Speaker cables: asked 100 times and never fully answered

Several Orders of magnitude, as in multiples of 10? Are you sure?

The question of suitability / stability is complicated by the frequencies of the music and the behaviour of the load (reaction of the crossovers / enclosure acoustics etc.) As for compatibility, I don't know (and why should I anyway?) What always strikes me is the lack of urge for audiophiles to change the cable runs to the speakers within the speakers or amplifier. The crossover to speaker wire in one of my speakers is over 1.5m, a fair proportion of the total wire length between transistor and drive unit.

Good spot. It's about 0.5 of 1 order of magnitude.

A man goes up to the bar and asks the bar person for five pints of bitter, six glasses of red wine, a lemonade and a tonic water. 

'That's a large order' says the bar person

The next man goes up to the bar and says 'I'd like ten times what he just had'.

'Now, that's an order of magnitude'.

You won't get an answer from Naim on this. Their advice is clear - use their own cables. 

Say they specified particular characteristics of a cable that would be OK. A new cable comes to market with those characteristics but assembled in different way, and the Naim amp is destroyed. The user can then demand a new amp because it met the parameters and can therefore prove that the amp is faulty, when in fact that the fault lies with the cable's construction. 

My law and knowledge of electronics may be suspect here, but I'm sure you get the idea. 

Why are people worried, almost to the point of obsessiveness, about this?

Because almost no other manufacturer stipulates this and Naim do not actually say that you cannot use a non Naim cable or even that it invalidates the warranty. They just don't recommend it.

Their statement clearly states you can use other cables and provides the specs of A5 for us to make an informed decision. The point I am making is that the cable specs don't let you make an informed decision; the tolerance range on the amps does and that is what we are not given. Statements like low and high capacitance don't mean anything - it is not a value. It is like saying it should be a long piece of string. Similar to A5 is also not a value. What is similar? +/- 1 milli ohm? +/- 10?

As to the other point, I see what you are saying but if a cable had those properties then it has those properties. If construction caused an issue that would be something that changes those properties so would be a moot point. For example, if the wire has inductance of 1uF/m and the plugs on the end change it to 20uF/x (just a number from thin air), then it's no longer a 1uF/m cable.

This has always seemed straightforward to me if you want to hear your amps as their designer heard them. But I realise that is not what you are asking about.

Chris

Question is, why would you want that? I surely don't care, I spend my money in order to hear music in the way I like.

The way I see it is this: Naim designed their amps in a specific way in a way that a cable with extreme measurements can cause issues. That's a design choice. At the same time, Naim is fully aware that people will use other types of cables, so they just cannot permit themselves to design an amp with small tolerances towards cables. Because even if you warn people for this situation, the the effect will be a load of malfunctioning amps and worse, massive damage to your companies reputation. "Told you so"will not work." Btw, this negative impact far outperforms the extra benefits for Naim selling cables.  

So you can be sure that while it's better to stay well within the boundaries, it's unlikely that you will hit the boundaries by using other cables. Never saw a post that somebody blew up his amp by using other cables on this forum also.  

Black / white statement: the "better safe than sorry people" and the "hear the music as meant by Naim" will stick with NACA5, The rest will try / use other cables. 

I tried NACA5, didn't like it. Not bad, but other options suited my taste much better. And that's where I spend my money. My taste, my preferences, my money. 

Fair enough, people differ.

C.

Naim's manual for its current integrateds simply states "Naim speaker cable will provide the best results, however, a wide range of speaker cable types can be used without risk of damage to the amplifier''. There's nothing there about avoiding high capacitance cables.

In their manual for power amps however, I found for the 250.2 that Naim advise against using other cables as damage can result: 

"Only Naim Audio speaker cable should be used to connect a loudspeaker to the output of the NAP 250. Custom Naim Audio loudspeaker connectors are supplied to make the connection and in order to comply with current European safety regulations these should always be used. DAMAGE TO THE AMPLIFIER MAY RESULT IF “HIGH DEFINITION” CABLE OR ANY OTHER SPECIAL CABLE IS USED TO CONNECT THE LOUDSPEAKER''

Naim don't give fixed rules as the conditions of use are variable.

It depends on the cables, the speakers and even the cable layout...

The length of the cables: As the cable is in series with the speaker, when using short cables are used, the reactive load that the amp will 'see' is dominated less by the cables reactance and more by the speaker's reactance.

