DF is limited by the inductor in series with the woofer to about 100. DF of an amp should be few times higher. Very high DF=4000 (Icepower) doesn't make much difference - at least at low frequencies. Low output impedance at higher frequencies helps to cut interaction between tweeter and midrange (especially with biwired speaker).
Power itself is not important - sound is. In order to get twice louder you need ten times more power.
Don't buy an amp for the specifications - buy it for the sound!!! |
So in our imperfect world, which is more important? Frankly, neither. The most important thing is how the system as a whole sounds to YOU in YOUR room. Buying equipment based on specs can be a very expensive endeavor. You might find the perfect set of specs and save a few bucks by buying it on audiogon, but if it doesn't sound good you you then it doesn't matter how much you saved. Take a look at the article on Wikipedia about damping factor and read some of the external links in that article. That should give you a good idea of what damping factor is all about. If you haven't already, look through the various audio forums and see what other people are using with their maggies or post a question looking for the same information. Also include some information on the "sound" you are trying to achieve as well as your budget. Finally, if you have (a) good local dealer(s) then talk to them. A good dealer should be able to help guide you in the right direction and most will also allow you to borrow equipment to try in your system. If your dealer provides this service and helps you out it would be wise to return the favor and buy whatever component you decide on from that dealer.... otherwise you may find that the good advice and loaner gear made unavailable to you in the future. |
The claim that high wattage combined w/ high damping factor is expensive is an urban legend. |
Damping factor needs to be at least above 10 and above 50 it will make very little incremental difference.
Watts are essential in any amp but around 200 Watts into 8 Ohms is all you need...after that it is diminishing returns. |
Shadorne is pretty much on the money, as are the other posts here. In audio its easy to make a small fortune- start out with a large one :)
In this case, buying by the specs (when it is your ears that ultimately matter), is going to flush a lot of money away. So be careful! If you would like some hints, the Magnaplaner people claim that the best they have heard their speakers is with tube amplifiers. They tell me that every time I see them.
Good Luck! |
Icepower is class D - has high power, high DF and is not astronomically expensive (to answer your question).
I stated previously what ideal DF should be and I feel I should explain it more. In ideal world high DF would be great - but in real life it is often achived by strong global negative feedbacks resulting in tons of TIM and unpleasant bright hi-fiish sound. Recently many companies design amps with shallow feedback or no feedback at all. Yes DF is low but they sound great.
About power: Quality of sound is inversly proportional to amp's power (revolutionary statement!). 100W amp sounds better than 200W amp - let me explain: For the same amount of money you can find better sounding 100W amp than 200W amp. There is a lot of estate needed to make 2x more power but you can barely hear the difference when power is doubled.
In addition to all this - different amplifiers (and technologies) sound so much different - don't go by the spects. Icepower has good spects but specific sound - I like it (and have it) you might hate it. |
There's a lot of myth about power, gain, and DF. There is no universal, all encompassing rule. The age old JBL reinforcement manual recommends an amp with 2x the rated power, in watts, of the speaker at like impedances for optimal driver control. Dare somebody with a pair of original Ohm Walsh's to try putting 2x the power into them!(If you're silly enough to try they still make replacement parts). Bottom line, read the specs only to make sure you're not doing anything dangerous, then trust your ears. It's far to easy and tempting or manufacturers to confuse, hide, or outright lie to end users on the spec sheet |
Damping factor is is an amplifier's ability to control a woofer's motion after the drive signal has ceased. For example: if you drive a loudspeaker with a bass drum whack, the woofer's inertia and resonance in the enclosure will cause it to keep moving after the signal has died away. a This is a form of distortion that alters the music signal's dynamic envelope. The diaphragm of the Maggie is so light, and it's excursion so short(because of it's surface area) that the damping factor of your amp is virtually a moot point. The MMg's are only rated down to 50hz anyway(they're already rolling off at 80hz). Bottom line: Don't buy an amp to go with your Maggies based on damping factor. I'm using a Hafler TransNova 9505(255wpc) to drive my subs because of it's high damping factor(1000) and fast slew rate(150v/ms). Not an expensive amp at all(for high power and damping factor). BUT- I love tubes for my Maggie mains, and my Cary SLM-100s do a wonderful job. I've driven the Maggies full range with the Carys(and the Hafler for the heck of it) and they don't sound bad at all. Nothing beats the dyanamic range and transparency I get by bi-amping though. One more thing- Damping factor is related to the amplifier's output impedance. The lower the output inpedance, the higher the damping factor. The impedance of the cable between the speaker and amp decreases the effective damping factor of your amp. Keep you speaker cables short. |
DF will not control the driver excursion. The driver will move somewhat with inertia when the signal stops regardless of DF. What happens is that the moving driver acts like a microphone as the cone moves in the voice coil creating an emf. If the amplifier DF is very high, it's low output impedance presents a short to the speaker. This prevents a back emf to the speaker which could cause the driver to oscillate as it repeats the process.
