Interesting read...FMD (doppler distortion) doesn't exist

[DrWho]

Haha, I was kinda hoping the title would catch some people's attention. I came across this article today that proved the whole concept of doppler distortion is false...however, the true phenomenon of phase modulation distortion does exist and its effects are identical to that of frequency modulation distortion. Just thought y'all might find it a very interesting read (PWK was the first reference on his list). Ok, so here's the link:

http://sound.westhost.com/doppler.htm

To sum up the article, the modulation due to cone movement occurs at the peaks of excursion, not at the midpoint where diaphragm velocity is greatest. The resultant modulation is due to differences in phase of the HF signal riding on diaphragm producing the LF signal.

There are also a bunch of other articles on this site that discuss the technical aspects of all sorts of audio related material. There are also many project designs with explanations and even kits you can purchase to build them yourself. Here's a clickable link:

http://sound.westhost.com

Enjoy!

[D-MAN]

If you really want to strain your noodle, the fact that air compresses easier than it "fills" back in a rarefaction is another physics problem associated with unequal excursion, FMD and its related kin. Now my head hurts...

DM

[sunnysal]

the writer proves that the term "doppler" was mis-applied, not that there is no effect...he shows a mild, but measurable, distortion based on the cones movement...PWK claimed there was no distortion at all...so...forget PWK and the words doppler and we still have the issue of a cone´s movement causing audible distortion to an input waveform...right? tony

[djk]

Posted by djk ( M ) on August 31, 2004 at 03:20:48

In Reply to: Re: I don't think so. posted by Steve Eddy on August 30, 2004 at 12:26:49:

Is this a bit like 'electron flow' vs 'current flow'(or 'watts RMS' vs 'average power')?

If so, then rather than argue whether FMD is technically correct or not I am going to start calling it 'Gargle', after all, that what it sounds like.

--------------------------------------------------------------------------------

Posted by djk ( M ) on September 06, 2004 at 01:00:05

Gargle! (the distortion previously known as 'Doppler' or 'FMD') is back in the news.

Linkwitz has examined Rod Elliot's claims, and states: "Either way of looking at the situation leads to the identical magnitude of the spectral components."

He goes on to examine the Seas W18E001, the very expensive 6-1/2" driver with copper shorting rings in the motor.

At 50hz with 5mm peak excursion, a 2Khz tone is examined.

Using the Elliot method gives us a 9.2 Hz deviation of the 2000 Hz tone... could generate up to 12.9 dB change in its amplitude.

"It is a linear form of distortion, at least to the first order, which means its magnitude is independent of the amplitude of the higher frequency tone and distortion increases linearly with the amplitude of the lower frequency tone.."

How loud do you think the 50hz tone was to cause a 12dB change in the amplitude of the 2Khz tone?

86dB !!!

Raising the crossover point to 200hz (a good point for a 6-1/2" driver) allows for 24dB more output for the same amount of distortion.

[DrWho]

Yes, the effects are the same and are the result of cone movement...I just thought it was extremely interesting that the system causing it has nothing to do with the speed of the cone. Only the title was supposed to stir the kettle and spark interest, not content of the post.

If pmd is such a nonissue then why does everyone claim that the reduction of cone movement is such an important thing (like how horns reduce the excursion)? Does it have more to do with the amplitude modulation? If so, does anybody have any cool articles like this that describe it in detail with pretty pics and all that?

[D-MAN]

The frequency modulation distortion (FMD) is the natural result of the imposition of 2 or more frequencies being simultaneously reproduced by the same cone mass. In essense, the lower frequency being the one "causing" the most cone extension as its natural effect, and the higher frequency being modulated upon it while it is in motion according to the lower frequency requirements and the physical limitations of the motor and cone structure. The lower frequency would control the excursion and the higher frequency would "ride" upon it. The higher frequency modulation would therefore be subjected to phase shifts (that is, non-signal related distortions) caused by the reversal of cone movement in either direction as the cone responded to the lower frequency wavelength. In reality MANY different wavelengths are produced simultaneously by superimposed modulations on the same cone mass and all but ONE (at best) are subject to phase shifts that are not signal related.

Due the motion of the cone mass, there HAS to be some resultant distortion of the waveform(s) caused by the imposed modulation. No way around that. Doppler shift measurements COULD be employed as they are certainly there. The question remains concerning FMD whether anything can be done about it aside from the discarding of cone (or diaphragmic) drivers.

But perhaps a larger concern of PWK was the EQUAL (or balanced) excursion of the cone mass in both directions, in particular because of his belief in the sealed back chamber for his front-loaded horn designs (Modulation distortion, sometimes called Inter-Modulation distortion or IM for short).