The speakers in use:  If the cable isn't long enough to completely dominate the load (it rarely is!), the load on the amp is a combination of cable and speaker reactance.

Cable layout:  The physical arrangement of the cable (coils, proximity to metal connected to earth etc) can influence the actual installed capacitance and inductance of a cable.

Ergo Naim can't always specify simple values for the cable's characteristics.

Zen,

What the hell?   Naim recommend Naca it couldn't be simpler.  I use another cable which predates  my purchase of Naim amps.  But before using it I simply asked my Naim dealer f it would be okay.

Regards,

Lindsay

Naim were selling their amps way before they ever got around to selling their cables.

c.1980, when I bought my Isobariks from Naim, JV simply recommended connecting those to my 250 using a good quality 15 amp flex from an electrical supplier. 79 strand "Monster" cable would have done the job just as well. He did make it clear that Lintz (?) type cables might cause instability so not to use them.

I don;t believe that since then Naim have altered their 250 lineage design to the extent that they are super-sensitive to the speaker cables or loudspeakers.

Of course Naim would like us all to buy their cables and their speakers, but that just ain't ever going to happen. So their best advice will be to avoid certain characteristics in cables and aim for a cable somewhat similar to "their ideal", whether that be NACA5 or Super Lumina.

At the start of the 1980s there were just two options - standard side-by-side cables, or the (then) new-fangled Litz (many thin woven strands) cables.

It was simple: Don't use the Litz cables.

Then the cable manufacturers got clever with the marketing tricks and made cables from a small number of strands woven together (e.g. Kimber cables).  Although non Litz construction, these looser woven cables are still not suitable for classic amps (and somewhat marginal even for the black integrateds, but can still be OK with short runs and some easy speaker loads).

So, how would you want Naim to express this information for these cables and other non side-by-side designs?

Do you expect them to test every cable, cable length and speaker combination?

Should they make a general statement of principle (as they currently do)?

Hi Philip,  you are right the speaker has a much bigger effect on amplifier load & especially its output stage stability.  Solid state amplifier output stages become unstable due to the ‘speakers capacitive reactance. This can be minor parasitic ringing up to full oscillation and although the problem occurs well above the audio band it will destabilise the amplifier throughout its whole operating range. Amplifiers have a Zobel network across the output stage to damp this oscillation. However with the variable capacitive reactance of the speaker load coupled to a length of speaker cable the Zobel circuit alone may be ineffective. This is commonly solved by placing an inductor in series with the amplifier output downstream from the Zobel shunt circuit,  the inductor value is set low enough to avoid high frequency roll-off in the audio range but high enough to provide the required damping at frequencies above that.  Some amplifier manufacturers such as Naim do not have an internal inductor and instead use the speaker cable to provide inductance.

Think of the cable inductance (or the amp internal inductor) as a barrier to damp (block) the effects on the amplifier output stage from the speakers capacitive reactance.    

I think some people are missing the point.

• No I don't expect Naim to test any cable. That is why the purpose of this post is to turn the argument on it's head and ask what the ideal load parameters of the power amps are, not the cable.
• Yes indeed, the speaker itself and the arrangement of the cable changes everything. I 100% agree, which is why again, it makes more sense to understand the issue from perspective of the amp, not the cable.
• Non litz designs that also present problems present those problems for the same reasons; ranges of Capacitance, Resistance, or Inductance (usually Capacitance for litz) that is non conformant with Naim's design goal. But again, this is fairly meaningless.

The cable is one part of an equation to be satisfied with the cable and the speaker on one side and with the ideal load characteristics of the power amp on the other. So far all discussions focus on the cable but all we really need is to understand is the problem Naim are trying to address with NACA5 (parameters of the amps). Then we can balance the other side. Even a litz cable could be perfect I suppose if it's properties, over a certain length with a certain speaker fell within the desired range.

I'm also not suggesting for a moment that we choose cables/speakers/flavours of ice cream based on technical values. I'm trying to get a baseline so that we can merely rule out problematic components from our decision making process. The cables I can safely choose with 5m runs to my PMCs might be totally different from Johnny Moneybag's scenario that needs 100m runs to an JBL Paragon, or Kenny Collegeboy and his 1m runs to clapped out MIssion 751s. In those scenarios NACA5 itself may be very unsuitable.

