Impedance matching and cables

How much does impedance matter in cables? Find out the truth.

[@mysock351C]

3:50 FYI I think Paul meant "extremely low" relative to the source. Extremely high impedance would attenuate the signal substantially.

[@marianneoelund2940]

These discussions always end up confusing cable characteristic impedance, with loop impedance. The former is a high-frequency consideration which rarely matters at all in audio, and the latter is mostly the copper resistance which we always want to be as low as practical.

[@thunderpooch]

@@marianneoelund2940 always cracks me up. By the time most audiophiles can afford high end gizmos they can't even hear above 10khz Hearing aids are cheaper than what I see a lot of geezers wasting fortunes on

[@marcusm5127]

@@thunderpoochThey should sell high end speakers with a discount for an inferior tweeter.

[@carlbarker4519]

Thank u Paul for your time!

[@CobraChamp]

If as you say, Paul, “Cables make a difference”, what is the electrical term that is responsible for them making that difference? I believe currently we use the terms capacitance, inductance and resistance to describe a circuits behavior. If we could just come up with an accepted engineering reason why “cables make a difference” then those that are disbelievers in the proper matching of cables to electronic equipment, would have an easier time buying into the concept that they make a difference. Without an engineering reason or (new) term, this will continue to be the third rail of audio circles worldwide. If what you’re saying is true, then there has to be an electrical explanation why certain cables perform differently.

[@user-od9iz9cv1w]

There was an article that talked about impedance matching speaker cables. I thought it was on PS Audio Forums but I can't locate it now. I had built hefty cables that were in a ribbon with conductors separated by an inch. (works well on interconnects). The author opined that this would not work well for speaker cables suggesting an impedance matching issue. I replaced the cables with another diy style that used braided wires totaling similar gauge. To my ear the author was correct and my system sound benefited.

[@beef405]

That could be for reasons other than impedance matching. Impedance does not have effects until roughly 1/5th the wavelength of the signal to the cable (hence making it a transmission line).

[@organicalgorithms]

I’ve been told not to extend speaker cables with anything like a female to female couplers because it, “messes with your impedance” and also not to use speaker cables over a certain length. Do these, or other jury rigging practices have negative affects to your sound system?

[@spacemissing]

All those parts do is secure the connections between pieces of wire. They have NO effect on impedance.

[@marianneoelund2940]

The only consideration for speaker cables, is their series resistance. You want that to be as low as practical; excessive length will compromise your amplifier's damping factor. Adding couplers isn't a problem if they are good quality, but the fewer connections in the speaker circuit, the less likely you are to have issues with them.

[@jkbrown5496]

Any connector introduces the risk of a discontinuity creating a resistance/impedance. It might be small, but there are videos testing electrical connectors under high current, some heat up and melt before others. That's from the resistance in the connection. For audio, you probably can't measure it without special equipment, but then the standard in audio is whether you can hear it. As for speaker cable length, I suppose someone could have used a long cable and accidentally created a long wire antenna for a nearby radio transmission frequency and thus developed the rule for their location.

[@marianneoelund2940]

@Sblackfll Which statement are you referring to? Some additional details: There are a few amplifier designs which require a specific *minimum* speaker cable length, to provide a small capacitive load to the output stage that's a necessary part of its stability network. In general, the complete amplifier/cable/speaker network never achieves damping factor anywhere near the specification for the amplifier alone, due to the driver voice coil resistances. That means a small amount of cable resistance will be swamped by other resistance in the circuit, so it will go unnoticed.

[@marianneoelund2940]

@Sblackfll I would certainly be interested in finding a few people with "golden ears" for an ABX test session. I'm not interested in doing the listening myself, as I'd rather be monitoring signals with test equipment. I'm open to suggestions regarding cable types to compare, and music selections, and I'm willing to bring test equipment to an audio installation where there is a speaker-cable difference claimed. If you're curious, I'm an EE with 40+ years experience in analog and digital equipment design/development, from power supplies to amplifiers, analog signal processing to complex digital systems used for commercial flight controls. I'm a hardware troubleshooter and analyst at heart, and have 100% success rate in solving design issues that have been assigned to me. I was often the engineer of "last resort," called in to solve problems that were stumping other engineers. If there is a difference between the signals delivered to a loudspeaker by various cables, I *will* find it.

