AudioCircle
Audio/Video Gear and Systems => Owner's Circles => Digital Amplifier Company Owners => Topic started by: AmpDesigner333 on 17 May 2018, 05:48 am
-
Speaker Wire resistance can kill damping factor. This is especially important with low impedance, high power amplifiers.
The load (speaker) is connected to the amplifier output in series with the amplifier's output resistance AND the speaker wire's resistance.
As a fun math experiment, compare the output resistance (impedance simplified) of a high performance amp, like Maraschino, which is less than 10mΩ before the connectors (both) and about 16mΩ after. This results in a damping factor of 500 into an 8Ω load. If the speaker wire has a resistance of 16mΩ, that would cut the effective damping factor in HALF (now effectively 250)!
Now, add TWICE (round trip) the resistance of a 6-foot wire....
10-gauge is 12mΩ
12-gauge is 19mΩ
14-gauge is 30mΩ
16-gauge is 48mΩ
Damping factor is just load resistance divided by output resistance.
So, an amp with a damping factor of 500 and (16mΩ output resistance into 8Ω) driving a load through a 6-foot 16-gauge speaker wire has an EFFECTIVE damping factor of only 125!
What do you think happens when an In-Line Maraschino is used, hanging from the speaker with 6" (that's INCHES!) long 12-gauge wires? Hint: 1-foot of 12-gauge is only 2mΩ....
-
Funny how a bunch of people just got done telling me that it doesn't do anything :roll:
-
Handy table....
1000 ft of 8 gauge: 0.628 Ω
1000 ft of 9 gauge: 0.792 Ω
1000 ft of 10 gauge: 0.999 Ω
1000 ft of 11 gauge: 1.26 Ω
1000 ft of 12 gauge: 1.588 Ω
1000 ft of 13 gauge: 2.003 Ω
1000 ft of 14 gauge: 2.525 Ω
1000 ft of 15 gauge: 3.184 Ω
1000 ft of 16 gauge: 4.016 Ω
1000 ft of 17 gauge: 5.064 Ω
1000 ft of 18 gauge: 6.385 Ω
1000 ft of 19 gauge: 8.051 Ω
1000 ft of 20 gauge: 10.15 Ω
-
Funny how a bunch of people just got done telling me that it doesn't do anything :roll:
It's not that it doesn't do anything, it's that when the damping factor gets well above 10 it's further impact quickly diminishes.
What they are hearing is the effect of added series resistance causing response changes with frequency. Just look at the Stereophile measurements of some tube amps (just about the effect).
-
The explanation covers only half the story.
Effective Damping Factor values for calculation includes:
> 1 run speaker cable > resistance of crossover elements > voice coil resistance > 1 run of speaker cable
(ignoring for the moment the effect of various connectors, etc which would further increase the above value).
A value of 20 or higher is adequate.
-
It's not that it doesn't do anything, it's that when the damping factor gets well above 10 it's further impact quickly diminishes.
What they are hearing is the effect of added series resistance causing response changes with frequency. Just look at the Stereophile measurements of some tube amps (just about the effect).
How do you come to this conclusion? Lower damping factor means less control over the speaker. More control typically means better, more accurate sound.
-
The explanation covers only half the story.
Effective Damping Factor values for calculation includes:
> 1 run speaker cable > resistance of crossover elements > voice coil resistance > 1 run of speaker cable
(ignoring for the moment the effect of various connectors, etc which would further increase the above value).
A value of 20 or higher is adequate.
How do you come to that conclusion?
Speaker system are typically designed to assume near zero source impedance. The voltage at the speaker terminals does NOT include resistances "inside the box". Think of it as a black box, and your job as an amplifier is to deliver the most accurate signal to the black box.
-
A value of 20 or higher is adequate.
+1
Or another way of putting it, anything over 20 brings you vanishingly little.
dave
-
Lower damping factor means less control over the speaker. More control typically means better, more accurate sound.
The 1st is true (but implies that each speaker needs maximum control). The 2nd is baldface misinformation.
A speaker/amplifier system to sound most accurate needs to have a range of total damping. A amplifer can provide a certain amount of electrical damping. A speaker has an inherent mehcanical damping. These add up to give total damping. If the amplifier has too low an output impedance (high damping) can overdamp a speaker and cause lean bass.