Anyway, I have posed the question to Naim. Let's see what happens. My money is riding with all of you who have stated that Naim will probably decline to answer this with anything other than "Use A5 or SL". I hope I am wrong but it doesn't hurt to ask.

Ideal load, easy:

3.5μH and 8Ω in series

Other than that, how exactly would you like to see Naim express the load requirements?  (I'm not expecting actual figures, just an explanation of the form in which you'd like to see it expressed and an example layout).

When you try to express the limitations in terms of reactive behaviour, you'll find that it quickly becomes complex (), and hard for anyone but electronic engineers to understand. (If you didn't understand the wink, then you're certainly not there!)

It'll also be completely useless, as few loudspeaker manufacturers provide enough detail of the electrical (and electro-mechanical) characteristics of their speakers for anyone to be able to calculate if a particular cable / speaker combination is compliant.

Huge you are spot on.

If addressed it needs to be within loose enough guidelines but with enough info to be useful. For example, it would need to be clear that the values represent a total of loudspeaker and cable together.

Optimal range

Resistance: X Ohms - Y Ohms 

Capacitance: X pF - Y pF

Inductance: X uH - YuH

Qualified operational limits

Resistance: X Ohms - Y Ohms 

Capacitance: X pF - Y pF

Inductance: X uH - YuH

Or, alternatively, to come clean and say

"The load and electrical characteristics of any given cable/speaker pairing is complex and difficult to accurately ascertain without in-depth electrical knowledge and measurement apparatus. As such we advise customers to use whichever sodding cable the like because it makes no difference to the average punter and strange lengths of our own cable or any of thousand types of loudspeaker on the market make predicting a load by you, the consumer, entirely pointless."

Unfortunately that doesn't work, ...

Even with values within those that are appropriate to specify for 'qualified operational limits', then, if you have most of the capacitance in the cable and most of the inductance is in the speakers, there's a good chance the amp will overheat and/or go bang, and at the very least will have a shortened operating life.

It's not just the total values but where they are in the circuit.

For the 'Classic series' you could say

Minimum 3.5μH in series with 2Ω minimum load (reactive) and first cable resonant frequency >6MHz

i.e. equivalent to 3.5M NAC A5 (plus a small safety margin on the resonant frequency).

Mr Zen,

I wouldn't worry too much. The Naim's don't have an output inductor but (ISTR) they do have the rest of the Zobel network (but they do have the small paralle capacitor and series resistor). As Mike says you normally have a small series inductor (2 or 3 microhenries), this is typically to stabilise the amplifier into a capacitive load.

So, if your speaker cable has a few microhenries of inductance you're fine. Litz cables (and similar) have lots of capacitance and little inductance and so you need loads (off the top of my head a 100 metres or so) of it to provide the desired inductance (take a look at Kimber 4TC for an example).

The Naims are reasonably load tolerant (albeit not a powerhouse design in my view) and (IMO) the design of the 250 etc is inherently tailored toward active operation as capacitance (again off the top my head) is more a crossover characteristic than a drive unit characteristic - hence the partial Zobel and some 'breathed on' mains cable for speaker cable works well in that situation.

Beyond that, don't worry, if it really was that much of an issue a variety of speaker would break the amps on a regular basis  or just make them sound very poor and the amplifier run rather hot. If it isn't doing that, it's working just fine.

I have never heard from or of anyone who has damaged a naim amp by using non-naim cables. I had a nait3 for years and never used naim speaker cables and no damage resulted and it sounded great.

I would also tend to take a practical-based approach to this.  If a cable actually sounds good, and the amp doesn't obviously start to, say, overheat when using it, then I'd take that as a conclusive indicator that it works fine with Naim.  If it doesn't sound good, then forget about it anyway.

Obviously this rule of thumb is only of use or help when cables can be trialled.....

Naim's service dept. has had plenty through where using the incorrect cable has caused instability and thermal damage.  Also bear in mind that an amp running on the edge of instability requires service attention way sooner than usual.

I remember trying some Kimber 8TC on my NAP250.2. 8TC is very high capacitance and low inductance - far from ideal for Naim amps.  The amp sounded nothing special and got hot quickly - not a happy amp.  Quickly removed.  8TC worked much better on a big valve amp.