[@hoobsgroove]

yes but capacitance and inductance does play a factor so length of cable does play a factor

[@Harald_Reindl]

If you don't plan hundreds of meters it's not relevant

[@SwirlingDragonMist]

Hey man, the last few videos of yours have felt like they were holding back, like it was being dumbed down for brevity. I don’t really feel like I’m deriving value from them anymore. I feel like were just skimming the surface, to the point of not getting anything substantive. There’s allot of videos out there that give layman advice, I feel it’s insulting to the consumer by leaving them in ignorance, I emplore you to shrug off that line of thinking which shrinks away from deeper content by demeaning what the audience is capable or interested in learning. Stay fly homie, Thanks :)

[@tuojisprotesiu]

Cheers from Roskilde, close to Copenhagen

[@godzilla12325]

Cheers from Falkirk, near copenhagen 😏

[@ferrousallotrope]

3:53 I think he meant the impedance is extremely LOW relative to the source.

[@jkbrown5496]

If you really want to start a cable controversy comment on Veritasium's 'The Big Misconception about Electricity' video from the audio perspective. Just remember to duck afterwards.

[@markmywords9649]

Will you make your advancements of the DSD recording technology available to other studios, like Blue Coast Records, Stokfisch Records, Reference Recordings, etc.? Oh and thanks for the great video, as always, and a good/healthy start into the new year!

[@Paulmcgowanpsaudio]

Yes, absolutely we will share what we find and build with any other studio wishing to learn/improve.

[@markmywords9649]

@@Paulmcgowanpsaudio great to hear, but I thought you would so. On a different matter: Will you be at the High End Munich Show 2022 with your new loudspeakers?

[@octaverecordsanddsdstudios1285]

@@markmywords9649 Maybe. We're still figuring that out.

[@markmywords9649]

@@octaverecordsanddsdstudios1285 I know a great number of people here in Europe, including me, would love to see and hear your new loudspeakers. Hopefully we will see you there!

[@VOLKOV9]

Is that Alex Gray art on that mousepad?

[@sirsmarty274]

Copper in and of itself acts as a filter, it is a filter. That is why there is a difference in sound from 1 cable to another. Length, AWG also matters. Copper is not just copper, there are different purities of copper. The best so far being OCC copper. OFC (oxygen free) being second, and everything else can have anything in it. Yes cable quality does matter in sound reproduction. I say if you really had the very best high end cables available at every point in your system, you may not like what you here most of the time because the clarity is so good you here more imperfections rather than coloration. But when you do here a good clean track you will be blown away at how good it is.

[@CarlVanDoren61]

Kimber Kable #1!!

[@txmike1945]

You call yourself Smarty and you write this phrase, "...you may not like what you here most of the time..." Not so smart, hear hear!

[@CarlVanDoren61]

@@txmike1945 Brilliant analogy buckwheat

[@txmike1945]

@@CarlVanDoren61 Well I am glad you approve, I was concerned that you might not and seems everyone who matters believes that you are King of u-tube discussions. Brilliant indeed!

[@cengeb]

Wire you making stuff up?

[@artyfhartie2269]

I think that bunching or coiling loudspeakers cables measuring over 6 feet or more create some resistance or impedance. Stretching them and keeping them to some extent away from power cables improves the sound. More solid bass and clarity in the upper frequencies imo.

[@corystevenponzo7007]

so a coil of wire creates variable resistance, inductance. not much though as cable has heavy insulation. A power cable I don't think would be an issue, remember power cables are throughout our walls. But a low voltage line cable can be suspectable more so than a high voltage speaker cable. A unbalanced cable low voltage ex. a unbalanced mic is susceptible to the condition you described. I highly doubt your example will produce the affects you describe in a balanced scenario. get a mic and get a daw with noise generator and frequency analyzer then you can test this all.

[@marianneoelund2940]

In extreme cases, coiling loudspeaker cables could conceivably affect the highest audio frequencies, but not bass.

[@joshua43214]

Not coiling wires is best practice for all types of wires from extension cords to interconnects. It will certainly have a very audible effect on long audio cables (Dave Rat demonstrated this a while ago). Will it be audible on a short cable? Who cares, follow best practices when possible.

[@artyfhartie2269]

@@marianneoelund2940 Tell that to my ears.

[@shuntachi]

Characteristic impedance and resistance are not the same. If you measure the resistance of a cable, of course it's very low. But characteristic impedance is different thing. Probably not so important for analog cables but it's very important in digital domain. Impedance mismatch creates reflections and deflect the digital signals.

[@marianneoelund2940]

I suspect there was a bit of a "disconnect" between the submitted question, and what Paul was addressing in his answer.