Now it is true that many (if not most) speakers available today are designed with the assumption that the amps has a high damping factor (ie >20). If you own one of those speakers you should get an amplifier with high damping or you may find that the bass is towards bloated/fat. But there are speakers available (even more common in the diyWorld) where the speaker likes to see a higher output impedance because they have high meanical damping so don’t need so much electrical damping.
This can be taken further. Since a speaker is a current driven device, there is an argument that an amplifier that is a current source will have all sorts of lesser distortions. So it sounds better. But only practical if the impedance is close to flat since the FR of a speaker driven by a current source will be convolved with the speakers impedance. Almost every speaker with a crossover will not satisfy this criterion — the crossover makes the impedance take a wild ride. Also we have the speaker’s fundemental resonance which is typically shown by a big rise in impedance, a sign of an underdamped situation where extra electrical damping is beneficial. So to use a current amp you either need a speaker with very well behaved impedance (very rare) or use current amps in an active system for mid/top where over the bandwidth of the directly driven loudspeaker the impedance is flat, and use a voltage amp for the bass because the speaker does not have a flat-curve.
Also, with a single driver FR system, the impedance rises at the bottom and at the top. and an amplifier with just the right output impedance can extend the flat frequency range both up & down. Experiments with a constinuously variable Transimpedance amplifier (output impedance can be dialed from near 0 to near infinity) that each different loudspeaker had an optimal setting.
So what is the best damping? It depends. The OP is trying to stir up a tempest in a teapot.
dave
-
+1
Or another way of putting it, anything over 20 brings you vanishingly little.
dave
Based on what? I've heard the difference, and much more. So you know I'm not just making baseless claims, I'm an electrical engineer with 30+ years experience in audio (consumer and pro), plus military, medical, fire safety, and even toys. That's design experience from concept to production. I've been able to connect the dots between measurements and sound. This is through circuit changes, measurements, and listening.
I can say for a fact that higher effective damping factor sounds better on systems that have adequate resolution. This is the reason I brought it up in the first place. Speaker wire (or "just straight wire") is low enough inductance that its inductance is negligible. Speaker wire is low enough capacitance that its capacitance is negligible. The resistance is NOT negligible, and damping factor matters!
Take a look at what people say about Cherry Amplifiers! Why are so many people saying it's the best sounding amp they've ever had? We do things right, and low impedance is one of those things. Pay special attention to the ILM (In-Line Maraschino) owners. Many of them HANG the amps from their speakers. The result? Speaker wires with less than one foot of round trip length ---- near-ZERO speaker wire resistance. We're talking about less than a milliohm! The sound delivered by these high performance (117dB+ SNR, 0.002% THD, 0-100kHz), small, light (they weigh about a pound), yet very powerful (up to 1000W into 1.8 ohms!) amps is nothing short of amazing. Here are some reviews on Cherry Amps:
https://www.audiocircle.com/index.php?topic=127654.0
If you think a 1960s technology audio system sounds just swell, than higher damping factor, wider bandwidth, DC coupling, true balanced operation, super low noise, super low distortion, and adequately low speaker wire resistance won't help you. Hey, some people are perfectly happy driving a Yugo.
-
Based on what? I've heard the difference, and much more. So you know I'm not just making baseless claims, I'm an electrical engineer with 30+ years experience in audio (consumer and pro), plus military, medical, fire safety, and even toys.
Me too. And possibly over a much wider range of loudspeakers. 45+ years in audio. And while no engineering degree (1st half of one, but i started with an honours math degree 1st) an engineering mind. And i like to leave it open. Engineering will let you comfortabley generate numbers, but if you include everything in the circuit and damping factor greater than about 20 adds vanishing little to the end result. Much as Johnny2Bad has already done for us in this thread.
dave
-
Me too. And possibly over a much wider range of loudspeakers. 45+ years in audio. And while no engineering degree (1st half of one, but i started with an honours math degree 1st) an engineering mind. And i like to leave it open. Engineering will let you comfortabley generate numbers, but if you include everything in the circuit and damping factor greater than about 20 adds vanishing little to the end result. Much as Johnny2Bad has already done for us in this thread.
dave
Dave,
So we can agree to disagree. In my experience, it matters more with high performance hardware (measurement and SQ wise). Lower speaker wire resistance won't do much if you're using an already handicapped tube amp with poor damping factor, rolled off highs, and high noise, for example. Have you ever heard a Cherry Amp? Have you checked out the link I included above? Have a great night. It's pretty late here in Allentown PA.... Thanks for contributing to my board.
