panomaniac said:
It kind of depends where you want to place the speaker cable in the circuit. If you consider it part of the amplifier, then it lowers the damping factor of the amp (because you can now consider the output impedance to be greater than the amplifiers original design spec.)
However, if you consider it to be part of the speaker, then it will affect the Q of the speaker. In the end, it would essentially be the same result, but two different ways of approaching it.
panomaniac said:
Panomaniac and others,
The qualitive observation is correct. It is when one looks at the quantative picture that the real situation becomes clear.
The equivalent diagram of any output circuit contains the loudspeaker as a generator with its complex generation of V and I, with a total resistance loading that, of the amplifier output impedance, cable impedance and loudspeaker (voice coil) resistance all in series. If we take the amplifier output impedance as negligible (folks are always boasting high damping factors) - or, let us say, for a DF of 40, as 200 m.ohm. Normal domestic cable resistances are of the order of another 200 m.ohm maximum if one does not use bell wire. The loudspeaker voice coil resistance for an 8 ohm system is about 5,6 ohm minimum.
This gives a total of 6 ohm resistance in the circuit. (We will ignore the reactive part of that, not to complicate matters - such would make the following argument worst case.)
The effect of the cable resistance mentioned above would have a worst case influence of 3,5% or 0,29dB. Now if anyone can prove that this would make a significant or even audible difference at all, I would like to see that. This, disregarding the order of magnitude larger variation in normal loudspeakers' impedances over the audio range, which makes cable parameters even less significant.
What some of us have tried to show for a long time now, is that if things are negligible, they are ... well, negligible.
Regards
Hey John- I think you're missing something. The added series resistance will attenuate the signal a little- but the main thing is that it will change the effective Qts.
That change in Qts will change the way the driver and the box behave. It will change F3 and it can change the frequency and size of any resonance peak in the box. It may have an effect on other things like pulse response, group delay, etc.
You should be able to hear those things - and much more easily than simple attenuation.
A 10 foot 24 gauge cable (Cat5 strand) will have a DCR of about 0.5 Ohms for both legs combined. A 12 gauge cable will have a DCR of about 0.064. That can do a lot to change the system Q.
However, if the same changes can be duplicated with a simple resistor in series, then who cares? Use the wire if the changes work for you, or a resistor, if you like. But remember attenuation is not the only change happening.
That change in Qts will change the way the driver and the box behave. It will change F3 and it can change the frequency and size of any resonance peak in the box. It may have an effect on other things like pulse response, group delay, etc.
You should be able to hear those things - and much more easily than simple attenuation.
A 10 foot 24 gauge cable (Cat5 strand) will have a DCR of about 0.5 Ohms for both legs combined. A 12 gauge cable will have a DCR of about 0.064. That can do a lot to change the system Q.
However, if the same changes can be duplicated with a simple resistor in series, then who cares? Use the wire if the changes work for you, or a resistor, if you like. But remember attenuation is not the only change happening.
Not to belabour this, Panomaniac -
but a cable series resistance of 0,064 ohms will affect the Q under 2% (24 gauge is a little thin for my liking). A system Q of 0,7 will become 0,714. You say this will have an audible effect, plus all the other things you mention? (No two loudspeakers of identical make are even this consistent.)
but a cable series resistance of 0,064 ohms will affect the Q under 2% (24 gauge is a little thin for my liking). A system Q of 0,7 will become 0,714. You say this will have an audible effect, plus all the other things you mention? (No two loudspeakers of identical make are even this consistent.)
Dumbass,
That is as I see it (but I have not had to do with this for over 2 years - still).
I understood from reputable loudspeaker manufacturers that characteristics of the same model loudspeaker can vary by up to 10% in production. The cable resistance quoted by Panomaniac will make a 1,14% difference. Then one disregarded amplifier output impedance, but which will also make a very small difference. (However, it is not constant in phase in many amplifiers, but let us not open that - it can be a can of snakes, not worms).
Anatech,
I would believe so. I am still not at home with my books and tables, but you could easily calculate the resistance increase. I understand that 100 degrees are not much for some drivers. The magnets can get hot.
Regards.
That is as I see it (but I have not had to do with this for over 2 years - still).
I understood from reputable loudspeaker manufacturers that characteristics of the same model loudspeaker can vary by up to 10% in production. The cable resistance quoted by Panomaniac will make a 1,14% difference. Then one disregarded amplifier output impedance, but which will also make a very small difference. (However, it is not constant in phase in many amplifiers, but let us not open that - it can be a can of snakes, not worms).
Anatech,
I would believe so. I am still not at home with my books and tables, but you could easily calculate the resistance increase. I understand that 100 degrees are not much for some drivers. The magnets can get hot.