The nature of the reactance that a respective horn imposes at the throat causes the cone to experience a resistance to movement, in that the horn channels contained a certain volume of air, and that the horn channel walls and expansion rate all added resistance to the movement of air through it, and the idea was to balance the resistance in the front of the cone (at the throat cavity opening) with the same resistance to compression at the back of the cone (the back chamber). The air column in the horn is reactive in that it also tends to fill the rarifaction (lower pressure area) caused by positive cone motion in a more rapid fashion that "free" air of the same volume would as it is a more enclosed and restrictive space due to wall-surface resistance and constriction. Hence, more reactance at the throat "seen" by the driver. This resistance to movement serves to naturally reduce cone excursion per given electical signal.

Where less cone excursion results in less distortion, of course, there is a typical loss of SPL. Which is where the horn steps in with its high efficiency. Levels of distortion can therefore be directly related to efficiency, as PWK has famously made known in the past.

DM

[Klewless]

Does all this mean that the train whistle tone does not go up as it approaches, and down as it departs? Did anybody remember to tell the whistle about all this?

Now my head hurts.

[sunnysal]

thanks dr. who for posting the article, it indeed is an interesting subject.

unfortunately no one has really tackled the subject scientifically. Unlike the good old days few are willing to put the work into it.

even the article mentioned in the first post would not pass muster with many who expect findings to be subject to peer review by proper authorities before being published and quoted as fact, that article does not meet those standards.

regards, tony

[cwm]

"Where less cone excursion results in less distortion, of course, there is a typical loss of SPL. Which is where the horn steps in with its high efficiency. Levels of distortion can therefore be directly related to efficiency, as PWK has famously made known in the past."

I think the only possible exception to this would be the case of large planar speakers e.g. electostatics, which although relatively inefficient (85-90db) would have relatively low diaphragm excursion due to the very large surface involved and the ultralight construction of the diaphragm. To me electrostatics and well engineered horns have always led the pack for very low distortion.........

[DrWho]

I find it ironic though that you never see horns or electrostatics inside the recording studio

[cwm]

Not strictly speaking true, in Europe some smaller recording companies do in fact use ES speakers (i.e. Quad 57) and in the U.S./Canada Altec A-7s have been used. The late pianist Glenn Gould quite frequently used the Altecs to listen to the final mix down.......There is a picture of him somewhere doing so and if I can find it I will post it. The audibility of FMing or IMD at "high" sound levels with many direct radiators is I suspect a reality. Interesting discussion....Yamaha NS-10s are used in nearfield monitoring and you know what some engineers say.."If it sounds good on an NS-10 it will sound good on anything!"

[sunnysal]

doctor who,

you are WAY off my friend...horns are present in great numbers in many of the world's greatest recording studios...you can find offerings from JBL, Altec, Tannoy, TAD and Westlake (among others) all of which utilize horns for their professional studio monitors.

Regards, tony

[DrWho]

Hmmmmm, well I suppose there are always exceptions to the rule...Of the dozens of studios I've worked in and the hundreds I've seen online and the dozens I've read about in recording books, I don't think I've run across a single studio using hornloaded monitors.I have seen a few mastering facilities that use them, but never in the tracking and mixing stages. And even those studios that do have horns, I'll bet good money that the majority of them still have normal nearfield studio monitors laying around. I'm yet to find a horn that can be used in the nearfield and still sound good. Though I don't believe in the mixing schools, all of them in the world are using direct radiators in their classrooms as well.

Back at the college of dupage (local community college), they dropped a ton of money on a studio and the professor dude brought in a ton of oldschool monitors...two of which were hornloaded JBL and Altec speakers...I can't remember the model and I doubt they're still there. Anyways, mixes never translated well on them and good music sounded crappy on them too (it was a lot more than an amp or room mismatch too). Nothing like a good pair of Mackie 828's or Event 20/20's. Another popular one out there is the KRK V8 (haha, it always reminds me of V8 juice). I have a friend with a pair of Tannoy monitors, but he always ends up mixing on his NS-10's

Anyways, not to get into any debates or whatever, but you are the first person I've heard claiming that horn loaded studio monitors are popular.

As far as the research goes, there is no real incentive for further study because the calculated results have the same effect on frequency response. Also, it is a flaw that would be present even in a "perfect" speaker, which essentially means the "perfect" speaker needs to be redefined. We're at a point now in speaker design that we've already tweaked out an already faulty design...It's time for some new methods of sound reproduction.

[D-MAN]

DrWho, until now, I always thought you "knew yer stuff"!