-Tommy O
p.s. Check out this thread started by a customer that went from a life of tubes to Cherry Amps ("SHATTERED GLASS … No more tubes – Class D all the way baby and loving it!"):
https://www.audiocircle.com/index.php?topic=154294.0
-
… you're using an already handicapped tube amp with poor damping factor, rolled off highs, and high noise, for example. Have you ever heard a Cherry Amp?
Your amps are on my long lost of amps i’d like to hear. And i’m guessing that the tube amps of which you are talking are those pretty much emulating the crap of the past. There are some VERY good tube amps.
To quote WindChaser from that thread:
… how much better it mated with my speakers…
Exactly. Just like i said in my long post… with his speaker. And i also said most loudspeakers need an amp like yours.
dave
-
Check out my post on this thread about solid versus stranded wire:
https://www.audiocircle.com/index.php?topic=157556.msg1688249#msg1688249
-
F. Langford Smith wrote about this in 1947.
Industrial audio consultant Dick Pierce wrote this paper in 2002:
"Damping Factor: Effects On System Response"
Dick Pierce
Professional Audio Development
http://www.cartchunk.org/audiotopics/DampingFactor.pdf
http://www.cartchunk.org/
-
F. Langford Smith wrote about this in 1947.
Industrial audio consultant Dick Pierce wrote this paper in 2002:
"Damping Factor: Effects On System Response"
Dick Pierce
Professional Audio Development
http://www.cartchunk.org/audiotopics/DampingFactor.pdf
http://www.cartchunk.org/
Not one mention of speaker reactance! Translated: obvious bias toward low performance amplification
-
Not one mention of speaker reactance! Translated: obvious bias toward low performance amplification
That's because system 'Qr' which is what this is all about is measured at resonate frequency where the reactances cancle out.
-
That's because system 'Qr' which is what this is all about is measured at resonate frequency where the reactances cancle out.
If you're looking at a simple resistance (representing output impedance of the amp at a given freq) driving a reactive network, there is no "cancelling out" of the reactance. Here's an interesting article I found regarding speaker reactance:
https://hometheaterhifi.com/technical/technical-reviews/a-secrets-technical-article/
Zero output impedance is ideal, although unattainable, so the lower the better, but where does it stop making a difference? I like to look at it as orders of magnitude, and two orders of magnitude (minimum, as a rule of thumb concerning impedance ratios) gives us a damping factor of 100. This is only attainable with adequately low speaker wire resistance. Note that some speakers dip very low impedance wise (under 2 ohms, like some ESLs, and Gallo speakers, those with transformers are notorious for this). However, this static impedance number only tells half the story, and is generally an over-simplification. There is a phase angle from the back-EMF of the speaker, and this can amount to a much worse situation than driving a simple 2 ohm resistor. Ever wonder why some amplifiers "can't handle" low impedance speakers? This is why some manufacturers say stuff like "stable down to 4 ohms" or "speakers less than 4 ohms are not recommended". It's like a hockey fight ---- harder to land a solid punch when you're sliding on the ice. The slippery ice is like high output impedance! Simple physics, related to opposing forces.
Amps with very poor damping factor (non OTL tube amps, for example) can't fight much against the back EMF of the speaker. This causes a "sloppy sound", especially noticeable in the bass. Also, most amps with great damping factor only have that characteristic at low freq (100Hz, where it's typically tested, and lower), then the output impedance climbs like crazy as the freq goes up. Some amps with high bandwidth damping attain this by applying tons of feedback, creating the need for wild phase shifts to allow stability. This is especially the case with Class-D that wraps feedback around the output filter. The same technique to get decent specs on paper when the amp is driven with sine waves of one or only a few frequencies (what's used for THD+N testing, for example). Music is not a continuous tone! The problem is that this same "high feedback" practice can cause an "etched" sound. How many times have you heard "great bass, but harsh sounding" about one of those pre-fab module based Class-D amps? We have conquered this issue with our patented and proprietary amplifier circuitry. This is why users of other Class-D amps, as well as tube fans, have migrated to Cherry Amps. Thanks.
