|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
In Reply to: making a cable snake posted by brain echo on March 23, 2002 at 20:10:56:
Look for foilscreened, offers the best screening, Starquad if you can afford it will help protect against RF interference. Oxygen free would be available at that price, but if you have lights on dimmers and radio equipement, the aforementioned screening is more the priority.
Regards
Roland
Follow Ups:
Roland,Foil screens give coverage against interfence above 1 MHz, but doesn't do much for noise below that. A high quality, tightly twisted pair with good double braided shield will give much better results for long distance audio.
The best cable for long distance balanced audio (analog or digital) is without a doubt Belden 1800F digital audio cable. This cable has extremely low capacitance (which reduces high frequency roll off). In addition it has French Braided double shields (two layers of braided shields with the braid patterns mirrored to minimise gaps in coverage).
http://bwcecom.belden.com/college/Prodbull/np127.htm
Another issue with snakes is avoid the temptation to bundle speaker cables or even line level signal returns in the same snake. Doing this can create an ultrasonic feedback condition that will roast your amps (and tweeters) and you can't even hear it while it is happening.
About Starquad, This cable has 4 times the capacitance per foot of Belden 1800F. When you need to run cables long distance this factor quickly becomes of primary importance. The reason for this is that the capacitance of the cable combines with the output resistance of the sending device to form a low pass filter. The higher the capacitance (which increases linearly with length) the more the trebles get rolled off. What may work great at 10 feet will often perform miserably at 100 feet.
Phil
I dispute your assertions. Belden cable is good cable no doubt, but there are many other very high quality manufacturers out there. Digital cables are normally designed for 110 Ohm impedance so are not suitable for mic leads, thats why manufacturers make both. The BBC typically use starquad mic cables, for extremely long runs (quite usually 100 metres or more). Any high quality mic cable should not exhibit any significant HF roll-off under a distance of well over 300 metres!! I'm talking about roll offs over 60khz!! PA companies regulary pass line and mic level signals up and down the same multicore without frying there systems. Who would dream of sending amplifier outputs down multicore mic cables? Starquad is particulary favoured by the BBC and other broadcasters for its excellent interference rejection, particulary as many of the sites that they can end up working with could be subject to interference from lights, transmitters, etc beyond there control.Regards
Roland
Hi Roland,Point by point responses.
> Digital cables are normally designed for 110 Ohm impedance so are
> not suitable for mic leads, thats why manufacturers make both.The characteristic impedance of a cable is only important for high frequency (RF) signals. At audio frequencies your cable length needs to get into the kilometers before you will notice any bad effects due to this. You are correct that digital audio cable needs to be 110 ohm since the edges of digital audio signals have a very high frequency component, but analog signals don't nor do they particularly care about the cables impedance. This means that very good digital cable can also be very good analog cable.
> Any high quality mic cable should not exhibit any significant HF
> roll-off under a distance of well over 300 metres!!The roll off frequency is determined by your source impedance and the cable capacitance. Lower capacitance is better.
> PA companies regulary pass line and mic level signals up and down
> the same multicore without frying there systems.PA companies also regularly recone smoked speakers and repair damaged amps. I don't know how ofter this is caused by ultrasonic feedback rather than just normal abuse, but I have seen this particular problem myself. Since then I have done research on this subject and discovered that this is not that uncommon a problem. In addition it is a very difficult problem to detect during a performance since nothing obviously sounds wrong, except your amps lack drive power and overheat too easily.
> Who would dream of sending amplifier outputs down multicore mic
> cables?Just about every snake manufacturer I know of carries a line of "Powered" snakes that combines speaker cables with mic cables. I agree this is a very dumb idea. I didn't want this poster to copy their mistake thinking it's ok since Whirlwind, Rapco, etc do it.
The reason why this particular Belden cable is so much better than the competition (including Starquad) is a combination of very low capacitance, very superior shielding (two thick layers of braided shields), very consistant construction and very good handling properties. The only think the 1800F is lacking in my opinion is a plenum rated version.
Phil
My point was simply that you stated that you could get significantly worse HF performance form extending a snake from 10ft to 100ft, and thats not true. Manufacturers of high quality microphones such as B&K state that there is no significant HF loss in good quality mic cables of 300-500 metres in length.Starquad offers the very highest RF rejection currently available, thats why it is the choice of many broadcasters.
Many other manufacturers now make cables of similar spec to those of Beldon. Klotz, Canford audio, Connectronics,etc.