What stops the cone (or controls it) is the air mass loading on the driver from the speaker enclosure (Q). Also, the crossover and the voice coil add to the overall DF in addition to the speaker cables so the amplifier DF will change once hooked up to a speaker. And with a planar speaker, the air loading is quite high which helps things out with low DF amps.
Then there are the unavoidable compromises in amplifier design: for an amplifer to have a high DF it has to have a very low output impedance which then will require more global negative feedback. This creates another problem to deal with in terms of distortion. Bottom line is that numbers don't tell you much. |
There are reasons (immaterial here) why some ribbon devices may sound slightly better with a medium DF than an extremely high one -- hence the use by some of tube amps.
Otherwise, your system looks fine on screen -- so the problem must be in the listening! What's bugging you? |
While Back EMF is not completely irrelevant in this discussion, I didn't feel it warranted inclusion in the definition. For a better understanding of "damping factor", see these: (http://en.wikipedia.org/wiki/Damping_factor) Particularly pg 1, para 2,3 / pg 4, para 1 (http://www.classic-audio.com/marantz/mdampingfactor.html) Notice how often the word, "control" is used. Tube amps almost invariably have low damping factors. To say that Maggies need a high one, but that they sound best with tube amps is contradictory. Here's an interesting little article on the MMg's you may enjoy Hlgoh: (http://www.enjoythemusic.com/magazine/viewpoint/1199/donibbles.htm) |
It's interesting how those three different URLs came out with the same "wikipedia" addy in the post! Hlgoh- The main point is: Don't be confused by peoples' opinions. Most only have those and not much in the way of truth. Do the research, and make decisions based on facts. If you want truth in your music reproduction: Listen to/get familiar with a lot of live music(amplified AND acoustic), and base the judgement of your system's reproduction on that. Enjoy your listening!!! |
The idea of damping factor having an effect on the sound of a speaker is mostly myth. However, it is a bit tricky seperating myth from reality. Here's what is really going on:
Almost any amplifier driving a speaker will have an output impedance that is lower than the speaker, but the important thing to understand is that the speaker is reproducing signals from the amp. As the output impedance of the amplifier approaches the impedance of the speaker, the distortion of the amplifier rises. It is this increase in distortion that accounts for the muddy sound that results- the so-called 'loss of control'. IOW the amplifier does not 'loose control', it simply distorts more.
To really understand why this is so, you have to understand the fundamental operating principle of the speaker itself. When the amplifier puts power through the voice coil, the result is that the diaphragm is deflected. The deflection is related to the power that the amplifier is making at that instant. As the amplifier makes more and less power, both positive and negative, the speaker diaphragm follows the waveform of power presented to its voice coil.