But if you're going to be bad-mouthing horns, PUT ON YOUR GLOVES cause there's gonna be some trouble!

DM

[DrWho]

lol, im not bad-mouthing horns...they're just hard to use in a studio setting. If you've ever mixed you'll know what I mean. It's so extremely easy to get something to sound great on any system...it's another thing to get something to sound good everywhere at the same time. Though I greatly enjoy horns for my playback system at home and even for PA use, you're simply not going to get a mix that sounds good in the car or on crappy HTIB systems (where 90% of listening is done anyway). The best mixers in the world are putting their mix through dozens of different systems before sending it off to be mastered. The studios with built in far-field listening speakers (like the horn monitors are probably used for) are using it as a reference mix...just like the NS-10's, which are used for a crappy speaker reference. I personally use a pair of Minimus 77's for my crappy mix and they do a real good job.

But when it all comes back to getting the most revealing sound, them direct radiators seem to be the option of choice...probably because it's easier to hear the actual sound source. There's just something about horns that make them wierd to mix with. I attribute it to the throat distortions inherant in every horn loaded design (another one of those built in flaws that you can only minimize). Some people are extremely sensitive to that sound which is why there are so many ppl that love to bash horns. That is very much my own personal opinion, but it seems to be a trend with all the mixers I know. Btw, it's not like I haven't tried hornloaded monitors either. The best of which are probably the few pairs of tannoy coaxials, but I can't get around their hashy midrange sound that they all seem to have. If you don't notice it, then I'm sure you'll love them, but I always find myself trying to compensate my mix for that sound...which of course leads to a mix that doesn't sound good anywhere else.

I must confess that most of my time in the studio has never been a full out production...it's just been fooling around with friends, or mixing special projects that only a few people will hear. The first real recording session I've done was just a few years ago with a band called Finding Philip. The recording made radio airplay down in southern illinois, though I must confess we did have some connections there. Nevertheless, it's a fun bragging right

[sunnysal]

I didn´t want to brow-beat anyone Dr. Who but MANY studios use horns and I did not want to let your original statement stand uncontested since some may read the thread and take it as a fact that horns are not found in studios.

I only know NY studios and virtually every one I have been in has a pair of horn speakers in the control room, the live room or somewhere.

a few examples are:

Electric Lady (you know this one, right? clients like, the rolling stones, led zeppelin, madonna...),

Sound on Sound (a few clients, rod stewart, steely dan, sting...)

Midnight Modulation (clients like; the band, NRBQ, dave matthews...)

Sorcerer Sound (clients like; norah jones, sonic youth, lou reed...) I cuold not find a photo of these...but I remember they had horns in the mixing/control room...anyway horns are valid speakers for mixing/montioring/mastering IMHO...

regards, tony

[D-MAN]

I won't argue that, I don't really know of any horns that I would want to use for close-field work, although they evidently exist.

That certainly isn't what they (meaning most) are best at or known for.

So MAYBE you really DO "know yer stuff", after all!

DM

[djk]

"I won't argue that, I don't really know of any horns that I would want to use for close-field work, although they evidently exist."

JBL 4430, you can listen to it at an arm's length.

It makes a B&W 801 sound 'colored'.

They sound good in a big room too, the local multi-plex theater has them for L-C-R.

[DrWho]

----------------

On 3/24/2005 4:40:54 PM djk wrote:

"I won't argue that, I don't really know of any horns that I would want to use for close-field work, although they evidently exist."

JBL 4430, you can listen to it at an arm's length.

It makes a B&W 801 sound 'colored'.

They sound good in a big room too, the local multi-plex theater has them for L-C-R.

----------------

hmmm, never heard either of them speakers...i'll try to make a point to some day (and if i'm lucky, i'll get to A/B them on the spot). The real question though is which one translates the mix better? I'll take a colored speaker any day if it means the mixes come out great.

I read JBL's pdf of the 4430 and wasn't very impressed with the frequency response either...+-3dB is a lot different than the +-1dB typically in published specs. Yes I know, specs aren't everything and the response is rather smooth (not jagged at all), it just struck my attention off the bat. Ironically, I noticed that the HF section is a lot rougher than the low end... (the implication is meant entirely in jest btw, lol)

[DrWho]

Hey no worries about brow beating or whatever. I think the happy tone in the discussion gets lost in the text.

I just wanted to point out that the "horns" in those pics are all mounted in the far-field. Also, notice the Mackie 828's and the KRK V8's sitting up close on top of the console. I can almost guarantee that they don't play through the far-fields until they've got a basic mix established (though I'm sure every mixer has his own techniques too).