advertisement


Speaker cable technicalities and related stuff

Arkless Electronics said:
Aha... I didn't know anyone had done it already...

It would be a bit more inductance than the 5 turns on a pencil and in fact be wound on high power carbon resistors to provide damping.


Jim. You wouldn't be thinking of a certain RF feeder there with that impedance figure by any chance?
I'd go for a rather lower figure myself... let's say 50R...
.
Click to expand...

I was thinking of winding onto something like a big polymer bolt or length of studding, then taping it down. Gives a more even spacing with a gap to reduce the self capacitance and prevent movement. But a simple DIY job.

I chose a value purely for 'stirring' purposes. :)

You *could* go for 10 or 8 Ohms, but that did get some people into trouble in the past... 8-]
 
MJS said:
I did rig an inductor a few years ago by cutting a notch out of a length of NACA5 and soldering in the inductor and resistor. Coated with heatshrink it was a bit Heath Robinson but it proved a point and stopped a very hot 250 from misbehaving. I've yet to turn it into a product because we have absolutely no idea if it'd sell or not. I don't think you could guarantee that it would enable the owner to use any type of cable and maintain stability. Maybe it could be sold as an RF choke to limit interference, as I've successfully used this cure here too.
Click to expand...

The above comment about it not necessarily being a guarantee is what would make me hesitate. Don't know enough about the behaviour of the amps in question to know what load impedance/frequency region needs avoiding. More than a simple inductance might not be 100% OK and someone might be unlucky. I suspect it is OK, but we'd need relevant test measurements to tell.
 
nobeone said:
As I understand it Naim made and sold amps like that before they sold cables, but sold through dealers that advised and supplied suitable cable, so in a system context nothing was unstable.
.
Click to expand...

IIRC the problem cropped up when the cables designed to have a characteristic impedance near to 8 Ohms appeared. Result of work by Hiraga and others I think.

These tend to have high capacitance/length and low inductance/length. Again IIRC this was when I encountered the effect when a Naim dealer loaned me an amp on the QT to examine and comment on, and I found some cables and loads let it oscilliate.

The Naim cable idea was then sold on the basis of 'sound', but ensured a capacitance and inductance that acted as the network to protect against instability. Again IIRC and from what other people told me at the time.
 
Well NVA majored in Richard’s day on minimalism, supposedly the best component being no component at all.
My A80s do sing, and with their music I am very happy, though I wasn’t aware they were also fragile in the output sense.
Thanks everyone for an interesting discussion.
 
chartz said:
Okay, so half marketing ploy, half science then. Still, silly practice when any old cable would have worked fine.
Was it exposure that did the same?
Click to expand...

No, it's just really bad engineering! There is no advantage in relying on the cable's inductance rather than using an internal inductor.
Naim use an 0r22 resistor at the output which is also there mainly for stability purposes by giving some isolation between the amp output and a capacitive load in the same way as an inductor but which limits damping factor to 16 with a 4r speaker. A few other companies also do this but most use less resistance, 0r1 or 0r05. It would seem that in the case of Naim it is not enough on its own to ensure stability.
 
Oh and as I have to keep banging on about NFB.... this is yet another case where the whole matter being discussed is due to the NFB!
 
For one who supposedly majored on minimalism, Dr. Bunsen Honeydew certainly offered a lot of loudspeaker cable variants.
 
Arkless Electronics said:
No, it's just really bad engineering! There is no advantage in relying on the cable's inductance rather than using an internal inductor.
Naim use an 0r22 resistor at the output which is also there mainly for stability purposes by giving some isolation between the amp output and a capacitive load in the same way as an inductor but which limits damping factor to 16 with a 4r speaker. A few other companies also do this but most use less resistance, 0r1 or 0r05. It would seem that in the case of Naim it is not enough on its own to ensure stability.
Click to expand...

Alternatively, the 0.2 Ohms may interact with some speaker impedances to tweak the overall frequency response in a way that makes the sound slightly 'different' and 'preferred' by some. :)
 
Jim Audiomisc said:
Alternatively, the 0.2 Ohms may interact with some speaker impedances to tweak the overall frequency response in a way that makes the sound slightly 'different' and 'preferred' by some. :)
Click to expand...

Indeed but in a pretty random way ie different for each speaker...

FWIW I've tried on several occasions adding such series resistors to reduce the damping factor and basically see what happens subjectively and its always had a negative effect. It can sometimes seem a little "warmer" and "less SS sounding" at first but I soon noticed a general muddying and reduction in transparency and focus.

The Avondale boards for Naim amps lose the series 0r22 resistor and add the missing inductor and in doing so puts right a couple of the most glaring problems with the original boards... quite possibly these changes alone are responsible for much of the upgrade that many seem to report from the Avondale boards.
 
Arkless Electronics said:
Indeed but in a pretty random way ie different for each speaker...
Click to expand...