Regards.
panomaniac said:
Panomaniac,
Can we assume, then, that the same changes to sound quality that we'd notice due to a change in Q of the speaker because of added series resistance (probably an amount less than the tolerances of DCR in inductors and drivers within the speaker itself), would be the same as what we would notice if we considered the series resistance to be part of the output impedance of the amplifier?
What I'm trying to get at, is what seems more significant: Lower damping factor due to the cable, or modified Q of the speaker due to the cable? It seems it can only be one or the other depending on what you consider "where the amplifier ends, and the speaker begins"....
I'd hazard to guess that it would seem less important if the DF of the amplifier was modified slightly, rather than considering that the speaker's Q has been changed? Would this, again, not be a case of the psychology of the matter, rather than the reality?
Johan Potgieter said:
Not the point. We are not swapping out speakers. We are swapping cables on the SAME speakers. So the differences in cable R should be heard, because the all other things stay the same. It may take a lot of R change to notice, but that happens. There are folks on this very thread who use 24 gauge wire, whether we like it or not.
I can make very small percentage changes to my speakers, stuffing, port size, port length, etc. and the differences are sometimes very large. Much larger than I think they should be. Other times, they don't seem to change much. So don't discount small changes in speaker Q - they may make a more noticeable difference than you think.
That said, the 10% difference in speakers will swamp the tiny differences in normal cables - between 2 sets of speakers. But when listening to the same speakers, there is no 10% difference, so the cable change will stand out.
Macgyver10,
Your question was to Panomaniac and I am not replying on his behalf - but I think one must not let this question of what is on the amplifier side and what with the loudspeaker get in the way. I understand the viewpoint, but current flows in a circuit, i.e. through all resistance (impedance) in the circuit, from any particular point all the way round back to that point, as it were. As such it does not make a difference what is where. I think part of the confusion (such as it is with various people) comes from this infernal definition of damping factor, which is a misnomer. (Loudpseaker impedance divided by amplifier output impedance is meaningless; those are not the only impedances in the circuit.)
Just wanted to re-establish that. (Now off to bed with me!)
Your question was to Panomaniac and I am not replying on his behalf - but I think one must not let this question of what is on the amplifier side and what with the loudspeaker get in the way. I understand the viewpoint, but current flows in a circuit, i.e. through all resistance (impedance) in the circuit, from any particular point all the way round back to that point, as it were. As such it does not make a difference what is where. I think part of the confusion (such as it is with various people) comes from this infernal definition of damping factor, which is a misnomer. (Loudpseaker impedance divided by amplifier output impedance is meaningless; those are not the only impedances in the circuit.)
Just wanted to re-establish that. (Now off to bed with me!)
Well, not off just yet then.
Panomaniac (your post #991),
You make a point, but also miss one. To take what you say to its logical conclusion, if you say that a difference of 1.14% should be audible, then what dramatic difference would there be between loudspeakers differing by 10%? Yet I have heard nobody complain that their stereo loudspeakers sounded vividly different, so how must I now understand that something contributing only 10% of such an unnoticable difference is going to be audible? I have also pointed out earlier that one is talking of about 0,29 dB difference. (You could transpose that to a Q-change and will get about the same order of influence.) With respect, I am afraid I cannot see how you can claim such order of things as audible. It is quite contrary to tested and accepted audio practice!
I think we must stop talking qualitively and start bringing figures to the party and put it in context with accepted practice of what is audible. (But thanks for an interesting debate.)
Now I will retire!
Regards.
Panomaniac (your post #991),
You make a point, but also miss one. To take what you say to its logical conclusion, if you say that a difference of 1.14% should be audible, then what dramatic difference would there be between loudspeakers differing by 10%? Yet I have heard nobody complain that their stereo loudspeakers sounded vividly different, so how must I now understand that something contributing only 10% of such an unnoticable difference is going to be audible? I have also pointed out earlier that one is talking of about 0,29 dB difference. (You could transpose that to a Q-change and will get about the same order of influence.) With respect, I am afraid I cannot see how you can claim such order of things as audible. It is quite contrary to tested and accepted audio practice!
I think we must stop talking qualitively and start bringing figures to the party and put it in context with accepted practice of what is audible. (But thanks for an interesting debate.)
Now I will retire!
Regards.
Johan Potgieter said:
No problem, in fact your response was ultimately what I was fishing for.
I was bringing up the point, because if you are going to "break" the circuit and claim that the DCR of the cable affects only the device connected at one end, then what end do you choose?
As you point out, choosing an end doesn't actually make sense.
No, I didn't miss the 10% thing. In fact it puzzled me all the way to work.