[@psyphonyxaudio]

Thanks Paul !

[@D1N02]

I hear people talking about impedance with phono interlinks. I mean from turntable to amp with phono preamp.

[@richardmackenzie9551]

with phono interconnects it is usually capacitance that is listed if i am not mistaken. I would love to hear more about that as well.

[@marianneoelund2940]

With MC (moving coil) cartridges, you want the phono preamp input impedance to be set correctly - typically around 50 to 100 ohms - for most accurate frequency response. With MM (moving magnet) cartridges, the standard load of 47k ohms usually works best, but you also want to avoid excessive cable capacitance (keep it under 150pF). These considerations are unique to phono cartridges. The more general concept of impedance matching of cables to device inputs is not a serious consideration for audio, but can be important at higher frequencies, such as for video or digital.

[@walterstorm9221]

i've experimented with making DIY cables out of good quality wire and realized how much the braiding and geometry affects the sound!! All I can say is that: 1) it is NOT subtle, and 2) more is NOT always better.

[@thunderpooch]

I'd love to hear what you've discovered. Since flexibility isn't an issue for me, stranded cables are really just piss poor implementation most of the time. Braiding extremely small individual wires, or braided litz is quite good theoretically, but the twisting and braiding needs to be tight. So the diy aspect of braiding your own wires isn't really feasible. Here's what I've discovered. Low dielectric insulation over solid core wire twisted tightly with one other conductor works remarkably well. Sure, skin effect might be an issue, but I think it's overstated. And by using one solid core 12awg or 10awg wire twisted with its same size gauge neutral, I know for a certainty I'm not getting minnie arcing that braided wire suffers from. anytime individual wire strands are packaged into the same insulator, there's tiny air gaps all along those wires, and there's a lot of needless arcing to jump those gaps. Stranded wire is truly for people that can't mange their cable runs and need flexibility. For those that can create nice runs, solid core is the way to go. Granted, Litz is essentially solid core since each individual strand has its own conductor, but I don't think the benfits outweigh the costs.

[@walterstorm9221]

@@thunderpooch I started with 4 lengths of 10ga OCC stranded wire and twisted two together to make a pair of twisted wire. I then twisted the two pair together in the opposite direction which "braided" the 4 of them. These sounded pretty OK, so I thought, what if I combine two of them to create a cable with 8 total 10ga wires :D yeah, that sounded terrible.. and I had to spend like $60 for terminators that would accept the monstrosity I created lol! It really sounded like the music was straining to get through, and the soundstage was flat and non-existent. midrange was thin, bass was hollow, and top end was brittle and fatiguing. I put the straight unaltered 10ga back on and everything was full again. And Im not talking audiophile subtlety here, like it was I thought I may have damaged my amp somehow kind of difference.

[@NoEgg4u]

@2:01 -- Dialog pertaining to the title of this posting begins.

[@iampuzzleman282]

Why didnt u explain impedance?..

[@bobmenke3211]

so there is no way to measure and explain why cables supposedly sound different?

[@sirsmarty274]

Find my comment above, it may help some, measuring ohms for a cable, really doesn't even matter, its really splitting hairs and bean counting, The basics are length, AWG, and quality of copper in the cable.

[@marianneoelund2940]

@@sirsmarty274 Connector quality and assembly workmanship are even more important than the cable parameters.

[@marianneoelund2940]

They have no engineering explanation, because Paul's (and his colleague's) ears are apparently endowed with better precision than any of their measuring equipment.

[@Paulmcgowanpsaudio]

None that has brought any satisfaction to those who don't agree or haven't actually tried them. There are a few difference tests online where you can see the actually differences between two cables but then those who don't want to hear that simply suggest that despite the difference it's not audible. From my view it seems a waste of time if one can simply listen.

[@bobmenke3211]

@@marianneoelund2940 The evaluation of cables has become a paranormal experience. I wish there were more double blind abx testing of cables.

[@allansh828]

would you talk about DSD sources

[@xanderguldie]

He just did yesterday. Pay attention bud

[@Metalhead-4life]

Exhausting??? You spent 6 minutes with us

[@budgetaudiophilelife-long5461]

🤗👍😍😍😍

[@theeprox]

Hahaha

[@bobross6802]

There is a difference between a resistance and an impedance.

[@txmike1945]

Obviously, otherwise you would only need one of those two terms.

