Impedances and transmission line effects...

[Cardigan]

I think Cardigan is right here. If the cable isn't being used as a transmission line (more strictly if the transmission line effects are negligible) then capacitance is the dominant factor. It can be treated as a lumped capacitance and the effects of inductance will normally be negligible in such applications. This would cover audio for most distances and RF for short distances where the wavelength is much greater than the length of the cable.

I would point out that in the past I've seen a short (<50m) run of twisted pair driven from a 'balanced' source that was anything but - and of about 1000 Ohms output impedance - suffer from notable treble loss.

The driver was an op-amp with 1000 Ohms in series with its output in the hot side of the feed: the cold side was a 1000 Ohm resistor in series with a 100uF capacitor to earth.

[Mark Hennessy]

The inductance can be measured by shorting one end and measuring the open circuit end.

Indeed, but it's not as much fun

Also, I've found that measuring Z gives more reliable results than measuring L, as I get different results from each of the 3 different bits of gear I have for measuring inductance.

[Mike Watterson]

I've found that measuring Z gives more reliable results than measuring L, as I get different results from each of the 3 different bits of gear I have for measuring inductance.

Could it be the frequency used to measure and or measurement method being affected by resistance or capacitance?

I mean it's not as if the inductance is really measured. Something else is usually measured. A nulled bridge with box of known inductances can be misleading.

It's hard to measure an inductance accurately if it has a metal / ferrite /dust core (non linear?), or self capacitance or resistance etc that's significant. Measuring Z for coax might be easier?

[pwdrive]

Just some general info (TDR):

http://www.montana.edu/blameres/courses/...r96a11.pdf

Lawrence.

[Mike Watterson]

Twisted pair:
Cat3 is about 60pF per meter, Cat5 less. So 1K (or sounds like it might even be 2K) driving twisted pair seems a bad idea for longer distances. Characteristic impedance might be about 120 ohms. Why were they making it bad by such large resistances and a "not really balanced" connection?

Measuring capacitance of cable can usually be done quite reliably at 1kHz or 10KHz etc, without difficulties of measuring cable inductance.

[Mark Hennessy]

I would point out that in the past I've seen a short (<50m) run of twisted pair driven from a 'balanced' source that was anything but - and of about 1000 Ohms output impedance - suffer from notable treble loss.

The driver was an op-amp with 1000 Ohms in series with its output in the hot side of the feed: the cold side was a 1000 Ohm resistor in series with a 100uF capacitor to earth.

With RG58 at 58pF/m - let's call it 60pF as we're engineers, not scientists - then 50m would be 3nF. With the total Zout being 2k, the -3dB point is ~26kHz. There would be some attenuation at lower frequencies, of course, but that doesn't seem like much to me. Measurably, certainly, and should be attended to in an installation, but would it leap out as being a big problem unless you were doing an AB comparison?

Regarding your point about balance, I'd say that actually, it was nearly perfectly balanced.

One of the biggest misconceptions that I have to clear up for folk at work is that balanced operation is nothing whatsoever to do with the voltages on the two lines. Rather, it's about having identical impedances to earth, so that an interfering signal affects both lines equally, so that the differential input stage receiving the signal is able to subtract the interference from the wanted signal.

I said "nearly" because as well as the resistance to earth, the capacitance ideally needs to match as well. Unlikely with a coax, naturally...

Of course, if you do drive the source in true antiphase, you get an extra 6dB headroom for a given set of PSU rails. So that's an obvious win, but it isn't essential by any means.

[Gryphon]

I would point out that in the past I've seen a short (<50m) run of twisted pair driven from a 'balanced' source that was anything but - and of about 1000 Ohms output impedance - suffer from notable treble loss.

The driver was an op-amp with 1000 Ohms in series with its output in the hot side of the feed: the cold side was a 1000 Ohm resistor in series with a 100uF capacitor to earth.

With RG58 at 58pF/m - let's call it 60pF as we're engineers, not scientists - then 50m would be 3nF. With the total Zout being 2k, the -3dB point is ~26kHz. There would be some attenuation at lower frequencies, of course, but that doesn't seem like much to me. Measurably, certainly, and should be attended to in an installation, but would it leap out as being a big problem unless you were doing an AB comparison?

Twisted pair is likely to have a rather greater capacitance per unit length than RG58, I would suggest.  You only have to double the figure you quote (and I seem to recall rather greater numbers being quoted upthread) to get the 3dB point down to 13kHz.  I suggest this would give noticeable treble loss, even if the clean original wasn't available for an AB comparison.

If it was the only line available, I'd give it to a journo for a speech piece.  I think I'd want something a bit better for Carols from King's.

However, in neither case do we really need to treat the line to transmission line theory, although of course that's what it is. Now let's look at the instance of trying to send 2MHz down a screened line 15m long, where the only known quantities are that (a) it's aircraft grade, and (b) the capacitance of the cable.  We can ignore (a), on the basis that the electrons don't know that they're at ground level.

2MHz corresponds to a wavelength of 150m in free space, so rather less than that in screened cable.  Someone upthread said that you need to start thinking transmission line theory if the line length approaches one-tenth of a wavelength (not the figure I would use myself, but never mind).  Mmm, 150/10 = 15.

[Cardigan]

Twisted pair is likely to have a rather greater capacitance per unit length than RG58, I would suggest.  You only have to double the figure you quote (and I seem to recall rather greater numbers being quoted upthread) to get the 3dB point down to 13kHz.  I suggest this would give noticeable treble loss, even if the clean original wasn't available for an AB comparison.

It was noticeable; it was a PC producing a frequency sweep to 20k. It was checked locally and via the long run of wiring, and the treble loss was significant.

[Mark Hennessy]

With RG58 at 58pF/m - let's call it 60pF...

Twisted pair is likely to have a rather greater capacitance per unit length than RG58, I would suggest.

Yes, somehow I missed that Cardigan said twisted pair. My bad

Now let's look at the instance of trying to send 2MHz down a screened line 15m long, where the only known quantities are that (a) it's aircraft grade, and (b) the capacitance of the cable.  We can ignore (a), on the basis that the electrons don't know that they're at ground level.

Just by way of a "heads-up", please allow me to point out that because you're new to this forum - and therefore we haven't yet got to know you - it's hard to be sure whether this last sentence is humour or sarcasm. As written, it could be taken either way. With a smiley at the end, I'd certainly chuckle

2MHz corresponds to a wavelength of 150m in free space, so rather less than that in screened cable.  Someone upthread said that you need to start thinking transmission line theory if the line length approaches one-tenth of a wavelength (not the figure I would use myself, but never mind).  Mmm, 150/10 = 15.

Well if Trevor's tie-line gives him grief at 2MHz, I've no doubt that Trevor will know full well that he needs to terminate it. Certainly, we've established that it's absolutely fine for audio...

So if not 1/10, then what's your preference, and why? Some say 1/4. Interested in your thoughts...

[Murphyv310]

Good Morning.
I'm off shortly for more tests regarding my health. Later on though I'll do some more tests and report back here. I'll even photograph the cable and it's stripped ends.
Up to the time I ran the cable into the house I have only used it for very short patch cables at Audio frequencies, its excellent screening properties and robustness were the main reasons for using it, plus I've plenty of it.