No argument that 10% is a lot more than you will find in normal cables. If you can improve your speaker by 10%, that's a good idea! Might cost you 100% more, though. All of us have stated that improvements to speakers and the room will yield better dollars for money than cables. But that doesn't mean that cables make NO difference. It might just be that the difference isn't important. Not worth the extra money, real as it might be.
Now that 10% speaker difference doesn't mean much, just floating there in thin air. What the heck is it? A 10% difference in frequency response, sensitivity, group delay, phase, impulse, some or all of the above? How do you figure out 10%??? 10% of what?
And even if you could figure it out, would that be a +/- 5% range in the production run of a speaker? If so, wouldn't the measured differences fall along the classic bell curve?
So most of the speakers of the same make/model will be grouped pretty closely on that curve.
But in practical terms, even that doesn't matter. We have the speakers that we have, we have to deal with them. As we are DIY types, we can modify the speakers if we want. Most people will not. Or if you spend $1000s on a pair of great speakers, may you would rather not. So what can you do? Better cables!! Will they make a difference? Some will, some won't.
It might be silly to spend a large % of your total system budget on cables, the money is better spent elsewhere. But to not try for better cables, amp, CDP, whatever, seems silly too. Isn't that what we all do? If we didn't, we would all have the same cheap amp and DVD player driving the best speakers we could afford. Most people here don't have a system like that. No need to be obsessive about cables, but they shouldn't be ignored either - just because some other speaker has a 10% difference from ours.
No argument that 10% is a lot more than you will find in normal cables. If you can improve your speaker by 10%, that's a good idea! Might cost you 100% more, though.
All of us have stated that improvements to speakers and the room will yield better dollars for money than cables. But that doesn't mean that cables make NO difference. It might just be that the difference isn't important. Not worth the extra money, real as it might be.Now that 10% speaker difference doesn't mean much, just floating there in thin air. What the heck is it? A 10% difference in frequency response, sensitivity, group delay, phase, impulse, some or all of the above? How do you figure out 10%??? 10% of what?
And even if you could figure it out, would that be a +/- 5% range in the production run of a speaker? If so, wouldn't the measured differences fall along the classic bell curve?
An externally hosted image should be here but it was not working when we last tested it.
So most of the speakers of the same make/model will be grouped pretty closely on that curve.
But in practical terms, even that doesn't matter. We have the speakers that we have, we have to deal with them. As we are DIY types, we can modify the speakers if we want. Most people will not. Or if you spend $1000s on a pair of great speakers, may you would rather not. So what can you do? Better cables!! Will they make a difference? Some will, some won't.
It might be silly to spend a large % of your total system budget on cables, the money is better spent elsewhere. But to not try for better cables, amp, CDP, whatever, seems silly too. Isn't that what we all do? If we didn't, we would all have the same cheap amp and DVD player driving the best speakers we could afford. Most people here don't have a system like that. No need to be obsessive about cables, but they shouldn't be ignored either - just because some other speaker has a 10% difference from ours.
panomaniac said:
I guess my approach to that, would be to obtain speaker cable that would have the lowest DCR for the lowest price, then.
Maybe from bluejeanscable perhaps?
The speaker cable would then be a constant, albeit minimal, contributor to DCR in the amp/speaker circuit and the 10% variances between loudspeakers could be fully revealed.
All this time we've been discussing the cable...but I think the quality of the interface between amp - cable - speaker probably has much greater direct influence on signal quality. Banana plug, spade, or direct bare wire? Is there a preference there?
I go for bare wire where possible, occasionally re-stripping and re-connecting if there is visible tarnish.
macgyver10 said:
That's what I do, without going crazy for low DCR. Unless the run is super long. Run some low voltage lights on undersize wire sometime. You'll see the difference, for sure!
Connectors are the important part. No need to go crazy there, just good clean, solid contact. Easier said than done. Remember all those crappy connectors from the 50s and 60s? The oxidized cheap RCAs, the flimsy screw terminals? The cheap spring clips of the 70s and 80s? We've come a long way since then. Most of those connectors work fine when they are in perfect condition. But, man, that perfection sure didn't last long.
When I was in pro audio 1/4" phone jacks were notorious for getting dirty and noisy. The jacks on mixing consoles were awful. Smoke, dirt, who knows what. A nice little 22 caliber gun brush worked wonders on those things. Cured many a problem. Same for speakers with 1/4" connectors.
Just keeping the connections clean and tight helps so much. I love gold connectors. Who cares if gold isn't the ideal conductor? It stays clean and gives a good strong connection. Not may other metals can claim that. I love vintage audio gear, but you can keep the cheap connectors!
- Status
- Not open for further replies.