[@andydelle4509]

Absolutely. Here is an example. I once worked on some 11ghz microwave gear. The input TNC (not BNC) connector had a 1/8in silver plated copper strap from the connector center soldered down to the brass sub plate of the device, about 1/2in long. A dead short, right? It would zero out any commom DMM. But at 11ghz, that was 50 ohms! It was also a "winding" on a coupling transformer to the secondary loop next to it - also a dead short at DC. At audio frequencies true impedance matching is of little consequence with an exception for tube amp output transformers..

[@bobross6802]

We're talking audio frequency here, not RF. Audio cables aren't measured like Coax 52/75 or 300 ohm balanced lead-in .... so I think the term to be addressed in audio freq. would be resistance. Like too small speaker cable gauge will cause signal loss at the speaker due to resistance, not impedance

[@txmike1945]

@@bobross6802 "Like too small speaker cable gauge will cause signal loss at the speaker due to resistance, not impedance." Yes, I witnessed that once. My friend had just added speakers in the corners of his room, maybe 12 feet from the sound source. The music sounded terrible, muffled with poor response. Then I discovered he had used lengths of multi-wire telephone cable, those tiny wires just couldn't carry the signal. I told him to buy some 14 ga speaker wire.

[@andydelle4509]

@@bobross6802 I know we are talking audio and not RF but my example was to show an extreme case where impedance does not equal resistance. A confusing eample is an 8ohm speaker. Measure one and it will read around 8 ohms resistance as well. And audio cable does have a characteristic impedance, not that is makes any difference these days. Belden 9451, the industry standard balanced line audio cable is 66 ohms. Interesting that that matches the output impedance most audio grade OPAMPS. With speaker cable we agree that resistance is the primary concern. But over 50 feet, inductance related issues start to creep in according to Roger Russell.

[@finscreenname]

"a bit exhausting" ....setting up a camera and answering sent in questions while the whole time having 170K subs, crap load of views and free advertisement for PS Audio....some should be so lucky to be that taxed. I shot 78 minutes of video yesterday, condensed it down to 7 minutes and when done only 20 people will see it.

[@txmike1945]

Yeah, well you are not as interesting as Paul is.

[@finscreenname]

@@txmike1945 Ya and I'm thinking that Paul wouldn't be as interesting in what I mainly do videos on either. Go ahead and defend a guy that thinks it's exhausting to answer a sent in and prescreened question in front of a video camera. IMHO it looking a gift horse in the mouth. Also, dont thumbs up your own post, its tacky.

[@paulb.3227]

@@finscreenname Well well, what a jalous guy you are !

[@finscreenname]

@@paulb.3227 Coming from a guy that's been on YT for 3+ years and never posted a video or even made a play list. I really value your opinion.

[@paulb.3227]

@@finscreenname O really? So I Have to post a video to have an opinion ??

[@cengeb]

Oy vey

[@ThinkingBetter]

Some things matter and some things don't. I'm happy Paul is not promoting cable lifters.

[@paulhunter6652]

CD cases upside down..almost free.

[@ThinkingBetter]

​@@sub-vibes When you spend US$200,000 on an audio system with cables costing US$20,000 of that budget, you need something to hold those cables for cosmetic reasons and if that is the argument, I'm all for it, but don't tell me the sound is any different. Well, our brain will make it sound different because we react to what we see.

[@iampuzzleman282]

Omg, get to point. Dsd?? Wtf.

[@simonlinser8286]

he says transcendence and you complain about work ha ha ha

[@CarlVanDoren61]

Personal time to get more sales, we know your angle Paul.

[@txmike1945]

And you find something wrong with that? You'd be a moron to not realize all companies do this in one form or another. The hope (for us) is that we gain some interesting insight in the process. But YOU do have an option, quit coming to the PS Audio u-tube site.

[@CarlVanDoren61]

@@txmike1945 lol, who peed in your Wheaties

[@txmike1945]

@@CarlVanDoren61 You posted a rude comment on Paul's own channel, please behave better!

[@mysock351C]

Pro-Tip: If you want good impedance matching to your speakers, don't use tube amps. Even the best ones from companies like McIntosh will still have about 0.5Ω of output impedance due to the need to use an output transformer. Many tube amps are far worse, and this causes the speakers dynamic impedance of the loudspeaker to pull the output of the amp around, with the net result that the speakers frequency response now bears the shape of its impedance curve if the output impedance is bad enough. And this doesn't even cover the issues of their very poor linearity in many cases. It's not musical, its DISTORTION. Solid state amps are typically no more than 0.1Ω, and don't have this issue. As far as the speaker cables go, there the concern will be the cables reactance that results from excessive length (due to parasitic inductance and capacitance). Even though 100ft of 14AWG two-conductor wire will only have about 0.5Ω of resistance, its reactance could easily cause an amplifier to oscillate. Linear amps will have Theil and Zobel networks to help guard against this, but still best to avoid it. This is not an issue for low level signals since they are so small. Their noise is the major issue.

