Amp placement short interconnect v short speaker cable?

[Markiii]

So assuming an individual amp per speaker to make life easy physically

Everything else being equal

What's likely to better from an sq perspective? Short interconnects from Pre-amp to power amp or short speaker cables from power amp to speakers?


Thoughts appreciated

[YNWaN]

It depemds on the amps and the pre-amp.

[The Grand Wazoo]

I think the generally accepted wisdom is that short speaker cables are best. However, that pre-supposes that your pre-amp has sufficient whoomptie to work with a long set of interconnects.

[DSJR]

It's worth checking with the preamp manufacturer to see if it's ok to drive long 5m plus interconnects as stated above and in any case, keep the speaker leads as heavy in gauge as possible, however long or short they are.

[Markiii]

I assume the theory is the shorter the speaker wire the lower the impedance? The easier time the amp has?

[Barry]

As short a speaker lead as possible. Provided, as others have said, you pre-amp can drive long interconnects.

An ideal situation would be to use a preamp having a balanced output with monoblock power amplifiers having a balanced input. In my system the speaker leads are < 1m long, and the balanced interconnects between the pre-amp and each monoblock are 12m long.

[MikeMusic]

Might tbe Naim or Linn specific ... I remember some advice being you must have a minimum 2.5 metres speaker cable length, could be Russ Andrews...

[Andrew B]

Might tbe Naim or Linn specific ... I remember some advice being you must have a minimum 2.5 metres speaker cable length, could be Russ Andrews...

Yes, Naim amps require low capacitance but also need relatively high inductance. That's why you shouldn't use short or uneven lengths with them. IIRC they recommend minimum 3.5m lengths of NAC A5

[hifi_dave]

As all the above posts. It's swings and roundabouts and the small differences are a matter of taste - if you can hear the differences.

Few amps are going to be bothered by a sensible run of speaker cable but many pre-amps are unhappy driving long runs of incon.

[Mr.Ian]

The designer of my valve amplifiers recommended short interconnects and long speaker cables.

Personally I though top end suffered very slightly, slightly duller in this set up relative to long interconnects (10M) and short speaker cables (1m) but it could easily be down to the change in speaker cables themselves (not that I am really a believer). The 1m ones used were Vecteur CV30 ( now there's a blast from the past - actually my speakers use it as std internal wiring) rather than the Belden at 10M

[hifi_dave]

The cables need to be the same for the two situations. The only variable should be the lengths.

[wee tee cee]

I run long interconnects one side 2m the other side 5m, Mark Grant will do any custom combos you require(previously did this with unequal length bi-amp cables and had four mono blocks in the cellar again, Mark Grant cables) to circumvent a fire place by going under it through the cellar.
I use temple audio mono blocks that are small enough and light enough to velcro( b&q double sided sticky velcro tape) to the back of each speaker.

One set of speakers is Spendor sp1s the other Quad 63s. It then lets me run speaker cable lengths, the length of bi amp jumpers (gives a bit of scope for trying exotic speaker wire that would otherwise be out of financial reach-most cable companies do pre terminated jumpers at achievable prices)

The temple mono blocks can be set at different impedance levels to tailor them to different speaker loads.
The pre amp is a passive Tisbury audio so really it just depends on the output of your source most DACs and CD players have worked fine in my set up....even a tiny Cowon portable was able to drive enough juice through the pre to the speakers.

Dont think a turn table would work without a phono amp in play though.

Initial worries about vibrations from the speakers effecting the amps haven't manifested.

[Barry]

The rules are not hard and fast - it all depends somewhat on both the pre-amp and the power amp.

Any interconnect will have capacitance per unit length. This capacitance, along with the input capacitance of the power amp, will form a low pass filter with the output impedance of the pre-amp. It is not difficult to show that the roll off frequency (-3dB point) is given by:


F= 1/(2π*C*R),

where R is the output impedance of the pre-amp and C the total capacitance. The value of R depends on the circuit topology. For a valve pre-amp it might be as high as 5KΩ. If we set the -3dB point at 20KHz, this makes C = 1600pF. Assuming a power amp input capacitance of < 200pF, then with an interconnect having a typical capacitance per unit length of 100pF/m, the maximum interconnect cable length is 14m.

Solid-state pre-amps might have R laying between 100 - 1000Ω, with some having R as low as 10Ω.



The length of the speaker cables to some extent depends on the output impedance of the power amp, and if the power amp is fitted with a Zobel network.

Naim amps do not use a Zobel network, nor have any series inductor at the output. They will suffer ultrasonic oscillation if used with cables having a low loop inductance/unit length. They recommend using 3.5m of NAC –A5 cable; providing 3.5μH. Effectively, the Naim amp circuit’s output terminations are moved 3.5m closer to the speaker by this method. In an average set up with around 4m of speaker cable the Naim system will work as if the speakers were directly connected to the amp with no speaker wire!

Solid-state power amps have an output impedance of < 0.1Ω. 5m of cable will increase this by ~ 0.1Ω so roughly, at least halving the damping factor.

Valve power amps have a higher output impedance ~ 1Ω, so long speaker cables will have much less of an effect on the (already lower) damping factor.

[Marco]

Happy 7000th post, Bazza! :)

As for cable lengths, always see cables (and connectors) as a necessary evil, and so keep cable looms as short as possible and minimise breaks in the signal path, hard-wiring instead at every practical opportunity.

Marco.