House speaker help please

[ericmark]

2.5mm cable is listed as 18mV/A/m since ohms = V/A then 0.018 ohms per meter and that's both feed and return.

However it is AC your feeding down the cable so the inductance and capacitance will have an effect. Normal method is to twist cable as in telephone cable to counter the effect. However with twin and earth that looks horrid. So twin flex is likely better.

2812X Twin Core Cable is available as 0.75mm and rated at 62mV/A/m so for 20m gives 1.24 ohms and as already said what one can accept will depend on speaker rating but it does look more like speaker cable.

3192Y PVC H05VVH2-F Twin Flat cable 300/500V is same size cable but with a outer Sheath. Also round Tower 2-Core Round Flex White 0.75mm² x 50m is not expensive at £10.50 a role and I would say any of these would likely do the job without using twin and earth which not only looks bad but could be mixed up with LV supplies in the future. You can get 2 core 2.5mm flex but price jumps to £50 for 100m and harder to find in 50m roles.

I think if I was owner of a house and I found an electrician had wired my speakers in twin and earth I would not be impressed even if it worked A1.

With a name like sparkmanc I am assuming it's not for his own house but one he is wiring for some one else? So not only does it need to work but also look right?

 

[chapeau]

2.5mm cable is listed as 18mV/A/m since ohms = V/A then 0.018 ohms per meter and that's both feed and return.

Yes, well that works out at 9mOhm per meter which isn't too far off the figure I gave. The reason yours is too high is because you took your figures from electricians books which give the value at 70C and mine is a more reasonable room temperature.

However I have just done the calculation from first principles using the resistivity of copper and it comes out at just under 7mOhm per metre.

Hardly worth quibbling about eh?

However it is AC your feeding down the cable so the inductance and capacitance will have an effect.

You won't be able to measure it with your ears though, not at the lengths of cable and the frequencies that go down speaker cable. It's probably as relevant as oxygen content i.e. irrelevant.

Remember you have a bloody big coil and magnet in the speaker, any cable effects will be completely and totally swamped by the coil and back emfs.

I think if I was owner of a house and I found an electrician had wired my speakers in twin and earth I would not be impressed even if it worked A1.

Mine is under the floorboards and in conduit behind the plaster, though I would agree with your sentiment.

What I'm advising the OP is that you don't need to spend more money on speaker cable than what you would pay for the price of conduit cable or twin and earth from screwfix.

 

[ericmark]

I would agree that no need to pay silly money and yes I did no correct for temperature.

I did learn how to set up transmission lines but since the frequency varies so much rather pointless.

And as you say the human ear is not that good anyway. I didn't do sound broadcast. Wish I had. We joined students on that course for maths and a few other bits and it did seem interesting. But had to decide between PLC's or sound broadcast and PLC's won.

 

[chapeau]

If you were to send the output of a powerful enough stereo amp down a transmission cable you might see cable effects but that's because those cables are measured in the tens of miles and not tens of feet. The cable starts to affect the signal when the length of the cable approaches a significant percentage of the wavelength of the signal. The wavelength of a 20kHz signal is in the miles, not feet, so a speaker cable in a house is not going to get anywhere close to affecting the frequencies found in your amp output.

The speaker has far more capacitance/reactance than any cable, the crossover has a capacitor and a coil in it which are build to tolerances that are far far far less than any cable effect. Your cable salesman will not tell you this of course, he will tell you that some cables are warm and some are bright and you match your cable to your amp. Absolute nonsense. Copper is copper and the stuff in your walls that carries 230v is built to a very high standard, the only question is how thick it should be. 1.5mm2 will suffice for anything under 10m I would guess, then you go to 2.5mm2.

Anyway, interestingly enough, if you take the figures in your OSG they are wrong if you calculate from first principles.

Suspect the OP is now too scared to ask what sort of kit to buy

 

[ericmark]

When I send a 145Mhz signal just 15 feet unless the feeder is balanced the transmitter will over heat and unless some auto shut down system is included in the design it will blow my output stage. I use 50 ohm coax and that is technically transmission cables.
At the frequencies used for sound it is very unlikely that the transmission cable will have any effect on output stage but once is still looking at impedance not resistance and just because both are measured in ohms that does not make them the same.
Band pass filters can and do separate the frequencies and direct them to the speaker best able to handle it.
This means sticking a DC ohm meter on a speaker network may not give one the impedance of that speaker network.
In the past I have made mistakes one was where to stop RF from getting into speaker cables I used coax for the speakers and this resulted in unwanted attenuation.
Tradition uses a twisted pair and the twist counters the capacitance but they don't look very neat so the figure of 8 has in the main replaced this. But also all unused cores of any cable should be grounded and to ground the centre core of twin and earth is a lot of work better not to have extra core.

However you are right that in most cases people will not notice any degrade of audio due to cable length or type. More likely to worry about echo where from the listening position one hears the PA type echo. And since it is near impossible to set up a system A1 in the home worrying about cable type is rather silly and I would just use any 0.75mm flex. If HiFi was required it would be limited to single room. As soon as it leaves a single room forget HiFi.

 

[chapeau]

Eric,

Just reread your first post on this thread and had to chuckle when you said

I hope this generates some more answers if only to say I have it all wrong.