[@spacemissing]

I don't think you fully understand tube amplifiers. (Not that I do either, but I DO understand why they are not problematic in regard to impedance.) I will leave it at that, and I will not get into a discussion about it.

[@mysock351C]

@@spacemissing How can you pass judgement on my understanding, when you yourself don't understand? I find it quite curious. At any rate, tubes have quite poor linearity in their native forms due to how they work. An additional issue is the fact that they have to work at high voltages in order for the cathode/grid/plate structure typically used in triodes to function. This necessitates the use of an output transformer since the tubes are inherently _high impedance_ devices. At best you will be looking at about 0.5Ω on the secondary. Given that its common now for audiophile speakers to have impedances that dip down into the 2-3Ω region, this means that the resistance on the secondary winding of the output transformer will act as a voltage divider in series with the impedance of the loudspeaker. The net result is that the lower the impedance of the loudspeaker goes, the more the output voltage of the amplifier is dragged down. The net effect is to change the on-axis response on the speaker and impress the shape of the speaker's impedance curve on it. While this won't be much more than 1-2 dB with a competent tube amp, it's still not negligible. This also assumes that the amp designer has not done something stupid like not using feedback or going with a SET design and making a guitar buzz box rather than a linear amplifier. This is all neatly sidestepped by solid state linear and class-D amplifiers with their feedback and feed forward designs. Switching topologies from companies like Benchmark and Hypex now have less distortion than even the best linear amplifiers. An additional bonus to feedback is it drives the output impedance down into the milliohm territory, which is what you want for a voltage amplifier. For what it's worth, PS Audio now almost exclusively uses fully integrated class-D modules in their amplifier designs.

[@marianneoelund2940]

There seem to be quite a few "audiophiles" who enjoy transformer sound, and believe that some low-order distortion adds a desirable quality to their music.

[@ronniefranks4351]

@@mysock351C I’m quite happy with my 2 w/pc single ended triode no feedback mono amplifiers. You listen to your specs but I prefer listening to music. Great specs doesn’t necessarily make for a great sounding amp.

[@mysock351C]

@@ronniefranks4351 Electronics are not supposed to have a “sound.” Dealing with the pathologies of speakers is bad enough, and they are already principally 2nd and 3rd harmonic distortion due to how they work. No need to add even MORE of this with tubes, along with a spray of other harmonics and IMD. Since you are not familiar with how these impact the sound itself, it’s not bothersome to you. It would be to me.

[@mattbonaccio3522]

What? Paul, you'd better check your facts, man. This is one of your worst videos yet, and probably the only one I've seen where you're just flat wrong. Cable impedance, expressed in ohms at a particular frequency, is not particularly useful in audio, but it does matter. Painfully, it matters most in the situation you explained that it doesn't, that of MC phono cartridge loading. Phono cables frequently have a capacitance of 30 to 50 picofarads per meter in parallel to the loading capacitor in the preamp. This means the two capacitances add together, and you may in fact run into issues with this with the wrong combo of cartridge, cables, and preamp. It generally matters much less for other cables (e.g. from a preamp to a power amp, or a CD player or DAC to a pre), but at least a few cable manufacturers and audiophiles claim that the capacitance of these interconnects is responsible for audible differences, so there you go. This info is no good, all viewers take note!

[@marianneoelund2940]

Capacitance is most critical for MM cartridges, not MC. With MM cartridges, if your total capacitance is about 150pF or less, it shouldn't be a problem. With MC cartridges, it's important to have the load impedance (amplifier input impedance) set correctly for flattest response (usually in the 50-100 ohm range), but cable + preamp input capacitance will not be a significant factor since the cartridge output impedance is only around 10 ohms. For interconnects, it takes rather extreme combinations to produce audible effects. Even for a high source impedance of 2.5K ohms, 1000pF of loading only attenuates 0.5dB at 20kHz.

[@RoderikvanReekum]

First !!!

[@NoEgg4u]

kzbin.info/www/bejne/nXSmdWdmmqdgmbc And giving your own useless comment a thumb's up is pathetic.

[@thunderpooch]

@@NoEgg4u giving your own comment a thumbs up is awesome sauce