Yes, Which may help explain why reviewers found 'magic' combinations of amp and specific speakers. The Isobarik being the classic example once you know how its input impedance varies with frequency, etc. However given that most of the reviewers couldn't twig the cause this remained as 'magic'... :)
 
Jim Audiomisc said:
However given that most of the reviewers couldn't twig the cause this remained as 'magic'... :)
Click to expand...

A corollary of Clarke's third law, "Any sufficiently advanced technology is indistinguishable from magic" is "If you don't understand the technology, it will all seem like magic to you".
 
Hi, everyone,

I'm the new owner of NVA - I purchased the remaining assets in August '19.

Just for clarity - it's absolutely fine to criticise our policy on speaker cables, our design ethos, or anything else. It's the internet, we are all entitled to our opinions and beliefs. It would be pretty boring if we all held the same values and if every bit of kit were designed in the same way. No offence will be taken.

I removed a couple of Ross's posts at HFS because they were repeating the same question over and over when a response had already been given. I suspended Ross for 30 days from HFS because he deleted / amended HFS Admin posts. The rules of the forum are the rules of the forum. Nothing personal.

Regarding the policy, an earlier post (Alan Brown) pretty much summed it up:

"I always found the NVA LS cables was great but Richard quickly tired of talking about alternatives. The truth is many alternatives will be fine but you need to understand the reasons NVA amps require certain things in order to choose. The upshot from NVAs point of view was it was better to adopt 'NVA cables only - or no warranty' as a policy rather than constantly explain things to punters (who then went and bought something inappropriate by mistake, and needed to get their amplifier fixed). I happily used the NVA cables and then Colin Wonfor designed cables (very low capacitance also) without any trouble at all. But personally I'd just use the NVA stuff."

NVA isn't going to publish a list of alternative cables. Ross was pointed toward the archive and advised that, ultimately, it is his decision. We simply don't have the time to monitor every manufacturer's cables to check if they are still suitable...we are building orders and preparing to exhibit at shows. And, yes, of course we want to sell our own cables.

I hope this is helpful.

Ross if you PM me your postal address, I will send you a free set of NVA cables (not the expensive ones, though ;)).

Best regards and Happy New Year to everyone,

Paul
 
Jim Audiomisc said:
Yes, Which may help explain why reviewers found 'magic' combinations of amp and specific speakers. The Isobarik being the classic example once you know how its input impedance varies with frequency, etc. However given that most of the reviewers couldn't twig the cause this remained as 'magic'... :)
Click to expand...

Tenuous :)
 
Jim Audiomisc said:
IIRC the problem cropped up when the cables designed to have a characteristic impedance near to 8 Ohms appeared. Result of work by Hiraga and others I think.

These tend to have high capacitance/length and low inductance/length. ...
Click to expand...
It isn't "tend". My 10 Ohm semi-rigid coax looks very strange. The inner conductor is almost the same diameter as the solid outer, capacitance should be about 150pF/m
This is specialist stuff for impedance matching RF power transistors
 
Vertical mosfet outputs are perfectly capable of oscillating at VHF frequencies where the cable is a transmission line.
Worrying about impedance matching at 20 kHz is daft.
 
Arkless Electronics said:
If your speaker cable is so long you worry about transmission line effects then you have other problems!!
Click to expand...
davidsrsb said:
… Worrying about impedance matching at 20 kHz is daft.
Click to expand...
Max Townshend seems to disagree. In a paper I picked up at the Bristol show last year I read:
"Critics insist that transmission line behaviour only occurs in 40km long cables at RF, but this is disproved by the simulation when distributed behaviour is analysed and shown in the blue traces above."​

The simulations do indeed show reflections and their effects from just 10 metres of "normal" cables - much reduced when using Townshend's Isolda loudspeaker cable.

The paper seems to be technically OK but I am sceptical about whether the simulation addresses reality. Its figures seem to show the audio amplifier producing a step output needing, I estimate, in excess of 50 megahertz bandwidth [1]. I am not sure how many real audio amplifiers can do that :rolleyes:. But if you have one then Max has a loudspeaker cable for you.

EDIT: re-checked the estimated amplifier bandwidth.

[1] The paper explains "A step waveform is utilised as it simulates all audio frequencies (and more) and is easily repeatable". "And more"? And how!
 
I have a lot of respect for Max Towshend as a mechanical engineer, but he certainly does not understand what he was measuring. Even Class D amplifiers cannot have fast rise times like that from the output switching or they would never pass CE approval tests
A lower test limit of 150kHz means that rise times cannot be faster than about 10uS
 
davidsrsb said:
It isn't "tend". My 10 Ohm semi-rigid coax looks very strange. The inner conductor is almost the same diameter as the solid outer, capacitance should be about 150pF/m
This is specialist stuff for impedance matching RF power transistors
Click to expand...

I'm making some allowance in case someone from Star Trek provides us with cables whose propagation speed is greater than 'c'. 8-]
 
You must log in or register to reply here.


advertisement


Back
Top