-Tommy
-
Zero output impedance is ideal
Given that this is false what follows does not have univesal application.
Ever wonder why some amplifiers "can't handle" low impedance speakers?
It is simple. The power supply cannot deliver sufficient current. Also the reason why some amps cannot do a decent job driving ESLs. With an ESL it is more a case of the speaker looking like a capacitor. Since Power =VA x cos(phase angle) and the closer to a capacitor the speaker is the closer cos(phase) gets to zero (and has little to do with back EMF. Esa Meriläinen’s book (https://www.current-drive.info/) has a very good analysis).
Amps with very poor damping factor (non OTL tube amps, for example) can't fight much against the back EMF of the speaker.
Once the output impedance becomes greater than the speakers (damping factor <1) we get into current amp territory. With a true current amp there is no EMF. The lower an amplifier’s output impedance the greater issues it has with back EMF.
This causes a "sloppy sound", especially noticeable in the bass.
This is very dependent on the mechanical damping of the loudspeaker system — if the impedance curve is flat at resonance the speaker has a well damped bass response and a high damping amplifier will sound very thin prodcuing lss than ideal bass.
most amps with great damping factor only have that characteristic at low freq (100Hz and down)
Duh. Given that damping factor only has a real effects in the region of the speaker's resonance frequency.
dave
-
\https://hometheaterhifi.com/technical/technical-reviews/a-secrets-technical-article/
The real purpose of this article is to show the Smith chart of impedance and the intro has some shortcomings in terms of not telling the whole story (and some untruths — ie a tweeter’s impedance is not capacitive). What we have with the 3 impedance plots shown here are just what one sees when one looks at the 3D impedance curve (Nyquist plot) collapsed into a 2d chart when looked at from a specific axis. Yje magnitude when one is looking from an axis where the imaginary parts disappear, the phase angle when collapsed from a view that hides the real data, and the Smith chart Variation of orthoganal to both of these (ie from the end). Much better, given computer screens these days would be a 3D display and the ability to rotate the view point.
(http://frugal-phile.com/images/3D-impedance-curve.png)
https://www.prosoundweb.com/topics/sound_reinforcement/the_nyquist_plot/#
Richard Heyser was a proponent of this and one of the audio magazines he did measures for showed a Smith curve (or similar). http://www.aes.org/technical/documents/openaccess/AES_TimeDelaySpectrometry.pdf
dave
-
Given that this is false what follows does not have univesal application....
Dave,
I don't understand why you disagree. The ideal amplifier is a variable voltage source, and the ideal variable voltage source has zero output impedance.
Also, you state power supply current as the culprit of load impedance minimums, but this is an over simplification. The output stage is in between the power supply and the load. So is the speaker wire. Connectors too, but since you think damping factor isn't important....
Here's an easy experiment you can do to examine the effect of output impedance driving a speaker.... Run a simulation with an ideal voltage source through a resistor into a speaker load model. The parameters of the speaker model can be simple (let's say, a two way system). Run plots of the response across the source resistor. Step this resistor value using a range equivalent to a damping factor of 1 to 1000, log. Check out the "loss" across this resistor!
Keep in mind many speakers use drivers that have a HUGE impedance spike at resonance, typically 30-100Hz, which is where (freq wise) manufacturers typically measuring damping factor. This happens to be where it matters the least. Typical amps the start losing control of the speaker as the frequency rises, ironically as the speaker's impedance goes lower. Many well respected high end speakers go below 2 ohms, for example. So, that leads me to another question....
Have you ever heard a Cherry Amp? Specifically our newer models, like Maraschino or MEGAschino? These are good examples of what amps with high wideband damping factor, low distortion, DC coupling (no bass phase shift), high bandwidth, and low noise can sound like. Plus, we take it through additional steps to optimize for sonics. This is why Cherry Amps don't suffer from the "dry" sound of other Class-D designs or the "sloppy" sound of tube amps. Those are bold claims, yes, but I'm not alone in my opinion here. Take a look at reviews on our Circle. The proof is in the pudding.
When it comes to damping factor, there's no reason for wanting it lower. Well, maybe if your selling amps with poor damping factor.