Well, and I mean this in a constructive way, you have it totally wrong. In fact your first post demonstrates a significant misunderstanding of AV issues. I prefer the term AV to HiFi which is really a term which belongs in the age when 33rpm LPs came on the scene.

Cheers

 

[plugwash]

When I send a 145Mhz signal just 15 feet unless the feeder is balanced the transmitter will over heat and unless some auto shut down system is included in the design it will blow my output stage. I use 50 ohm coax and that is technically transmission cables.

Assuming a velocit of 2x10^8m/s (it varies by insulation material but this will be in the right ballpark)At 150MHz your wavelength is 1.3m so your 15 foot cable is going to be about 3 wavelengths long so transmission line effects are going to be very significant and you will have big problems unless your transmission lines "characteristic impedance" (which is only sort of the same thing as impedance but we won't go there for this post) is matched to the impedance of your source and load. At 20KHz your wavelength is about a kilometer so transmission line effects are not going to be an issue in the cable runs seen in a home install.

skin effect is another potential issue and cable manufacturers claim it as a reason for pushing OFC. However according to wolfram alpha http://www.wolframalpha.com/input/?i=skin+depth+in+copper+at+20khz is just under half a millimeter so I doubt this is anywhere near as big an issue as the proponents of OFC claim.

In general I still maintain that unless your runs are insanely long the important thing is to make sure you have plenty of copper.

At the frequencies used for sound it is very unlikely that the transmission cable will have any effect on output stage but once is still looking at impedance not resistance and just because both are measured in ohms that does not make them the same.
Band pass filters can and do separate the frequencies and direct them to the speaker best able to handle it.

True

This means sticking a DC ohm meter on a speaker network may not give one the[ impedance of that speaker network.

In the past I have made mistakes one was where to stop RF from getting into speaker cables I used coax for the speakers and this resulted in unwanted attenuation.

Can you be more specific about the run length, the speakers impedance and the type of coax so I can try to figure out what caused your attenuation. I suspect the most likely cause was that the inner core was simply too thin.

Tradition uses a twisted pair and the twist counters the capacitance

Twists reduce electromagnetic interference pickup and radiation (and as a result crosstalk) but I don't think they have any significant effect on capacitance.

but they don't look very neat so the figure of 8 has in the main replaced this. But also all unused cores of any cable should be grounded and to ground the centre core of twin and earth is a lot of work better not to have extra core.

Unused cores in mains cables should but i've never heard such a thing claimed for ELV cables. Personally if I had to use T&E for this I'd probablly connect the earth in paralell with the feed to the speaker. This would both reduce the resistance of the cable and as a bonus would hopefully blow a fuse if some idiot connected mains to the cable.

Having said all that I wouldn't use twin and earth for speaker wiring if I had a choice simply because of the risk of some idiot confusing it with mains wiring. Mains flex has lower risk of this and dedicated speaker wire lower still.

 

[ericmark]

Assuming a velocit of 2x10^8m/s (it varies by insulation material but this will be in the right ballpark)At 150MHz your wavelength is 1.3m so your 15 foot cable is going to be about 3 wavelengths long so transmission line effects are going to be very significant and you will have big problems unless your transmission lines "characteristic impedance" (which is only sort of the same thing as impedance but we won't go there for this post) is matched to the impedance of your source and load. At 20KHz your wavelength is about a kilometer so transmission line effects are not going to be an issue in the cable runs seen in a home install.

Your very close in free air 300/145 = 2.06896 meters but because of the velocity factor K of the conductor, the Physical Electrical Length is less than the Free Space Length. Normally, for practical purposes, K is taken as 0.95. To obtain Electrical length for practical applications, K is used as a multiplier, it will change as (300 x 0.95) / 145 = 1.96551 m.

The run was quite short and can't remember part number of cable seem to remember I bought a whole role of 50 ohm and it had no problems with 3 to 4 times that length transmitting at 100W into a HB9CV. It was because I was transmitting on 2 meter band I had used coax as when using side band it seems very good at breaking through.

Now I only use 5W on hand helds so not a problem any more. But having had problems I realised any old cable doesn't work. No longer got my books to check on calculations. Since I had gone to all the trouble to put cables in the wall I was not happy to then find they were no good.

But as far as this thread goes I would still use a standard 0.75 flex.

Anyway seems I have too many turns on the coil and am an old one and HiFi is no longer the "In" word. Likely went out with triode and grids and gone are the days of steam radio. Even when we use valves today all air cooled and whole different world.

So I'll leave it to you young ones to work out the modern Op-amp world and I'll return to grid bias.

 

[chapeau]

But as far as this thread goes I would still use a standard 0.75 flex.

Well it will work but over a 20m run the resistance of the cable is getting to be a significant proportion of the effective resistance of the speaker. So compared to 2.5mm2, and all other things being equal, your amp will have to be turned up higher with .75 flex to get the same volume. Amps work better (i.e sound better) when not being worked too hard. Especially cheap amps where the power supply is cheap.

Volume is measured in decibels and decibels are logarithmic. I would rather not lose 25% of my amp output volts in the cable if I could help it and if a better solution (thicker bog standard copper cable) cost a pittance compared to the price of the rest of the kit.

Your money should go into speakers.

The comments on mistaking mains cable for speaker cable are valid of course, but my point is mainly that you don't need to spend money on special copper as the copper in twin and earth is good enough. I used it because I had a load lying around, and it is labelled as such along its length.