Regards
Roland
I feel I need to interrupt your friendly debate. Roland is correct in saying that adverse HF effects of cable lengths occur at distances over 300 meters (or 1000'). This is true assuming the cable is properly balanced. Lengths under that are not subject to appreciable losses. Now to address the impedence issues, 110 ohm cable is best suited for AES and other digital audio signals. To boot, StarQuad is the recording studio standard; It has the best shielding against EMI and RFI signals, making it the most appropriate for analog audio signals.
Now, live sound. The ultrasonic frying of amplifiers and speakers is a bit over the top. There are plenty of possible factors to the frequently inefficient sound reinforcement systems we use in clubs & arenas. As far as cabling, for years people have been running mic and line level to FOH, monitor & FOH feeds (at line or speaker level) to the stage, AC cables, clearcom, MIDI, etc within inches of each other. In theory it's problematic and of course it is not recommended. Much of the time, however, it happens anyway, and it's fine. We can't all afford fiber optic snakes and sometimes there isn't enough time to separate the cable runs.
For a studio or any permanent installation situation, you absolutely should keep mic level, speaker level and AC cables at least a foot apart from each other. This is fairly common knowledge.
Get StarQuad for your XLR cables, make sure you have a shield around the pairs, and strain reliefs at the ends. You'll be fine. Oh yeah, don't forget to feed the back shell of the connector on the cable BEFORE you solder the insert! ;-)
By all means feel free to jump in, but please don't ignore the reality of what StarQuad Cable is and what it is intended for. StarQuad is an excellent cable for rejecting EMI and RFI interference, which is very valuable in a broadcast studio setting. Unfortunately, the design of the StarQuad cable results in a much higher capacitance per foot than other cable designs.From the Canare website, StarQuad cable capacitance is listed at 145 to 170 pF per meter. Lets assume an average of 158 pF/m for the sake of argument.
From the Belden website, 1800F cable capacitance is listed at 43 pF/m.
Assume a typical mic output impedance = 600 Ohm. Assuming the load impedance is much greater than the source impedance, then the -3dB roll off frequency of a cable can be calculated as follows:
Fc = 1 / (2 * Pi * C * Z_Source)
For a 100 meter cable of Canare StarQuad the -3dB frequency would be:
Fc = 1/(6.28 * 158 pF * 100m * 600 Ohm) = 16800 Hz.For a 100 meter cable of Belden 1800F the -3dB frequency would be:
Fc = 1/(6.28 * 43 pF * 100m * 600 Ohm) = 61720 Hz.Based on these figures I would be very comfortable using a 100 meter Belden 1800F cable, but would be somewhat concerned about the audible sonic effects of using a 100 meter Canare StarQuad cable. The -3dB frequency for a 300 meter StarQuad cable is a totally unusable 5600 Hz, while the Belden 1800F cable -3dB frequency is still a perfectly respectable 20573 Hz.
Belden 1800F has the same or better EMI/RFI interference rejection properties (due to it's double braided, French shield) as StarQuad cable without the high capacitance penalty. I can understand StarQuad being used in a studio where a 50 meter mic cable is considered a long run. Live audio is different from studio use though, cable runs of 100 meters or more are not at all uncommon and even longer runs are encountered in concert settings. StarQuad doesn't do anything better than Belden 1800F, and 1800F will do it for further distances at a lower cost.
Phil
Well assume for a start that professional mic impedance is of the order of 150 ohms not 600! I'm not an electronics expert, but I do use runs of up to 150 metres, with absolutely no discernable downside effects. Being that I do a lot of critical Classical recording, and the questioner was looking at a stage snake, I think that belden as against starquad is a nonsensicle argument. In a PA situation RF and EMI are going to be his major problems. chances are that his snake won't even be 100 metres long. Maybe Keith AKA SSL tech would comment, but I'm sure that your maths is incorrect. I don't think that any commercial quality cable even at 500 metres would exhibit HF loss close to that you are suggesting for starquad at 300 metres.Regards
Roland
Roland,Why would you assume that 600 ohm is incorrect? The standard mics you see every day in live sound are still the Shure SM57 and SM58 and these are 600 ohm mics. Don't confuse your experiance in recording with live sound reinforcement.
I got the cable roll-off formula from my copy of the "Yamaha Guide to Sound System Engineering". This is actually just the standard calculation for capacitive impedance, the -3dB point occurs when the cable impedance equals the source impedance.
