln.Tosh said:
There is a drawback. Connecting wires in parallel leads to an increase in capacitance. Zctot = Zc1 + Zc2 + .. + Zcn. The same is true, if you twist wires together. By twisting them you bring them nearer to each other, thus raising capacitance even more. To bring capacitance down, you need to make the space between them as big as possible, their diameter as small as possible and their length as short as possible. The contradicting demands we all love so much, because they force us (and cable makers) to make compromises.
For low frequencies you want low inductance and resistance, i. e. big diameter and as many wires in parallel as possible. For high frequencies you want low capacitance, i. e. small diameter and as few wires as possible.Tosh said:
In theory bi-amping with stranded wire and big diameter for the lows and single wire with small diameter for the highs is the perfect solution. In practice there are a few questions with no simple answer. Where is the frontier between lows and highs? And does it coincide with the speakers crossover frequency for bi-wiring? What about the mids? And what about the interaction between cable impedance, speakers and amplifier?
On the other hand, there are inductors in the path before a bass driver with several mH of inductance (and several percent of tolerance!). Do a few µH of cable inductance really matter? And there are capacitors in the path before a tweeter with several µF of capacitance (and also several percent of tolerance). Do a few pF of cable capacity matter? In either case the cable impedances will be hundred or more times lower than the tolerances from the left speaker to the right. Interesting read here with surprising conclusion.
You will have to experiment to find the solution that works best for you and your speakers (and amp).
Thanks for that, Pacificblue. I had read NP's old paper a few days ago and overall was a bit disappointed by a lack of preference by NP. Perhaps NP has some updated opinions on speaker wire?
Figure 6 (response deviation on a D-W full range ESL) brought it to my attention that there might be unique problems for an ESL that I could nip in the bud...
Figure 6 (response deviation on a D-W full range ESL) brought it to my attention that there might be unique problems for an ESL that I could nip in the bud...
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.