I want you to know that I do appreciate your use of math and science. That is refreshing, and even if we don't agree on some narrow issues, we both believe in engineering.
-Tommy
-
=I don't understand why you disagree. The ideal amplifier is a variable voltage source, and the ideal variable voltage source has zero output impedance.
Many of the speakers we design do not like to be driven by low output impedance amplifiers. So we have direct counterexamples that disprove your assumption of what an ideal amplifier is.
An example (in fact, almost all of the Fostex FE series want to be driven by a high output impedance amplifier):
(http://p10hifi.net/FH/images/Aiko-FE108eS.jpg)
you state power supply current as the culprit of load impedance minimums
It is certainly the biggest reason.
Here's an easy experiment you can do to examine the effect of output impedance driving a speaker.... Run a simulation with an ideal voltage source through a resistor into a speaker load model.
How about a real world version of that where one puts a resistor in series with the speaker. With the appropriate speaker the bass improves.
(let's say, a two way system)
Let’s say a 1-way speaker.
Have you ever heard a Cherry Amp?
You asked that question already and i said no. And with so many amplifiers to try, and the more you show your ignorance of this subject the lower your amp’s get on the list.
...amps with high wideband damping factor, low distortion, DC coupling (no bass phase shift), high bandwidth, and low noise...
We like most of those, but high damping factor is very speaker dependent (with the right speaker, a current amp can provide the best performance), and one has to ask how you are quantifying low distortion. It is well understood that THD is meaningless.
Cherry Amps don't suffer from the "dry" sound of other Class-D designs or the "sloppy" sound of tube amps.
I interpret “dry”as lean, which is often caused by overdamping a speaker, and there is as much variation in quality of tube ams as there are in SS amps. Many of today’s tube amps are just clones of vinatge amps with all their same problems, and Single Ended amps with their highish output impedance want to be used with an speaker that likes that (like the Fostex mentioned)
Those are bold claims, yes, but I'm not alone in my opinion here.
It only takes one counterexample to show that your opinion is out-to-lunch. A speaker & an amplifier need to be considered as a system. There are speakers that will like your amps, and there are speakers that they will not work well with. My bet is that all the positive reviews are people using the amps with speakers that like them (in which case they may sound spectacular).
When it comes to damping factor, there's no reason for wanting it lower.
Except when you have a speaker that likes a high output impedance amplifier.
dave
-
Many of the speakers we design do not like to be driven by low output impedance amplifiers. So we have direct counterexamples that disprove your assumption of what an ideal amplifier is.
An example (in fact, almost all of the Fostex FE series want to be driven by a high output impedance amplifier):
(http://p10hifi.net/FH/images/Aiko-FE108eS.jpg)
It is certainly the biggest reason.
How about a read world version of that where one puts a resistor in series with the speaker. With the appropriate speaker the bass improves.
Let’s say a 1-way speaker.
You asked that question already and i said no. And with so many amplifiers to try, and the more you show your ignorance of this subject the lower your amp’s get on the list.
We like most of those, but high damping factor is very speaker dependent (with the right speaker, a current amp can provide the best performance), and one has to ask how you are quantifying low distortion. It is well understood that THD is meaningless.
I interpret “dry”as lean, which is often caused by overdamping a speaker, and there is as much variation in quality of tube ams as there are in SS amps. Many of today’s tube amps are just clones of vinatge amps with all their same problems, and Single Ended amps with their highish output impedance want to be used with an speaker that likes that (like the FOstex mentioned)
It only takes one counterexample to show that your opinion is out-to-lunch. A speaker & an amplifier need to be considered as a system. There are speakers that will like your amps, and there are speakers that they will not work well with. My bet is that all the positive reviews are people using the amps with speakers that like them (in which case they may sound spectacular).
Except when you have a speaker that likes a high output impedance amplifier.
dave
Dave,
I suspect that the speakers you show would have difficulty reproducing realistic (live level) SPL. Do you use subwoofers with these? I'd be worried about popping that little cone onto the floor. :o No idea how good it sounds at low SPL, but everything gets easier when you're only trying to reproduce a "miniature version" of a musical performance. This is not to say that a system is only good if it can "play loud". I do like the idea of no crossover, but I remember a similar system at 2016 RMAF that came with tons of hype and the sound of a table radio.
We like to make systems where you feel like you're listening to a band, not a stereo system.
Yes, I asked the question about listening to Cherry Amps more than once ---- for a reason. Think about it.
Please proofread your comments. Misspellings and grammatical errors sometimes change the meaning. I suggest you re-read your comments to see what I'm talking about, like "read world" where it appears you meant "real world". The quality of one's communication means a lot.
I prefer the term "nostalgia audio" over "vintage audio". There have been many improvements in audio technology over the years. Going back is typically just that. Nothing personal.
Thanks for contributing to my thread, but I'd appreciate a less insulting tone moving forward. Let's keep it classy. Also, this is supposed to be a fun hobby. Animosity is not a good spice to add to the broth.
Back to damping factor.... If your speaker "likes" series resistance, why don't they just build that into the speaker? Of course, this means you can get away with poor damping factor, but why kludge the amplifier to fix a speaker problem? Damping factor is not the only thing about our amps that sets them apart. The amp as an isolated element shouldn't be handicapped to accommodate the speaker. Our philosophy is that the amp should do the best job it can to be as close to an ideal amplifier as possible without compromising it's ability to deliver a fluid, musical experience. This also allows bringing new life to old speakers!
-Tommy O
-
Assertions that are not true need to be challenged — like yours that an ideal amplifier has zero output impedance. There are for instance arguments that suggest an amplifier with a very high output impedance (a current amp) is best. But there is no best in hifi, far too many compromises need to be made for that to be possible.
Every loudspeaker has a series of compromises. A good FR speaker does things that no multiway can achieve. For some this is seductive. All the ones i deal with are diy because good commercial ones are few on the ground and often very pricey.
dave
-
Assertions that are not true need to be challenged — like yours that an ideal amplifier has zero output impedance. There are for instance arguments that suggest an amplifier with a very high output impedance (a current amp) is best. But there is no best in hifi, far too many compromises need to be made for that to be possible.
Every loudspeaker has a series of compromises. A good FR speaker does things that no multiway can achieve. For some this is seductive. All the ones i deal with are diy because good commercial ones are few on the ground and often very pricey.
dave
Are you familiar with the work of Stewart Hegeman?
-
Yes.
dave
-
Yes.
dave
So you know about his famous "straight wire with gain" stance?
-
Yes.
dave
-
Yes.
dave
Dave,
Do you agree with Stewart Hegeman? If you do, you agree with the "ideal voltage source". This doesn't mean anything bad. If you don't need an ideal amplifier for your particular speakers, or something even close, there's no harm in that either. Thanks again.
-Tommy O
-
I dodn’t think that that says “ideal voltage source”. A speaker is a current device. Using the impedance of the loudspeaker as the resistor in an I/V converter is fraught with problems. Amplifiers as volatge sources made it conveinient for loudspeaker makers to move into a corner of the possible design space.
dave
-
Just revisited this thread after a long time. There are many pseudo-science and even anti-science claims made by fancy cable makers. Stuff about skin effect, which is not significant at audio frequencies, polarizing shields, which is just silly, and other similar junk (like various metals, etc.).
When it comes to designing electronics, or electromechanical devices, it takes education AND experience. Each is not a substitute for the other, no matter the magnitude.
Regarding speaker cables:
1. Shielding isn’t necessary because the impedances are so low.
2. The lower resistance, the more damping factor is preserved.
Regarding amplifiers, the ideal voltage amplifier (as 99.99% of audio amps) has voltage gain and no output impedance, regardless of frequency, amplitude, or load. This is fact, and not debatable. If you’re an electrical engineer, this is something you were taught early in school, as soon as the core courses were out of the way, and simply second nature as a practicing engineer, if you are fortunate enough to be one. Some engineers get the degree and never wind up doing any actual engineering.
Now, we have speaker cables available, although we don’t push them or try to make any significant margin on them. They are designed for easy use (physically flexible) and LOW RESISTANCE! Our current top-of the-line model is the MEGA Snake, a 9-gauge equivalent cable! Here’s a photo....
(https://www.audiocircle.com/image.php?id=197888)
Fact is, there are many companies, and individuals, that make up problems so they can “solve them”. This stuff comes from marketing, not engineering.