The reality can actually be worse than I stated since some mixer boards don't have particularly high input impedance mic preamps (at one time the standard for these was 600 ohms, but modern consoles are now normally somewhere around 6K-7K ohm input impedance). If your mixers input impedance is on the low side then you will see the effects of excessive cable capacitance even sooner.
I'm not just making this stuff up. Read any audio engineering book and they will agree with me, excessive cable capacitance is not a good thing, especially in a snake cable. My snakes (I have a 150 ft, and a 250 ft snake cable) both have multi-pin connectors which means I can easily daisy chain them in case I ever need an extra-long run. Why would I want to buy a snake cable that won't let me do that?
Phil
Phil,
Certainly, excessive capacitance is not a good thing. -It is the single biggest limiting factor on hf extension. -However, I'm mot sure that I would agree with the line of reasoning that suggests that lower input impedances aggravate the problem...-If you were to exaggerate the effect and strap a 10 ohm resistor across the load, -making the effective load impedance 10 ohms, then the parallel effect of the capacitive reactance becomes less significant, not more. ...Although a 10-ohm load would be ridiculous, since it would most likely sound awful due to the colossal impedance mismatch, rather than capacitive load issues!
Also, 600 ohms for an SM58 or an SM57? -Not according to Shure. -See the attached pdf link for the full shure specification... 150 ohms is in fact exactly a specified source impedance for both models.
Sanity in a Mad, Mad, World! lol. As Keith states, almost all pro mics are around the 150 ohm load, the 600 ohm were more commonly cheap band mics, like the shure prologues and unidynes.regards
Roland
So here's how I have the Neutrik A2 set up:Generator impedance = 200 ohms, (No 150 ohm setting unfortunately)
Generator frequency = 40kHz (as high as it goes for this set!)
Measurement input impedance = 600 ohmsA quick try of a handful of standard values available to me right this minute gives a 2.4dB drop a 40kHz when I strap 27nF across the line. 27nF = 27,000pf which at 150pF per metre, is equivalent to 180 metres.
-At 150 ohms source impedance, the drop off would be lower still, so it definately seems to disagree with getting a -3dB point at 16,xxxHz with only 100 metres of cable...
As for personal use, there are a number of factors governing this kind of stuff. I have always used starquad-type cable for individual mic runs in noisy environments, -Bundling cables produces a de facto improvement in shielding, but sometimes it all gets a bit to big, so a good multipair snake is a better choice as soon as outside dimension becomes a limiting factor.
On the subject of digital cable, -I have a few hundred metres of it (Clark Wire & Cable part # 801 and Gotham GAC-2) here and I've been forced to use it in a pinch for analog connections... -it's about as microphonic as anything I've ever tried. -Now I realise that other people might make a decent 110-ohm digital cable which is softer than this stuff, and may well be a little less ill-suited, but trust me... you wouldn't want to use this for analog connections -ANYWHERE!!!
-If anyone has any other types of 110-ohm cable, I'd be interested to know if they can hear anything when they flick the cable with their fingernail...Technically yours,
Keith
...In fact the math using a more correct source impedance of 150 ohms gives a -3dB point at 67kHz using the starquad at 158pF per metre, and a -3dB point of 246kHz for the Belden. -Now that seems more like it!!!-Certainly, the lower the capacitance the better, but for short runs in hostile environments, the Starquad -or similar- always gets my vote. -We have an account with Clark, so I normally go with their 'mink-4' cable, which is also a quad.
Frequently overlooked in installations (in addition to the very valid point of spacing cables where possible) is the grounding scheme. -Sometimes you have to star ground with far-end shields lifted, sometimes this forms a nice little antenna-array, and you have to take some of the far-end shields to ground with a small capacitor... sometimes you have to connect both ends... sometimes you're connecting unbalanced sources to balanced inputs, sometimes it's the other way round... Some manufacturers (Like MCI!) used to make the decision for you by installing the shield capacitors on the unit!
-An expensive cable will do you no good at all if the grounding scheme isn't properly worked out... -In fact a friend of mine recently invited me over to hear his new home system. -It took me less than three seconds to tell him that the whole thing was 'out of phase'... -We checked the connections... everything looked good... -finally I got the trusty ol' fluke meter out of the car. -Turns out that one of his (rather expensive, hand-made audio-salon type) speaker cables had a polarity flip... -He'd been listening fairly happily to the system for some weeks... -I'll never trust an opinion from him again!!! ;-)
Technically yours,
Keith
This post is made possible by the generous support of people like you and our sponsors: