Monoblocs - relative length of cables, interconnect vs speaker

[tjobbins]

Hi guys,

The manual for the Tag 125M lists as one of its advantages the fact that the monoblocs can be placed very close to the speaker it is driving, thus allowing for a very short length of speaker cable. This would then require longer interconnects from the pre-amp to the monobloc.

I was wondering if everyone follows this design, or whether there are advantages to reversing the lengths - having short interconnects and long speaker cable.

Intuitively it seems to me that it is better to minimise the length of the interconnect: the interconnect carries a low voltage signal, so any interference added at this level will then be multiplied by the power amp. This is not true if the same amount of interference is added to a speaker cable.

Therefore interference on the interconnect will have a greater proportional effect on the signal than intereference on the speaker cable.

I don't know if this argument can be expressed in simple numerical terms, but:
if we assume the interconnect carries a signal level of '10', which is amplified 100 times to '1000', then adding an interference signal of level '1' results in a 10% effect if applied to the line-level cable versus only a 0.1% effect if applied to the speaker cable.

Any of that make sense?


Tom

[bushy30]

Hi Tom,

Don't let Udo hear you say that LOL

The manual is correct ! I know this from doing PA work, always keep the power next to the speakers, the signal goes miles to the mixing desk and back in comparison !

I can't remember the exact technical reasons, but if you look at recent posts in the TAG owners forum, Udo actually explains it !

Cheers

Bushy

[tjobbins]

Originally Posted by bushy30

Hi Tom,

Don't let Udo hear you say that LOL

The manual is correct ! I know this from doing PA work, always keep the power next to the speakers, the signal goes miles to the mixing desk and back in comparison !

I can't remember the exact technical reasons, but if you look at recent posts in the TAG owners forum, Udo actually explains it !

Cheers

Bushy

Hehe, thanks Bushy. I certainly did not mean to impune Tag's recommendations! I was more wondering why this is the case, as it seems counter-intuitive to me. But then I am probably missing something important!

[bushy30]

No worries,

Graham, Steve et al will probably be able to answer it too !

Cheers

Bushy

[pjskel]

It boils down basically to the fact that nearly all interconnects have a screen to mop up all manner of RF and EMI, whereas the speaker cables tend not to have any screening. Mind you, I've had plenty good results from longer lengths of Purist Audio speaker cables.
A lot will depend ultimately on the construction of the cables being employed, and if stereotypical in design, longer line level connections won't unduly affect the signal, providing the screen has been properly gounded.
Used to love the Furukawa cables of yesteryear with their flying grounding leads attached. Never got to try a full system's worth of it sadly.

[johnson]

Hi
Also more recent TMA monoblocks have balanced inputs to negate the influence of noise.
Regards

[Kenny Glasgow]

Tom

Get yourself a good set of interconnects from Mark Grant in the Power Buys of the cable section "above". I can't recommend them highly enough

I've always wondered why people have long runs to their subs via interconnect and the have long runs of speaker cable to their speakers

[Dr Udo Zucker]

Originally Posted by tjobbins

Hi guys,

The manual for the Tag 125M lists as one of its advantages the fact that the monoblocs can be placed very close to the speaker it is driving, thus allowing for a very short length of speaker cable. This would then require longer interconnects from the pre-amp to the monobloc.

I was wondering if everyone follows this design, or whether there are advantages to reversing the lengths - having short interconnects and long speaker cable.

Intuitively it seems to me that it is better to minimise the length of the interconnect: the interconnect carries a low voltage signal, so any interference added at this level will then be multiplied by the power amp. This is not true if the same amount of interference is added to a speaker cable.

Therefore interference on the interconnect will have a greater proportional effect on the signal than intereference on the speaker cable.

I don't know if this argument can be expressed in simple numerical terms, but:
if we assume the interconnect carries a signal level of '10', which is amplified 100 times to '1000', then adding an interference signal of level '1' results in a 10% effect if applied to the line-level cable versus only a 0.1% effect if applied to the speaker cable.

Any of that make sense?


Tom

Hi Tom, when it comes to signal impurities then there is more to it than signal amplitude. We need to look at the way interference can "attach" to the signal and ways we can prevent interference (addition) and signal degradation (distortion). Let’s do this:



Looking at the connection from a preamplifier to a power amplifier, we have, as you rightly say a small signal, but actually not that small. Typically there might be a voltage between say 1V and 2V. The purity of that signal has to be maintained, no doubt, hence it is output (in most designs) by a low impedance pre-amplification stage, then “carried” via the cable before fed into a high impedance input stage at the power amplifier. That way (high mpedance input) we have almost no current flow, hence the cable’s resistance is of no real interest and almost more importantly we can now safely screen the connection cable without detrimental effect to the signal. Now, screened and voltage based, it is very difficult to interfere with or distort the signal whilst it “carries along the cable”. However, it is getting better: The low output impedance at the preamplifier prevents that any potentially interfering signal has a real chance to add significantly to the signal. We therefore can safely say that the chosen design allows the “clean” transport of the audio signal over a long distance between preamplifier and amplifier and it will be (almost) only mains hum issues we might suffer from due to the reference grounds being a long distance apart and therefore often different. Of course that effect can be eliminated by using a balanced input stage at the power amplifier end.