IOW the diaphragm is powered **all the way** through the amplified waveform. There is no place at all where the amplifier 'lets go' of the speaker. In understanding of this fact is also the understanding of why 'damping factor' is one of the more misunderstood myths of audio. |
Atmasphere- The URLs that I mentioned above contain sites and papers written by engineers in the fields of audio and electronics. Are we to believe that their understanding of damping factor as it relates to speaker control is wrong,and you are right? |
Atmasphere- I'm re-reading your posts in these forums and trying to get a handle on your resistance to the idea of damping factor and speaker control. In another thread you advised the poster to avoid his "very long" speaker wires because his system would be negatively affected. What is your understanding of the effect that long cable would have if not raising the impedance of the amp/speaker interface, thereby lower the damping factor? Perhaps our semantics disagree? |
These comments from recent audio shows are about rooms that used an amplifier having a damping factor of about 1.2 (that's "one point two"):
"Bass was excellent. I think I was told [the speakers] went down only to the mid 30s (I could be wrong) - sounded lower than that." - Paul Folbrecht on AudioCircle, RMAF '06
"The bass was powerful and tight while the overall presentation was wall to wall and incredibly precise." - Josh Ray, Sonic Flare RMAF '07 coverage
"Of all the rooms at the show, big or small, this room... [was] the finest thing I heard." - Thomas Portney, reader comment on Stereophile RMAF '07 blog
Duke
dealer/manufacturer |
I'd bet the farm that those rooms utilized speakers of high efficiency and extremely low mass. Can you be more specific(amps/speakers)? There are a lot of designs out there that cater to the SET lovers for instance. Speakers that(by design) don't require a lot of power or damping factor. |
Rodman99999, in the thread wherein I mentioned speaker cable length, I also mentioned 'characteristic impedance'.
As to your earlier question, the answer is 'yes'. I can point to a number of other examples wherein we find that a mythology has been created- where a mistake was made, something was considered negligible that later turned out not to be, or where something was said because a lot of money was involved.
Reread may statements about how the speaker works. Anytime a speaker is reproducing a signal, there is a cone position that corresponds to the amount of power in the voice coil. There are no musical signals that require full excursion and then have the amplifier 'let go' of the speaker. At any time when a speaker is making sound, the amplifier is **powering** the speaker to do that, whether it is full excursion or very nearly rest.
There is a story about the Emperor's new clothes. It is there to allow you to understand when something is being made up, and to see through it, even when the entirety of the culture is saying that the made up thing is real. Yet, the fact that an amplifier puts power through a speaker cannot be denied. So- you can go ahead believing the 'damping factor' story and that will not hurt you, but OTOH if you tend to look at these things closely, you will see after a while that that story can generate more questions than answers. |
Atmasphere- I've based my belief in my understanding of electro-magnetic fuction and Newtonian physics. Inertia= When a body is in motion, it tends to stay in motion unless acted upon by an outside force. Yes- the music signal from the amp is driving the speaker(electro-magnetically). While this speaker is going through the motions of reproducing sound, it's mass is also trying to continue to move in the last direction(large transient) it was told to go by the amp(inertia). Of course- the higher the mass, the greater the inertia. The amp's ability to counteract this tendancy determines how faithfully the signal is reproduced. The effects of back EMF(and, perhaps, quantum physics) not withstanding- Is there something flawed, or mythological to this? If I am wrong(unlike some), I like to be corrected. Truth is more important to me than holding stubbornly on to a misconception. Please- enlighten me. |
Rodman99999, I'll take you up on your bet. My house up against your farm. Here goes:
The speakers used were 92 dB efficient two-ways.
The woofer in the RMAF '06 room had a moving mass of 79 grams; Fs = 40 Hz; Qes = .35; Qms = 11.69, and Vas = 77 liters.
The speaker in the RMAF '07 room had two woofers, each with a moving mass of 41 grams [combined moving mass of 82 grams]; Fs = 31 Hz; Qes = .32; Qms = 4.35; and Vas = 121 liters [combined Vas of 242 liters]. This speaker is a bipolar, so the efficiency is 3 dB lower than the combined T/S parameters would indicate.
I am the speaker designer (thanks for asking!). The amplifier was an Atma-Sphere S-30.
So tell me.... just where is my new farm located??
;o)
Duke |
Not exactly INefficient though, are they? So tell me: Were you, by any chance, using any kind of auto-xformer between the OTLs and your speakers? Gotta keep digging for factors here before I give up any ground! |
Rodman99999, on just about every speaker that I can think of, the suspension gets stretched as the speaker diaphragm moves away from the resting point. Furthermore, the speakers that have the most back EMF, i.e. the ones that are the most reactive, tend to be high efficiency designs, not low efficiency. In most high efficiency designs, the speaker is usually moving less, not more, due to the nature of the design.
The back EMF of such speakers is often the reason that amplifiers with large amounts of feedback (and often higher 'damping factors') usually sound more shrill on these speakers, as the back EMF becomes an unintended part of the feedback signal.
This is probably not what you would have initially thought. I know it was not for me! It turns out to get highly reactive speakers to calm down, you need an amp without feedback, i.e. low damping factor. In fact what you start to get about this is that the damping factor is for the most part irrelevant, whereas the amount of loop feedback is (the less the better).
So anyway, the issue of stopping the mass of the speaker is almost non-existant, unless you intend to amplify non-linear /non-musical signals. An example of that might be a linear motor in a disk drive. Damping in a situation like that would be quite important as the driving signal is not a function. But for audio, the amplifier is always sending the speaker a signal (power), and this power is always moving the cone incrementally in a new place.
To be otherwise suggests that the signal is not a function (which it has to be by definition). Like I mentioned before, the amp does not push the speaker to full excursion and then 'let go'. Since the signal is a function the cone is literally under power all the way back to zero and then beyond.
When I started this business I believed in damping and a lot of other things that I later found to be made up. I was fortunate though that I had enough exposure to the right classes in college and had read enough texts from the 'old days' that I was encouraged to look past the 'company line' to see what was really up. Turns out there have been two design paradigms in conflict in audio for a long time:
http://www.atma-sphere.com/papers/paradigm_paper2.html
And I definitely subscribe to one and not the other, just like everyone else in audio :) |
I don't know anything about the technical stuff, I do know that my speakers need low damping to sound their best 9they are almost universally used with tube amps), preferably around 10, much higher and the bass sounds choked, with pther speakers you need high[er] damping or the bass sounds loose and out of control -- so I imagine the answer depends very much on your speakers and how they are designed. |
Audiokinesis- It's been a few years(like maybe 26) since anyone threw any Thiele-Small parameters at me, so I'm still picking up the pieces of my shattered brain here! From what you're describing: I'm going to guess the woofers are about 8" based on the Vas? If that's so: with a fairly high mechanical(11.69) and electrical(.35) system loss, they sound like they might have a relatively stiff compliance. The moving mass isn't all that high, so I'm thinking a fairly short throw(or single/flat wound) voice coil. If I'm all wet here, I'm certain you'll let me know. I don't expect any mercy!! |
Hi Rodman99999,
I don't blame you one bit for fighting to keep the family farm.
92 dB is higher than average efficiency, I'll admit. Whether or not it qualifies as true "high efficiency" is a judgment call - over on Audio Asylum they discussed at length where to draw the line, and if I recall correctly the consensus was 96 dB. You and I will probably have to hire lawyers and go to court to settle this issue. Or maybe we could have a cage-fight? We could sell tickets and split the proceeds, maybe both be able to buy some new gear (allow me to recommend a nice low-damping-factor amplifer...).
Sorry, couldn't resist.
Okay, back to the RMAF rooms. No autoformers were used. The only things that made the speakers "low damping-factor friendly" were a smooth impedance curve and a fairly low port tuning. The port tuning is actually user-adjustable, so the speakers can be adapted to different room acoustic situations, but this feature is also useful in amplifier matching. With a solid state amp, a higher tuning frequency would probably work better.
Duke |
Funny thing Atmasphere: I've got that very paper saved on my computer. I've never liked negative feedback loops, probably what endeared me to David Hafler(his early ideas). I wondered if there was a connection between your name in here and the OTLs. The lack of a hyphen threw me(not so hard to do at my age). |
Audiokinesis
Thanks for the info on Atma-Sphere S-30 - I checked and it looks very very interesting.
|
Rodman99999, I chose those woofers based on low thermal compression, suitable combination of bass extension and efficiency in a size box that I could live with, and sufficiently smooth response between 1 kHz and 2 kHz. I'd have liked higher voice coil DC resistance in both cases, but went with what was available. These woofers were not picked out as being uniquely "low damping factor friendly".
The heavier-cone woofer is a 12" with plus or minus 7 mm of linear excursion, and the lighter-cone one is a 10" with plus or minus 6 mm of linear excursion. Both are prosound drivers with unusually low efficiency (and consequently good bass extension) for prosound units of that size; most prosound 10" and 12" woofers are really midranges with efficiencies in the upper 90's.
Duke |
Thanks, Kijanki.
One other thing I should mention: I don't intend to give the impression that a low damping factor amp will work with just any speaker. I'd want to see the impedance curve and possibly the frequency response curve before making that call.
That being said, a couple of weeks ago a friend of mine wanted to try a low damping factor amp on his 89 dB, 5-to-20 ohm two-way monitors. I was skeptical because of the impedance curve. He liked the sound, but there was a lower treble emphasis associated with an impedance peak in the crossover region (which I had predicted, having seen the curve). So I calculated and assembled a circuit that smoothed the impedance peak, and that solved the problem. Maybe that would work with other speakers as well. I'd like to adjust the bass tuning in his speakers also, but we haven't gotten to that point yet.
Duke |
It's interesting to note that the highly regarded Pass X350, X600 and X1000 all have damping factors well below 100. Thanks, Duke, for your insight. |
Audikinesis
DF=1 is a little extreme. On the other hand Atma-Spere design does not have transformers and capacitors in the signal path !!!. I would stay away from amps with DF>500. Icepower, I use, is a different beast and has inherently high DF (speaker always connected to V+ and GND - only direction changes). In addition TIM is a result of the charge traped on the junction of output transistors when overdiven momentarily - does not happen in Icepower.
Some amps with high DF (suggesting deep feedback) don't sound harsh or bright but make you tired after a while. It is because our brain fills the gaps when amp is choked-up after transitions (TIM). I would stay away from very good spects - something has to give. TIM was uknown 40 years ago!
As I remember, long long time ago 16 Ohm or even 32 Ohm speakers were typical and damping was easier. Lower impedance came with solid state's (I suspect) desire for higher power. Also, underhung speakers (narrow coil in wide gap) are rare inspite of their lower distortions - is it cost? |
I agree with Duke - amplifier damping is secondary to box and woofer selection...by a long way...it does not surprise me that pro drivers with underhung coils would sound tight as the woofer is always fully in control of the amp as Atmasphere describes (sees full force of the drive signal throughout its excursion and therefore behaves in a very linear fashion). Poorly controlled woofers with long coils in narrow gaps will be less well controlled no matter what amp they are connected to (as the coil moves in and out of the linear portion of the magnetic field they see less drive signal).
An under damped speaker may gain from a higher damping factor (10 or so), however a critically damped or over damped speaker is unlikely to benefit much from damping factor (it becomes a moot point as the mechanical behavior dominates).
However, a low damping factor (or high output impedance) coupled with a speaker with a large swing in impedance will certainly tend to emphasize the "peaks" in the speaker impedance response (often resonance for a woofer but a port can change this).
When under damped speakers have a typical bass hump in frequency and also have a high impedance at their resonance then a low damping factor amp will almost certainly give you bass bloom as both factors discussed above may combine to give much stronger bass. |