The situation at the power amplifier’s output is different. Here the issue is less with noise (as the signal is large- didn’t you say it!) and the amplifier’s output is really, really low impedance. However, we have to make sure that the power amplifier is driving the speaker and partially the speaker cable. Therefore we have to prevent inductance (L), capacitance (C) and resistance (R). Well, the best way to achieve this is to remove the cable all together: L=C=R=Zero. If that isn’t possible we need to minimise each parameter. We could make the cable “thicker” to reduce R; we do not screen the cable to drop C (we accept the added noise) but reducing L isn’t easy and most frequently adds C. Of course speaker cables can be constructed to minimise these parameters for any given length, but no measure taken/money spent will have the same effect as simply reducing the cable’s length. Therefore, why spending so much time/money designing a long speaker cable, if we simply have to shorten it to get an even better result? The shorter, the better- it is simply that.



Personally I replaced a 2.5 m ultra-expensive Transparent Reference XL SS cable, which uses passive filters, with a 30cm TAG McLaren speaker cable. Now I have the “ultimate” speaker cable due to minimal length as little cost, whilst yet another two pieces of audio experience are “dusting” along in my cupboard.



I hope above helps, although purists of the engineering society might say that I should have used a more scientific language.



P.S.: Above is only true for intrinsically stable amplifiers. If that isn’t the case the speaker cable’s L is essential to stabilise the amplifier and hence such amplifier usually requires a very specific cable as otherwise it will self-destroy or oscillate. To sell amplifiers which are not intrinsically stable, whatever the their load, isn’t a good move in my opinion!


Typo, spotted by Steve, has been corrected. Sorry for any confusion and my thanks to Steve!

[bushy30]

Hey Tom !

Hate to say I told you so !

Thanks Udo, I'll try and keep all that in my pea brain for next time !

It's great to be able to discuss information like this, now I'm just wondering how I can get my 700:7R 30cm away from all of my 5 speakers !!!!!

Bushy

[Dr Udo Zucker]

Originally Posted by bushy30

...It's great to be able to discuss information like this, now I'm just wondering how I can get my 700:7R 30cm away from all of my 5 speakers !!!!!
Bushy

Hi Bushy, it isn't difficult to place 5 speakers around a 700:7r using short cables, but rather an achievement to find a suitable listening position then. Maybe a good case for TMREQ (or is it ALREQ today)

[Stevesky]

The purity of that signal has to be maintained, no doubt, hence it is output (in most designs) by a high impedance pre-amplification stage, then “carried” via the cable before fed into a low impedance input stage at the power amplifier.

I think there is a typo here as this is mixed around. The preamp output stage will have a relatively low output impedence, while the power amp input stage should have a relatively high impedence.

For example an AV32R output impedence is 47ohms, while a 250MR input impedence is 25K Ohms.

[bushy30]

WOW ALREQ !

Maybe if I could pay John to do a software revamp.............



Bushy

[tjobbins]

Originally Posted by Dr Udo Zucker

I hope above helps, although purists of the engineering society might say that I should have used a more scientific language.

Absolutely it does, thank you! I do like to understand why something is the way it is - sometimes it has a practical benefit, othertimes just out of pure interest

I am going to get the Mark Grant interconnects, place my amps at the base of my speaker stands, and connect them using 0.75m lengths of multiple CAT5 cables, home-made (Learnt that trick over at the Cables & Cabling forum!)

Thanks again everyone, particularly Udo!

[Miron]

Originally Posted by Stevesky

I think there is a typo here as this is mixed around. The preamp output stage will have a relatively low output impedence, while the power amp input stage should have a relatively high impedence.

For example an AV32R output impedence is 47ohms, while a 250MR input impedence is 25K Ohms.

Steve, he is simply enjoying portuguese sun (and everything what comes on the top) for far too long

[Dr Udo Zucker]

Originally Posted by Miron

Steve, he is simply enjoying portuguese sun (and everything what comes on the top) for far too long

Hi Miron, not really - it was rather my 18month old son who came into my office with the clear intention to pull every book from the sheves and to push every CD ROM onto the floor