Diffraction and its effects . .

[jimdgoulding]

is something I've made a study of and custom make an affordable after market product for its removal. For those of you who don't know, diffraction is the reflection of waveforms shorter in length than the width of your speaker cabinets or those produced by your tweeters. This is not something owners of Gallo speakers, for example, or any baffleless speaker to include planars, need be concerned with. It is for the owners of box speakers with right or near right angles. Diffraction in action:

http://www.silcom.com/~aludwig/images/diffdem.gif

Those green and red circles in the illustration by Al Ludwig are not a part of any recording that you own and are additive in your frequency and time domains. They are redundant and late arriving, out of time and phase. Some speaker manufacturers, not very many, compensate for a rise in the frequency domain caused by diffracted waveforms summing in by reducing the tweeters output just above of the crossover, but the time and phase anomaly persists to the detriment of the clarity and dimensionality inherent in good recordings.

I use organic wool and individually fit to the dimensions of most any speaker baffle and offer with a 30 day trial period. You should give it a try. You just might find that it is the best value audio investment you've ever made.

Some user and reviewer comments can be seen on the back page at my website: www.diffractionbegone.com.

An example can be seen in my avatar.

[zanash]

think you need to go back and re check your assertion visa vi.. planars and baffels don't suffer.

I to have studdied this effect. its a simple matter to show that a wave propgated from a plane on reaching an edge is refracted .

you can set it up in your sink if you have a mind ...

[jimdgoulding]

Zanash- Hi. Kindly give more information. I can't fathom what you mean by kitchen sink. Thanks.

[StanleyB]

Aha, the good old water ripple experiment. Didm't that Dutch guy Huygens did something similar?

Good luck to you Jim with your R&D and product development :).

[YNWaN]

I don't know if you will read this Jim, but Zanash is referring to the classic wave propagation/reflection experiment that involves making ripples on water and looking at how they are reflected.

Planar speakers are not entirely free of a baffle though, as you suggest. At best they have a vestigial baffle - very few speakers have no support frame to hold the drive units in place (MBL perhaps and a couple of others). Planar speakers still produce horizontal dispersion and any edge will cause secondary diffraction to occur (dipole operation doesn't stop this).

[John]

When I had standard box speakers this tweak really helped

[bobbasrah]

From the drive there is frame resonance (principally from the HF drive), and from the cabinet there is case resonance. The diagram implies there is a short wavelength waveform travelling almost like a skin effect from the tweeter. Is this intended to indicate the drive frame resonance?

With tweeters face mounted on the baffle, could the drive resonance not be better deflected by a stubby cone fitted around the drive or rebating the drives?

As my own DIY's were deliberately built with extra thick mdf and chamfered on all edges to reduce and mitigate cabinet edge resonance, and the flush faced drives filleted in rebates to deflect drive frame resonance, I thought I had dealt with all the possibilities.

And no, I am not putting them into the bath to double check..........

[YNWaN]

From the drive there is frame resonance (principally from the HF drive), and from the cabinet there is case resonance. The diagram implies there is a short wavelength waveform travelling almost like a skin effect from the tweeter. Is this intended to indicate the drive frame resonance?

No, the diagram represents the actual sonic output of the drive unit and the secondary diffraction of this sound wave as it meets edges such as the edge of the cabinet, or tweeter mounting plate etc.


With tweeters face mounted on the baffle, could the drive resonance not be better deflected by a stubby cone fitted around the drive or rebating the drives?

Both of these options will effectively horn load the tweeter to some extent and significantly alter its output. Even then, secondary diffraction effects will still exist at the edge of the cone, or the hole that the drive unit is rebated into.


As my own DIY's were deliberately built with extra thick mdf and chamfered on all edges to reduce and mitigate cabinet edge resonance, and the flush faced drives filleted in rebates to deflect drive frame resonance, I thought I had dealt with all the possibilities.


Both of these are excellent techniques for dealing with this problem. However, a sharp edge of a 45 degree chamfer will still cause an element (but reduced) of diffraction. Many speakers have rebated tweeters but relatively few use chamfered cabinet edges.Even then, in many cases, there exists the potential for diffraction to take place where the actual tweeters are mounted to the tweeter face-plate (although Jim's solution doesn't address this issue (some tweeter manufacturers seem to consider it more than others).

[bobbasrah]

Ok Mark understood that this is from the drive element itself, but is this an effect only with HF and tweeters, and with respect to all tweeters?
The reason I am asking is that when I checked this out with the manufacturer of the ribbon tweeter, the response was that such incorporation was unnecessary due to the cavity design, but if I had other reasons to do so (which I did) then proceed. I supposed it is a form of waveguide...

[YNWaN]

The effect becomes less relevant as frequency descends and the wavelength lengthens.

I would say that what the ribbon tweeter guys have told you isn't really true as such. Mounting the ribbon in a recess doesn't directly address the issue, which is endemic to the propagation of high frequencies. However, the more directional the output of a tweeter is, the less likely the problem is to occur.

[sq225917]

I found Jim's diffraction guides to be very worthwhile on my ES14's they even had an audible effect down into the mids IME.

[bobbasrah]

The effect becomes less relevant as frequency descends and the wavelength lengthens.

I would say that what the ribbon tweeter guys have told you isn't really true as such. Mounting the ribbon in a recess doesn't directly address the issue, which is endemic to the propagation of high frequencies. However, the more directional the output of a tweeter is, the less likely the problem is to occur.

Ok, so it applies to HF only.
Thanks for that.
I can envisage the phenomenon for domes etc where the radiation from the drive is pseudo-radial, and however fractional the power, may interefere with angled surfaces which are tangential near the mounting surface, but beyond that I'm getting a bit confused as to my understanding of how a radiator behaves in a waveguide, the resonant frequency of the casing, and the frequencies involved.
Can you explain please as kinda lost now.

[jimdgoulding]

Nice discussion, fellas. I do understand the ripple effect, of course. The purpose of the wool pads is to absorb and dissipate short and spreading or rippling high frequency waveforms produced by surface or flush mounted tweeters for those who are following the discussion and may not be quite so scientificly endowed, so they are not re-directed into the soundfield arbitrarily (see link) and distorting what the microphone(s) captured on location. I've done my homework. They work real good. Again, great discussion.

http://www.silcom.com/~aludwig/images/diffdem.gif

[jimdgoulding]

You mean these, Sq (at the bottom)?
http://surrounds
Thanks for chiming in. Here's a review from Filterlab posted right here some months ago:

"Like most audiophile chaps, I'm definitely up for a tweak which is preferably all of the following things:

1. Effective
2. Easy to install (i.e. doesn't require a degree in electronic engineering!)
3. Cheap.

So as if by chance the perfect tweak offer dropped into my lap seemingly out of nowhere! After extensive listening and loving of my Revel M22s and having tried them on different surfaces and with different cables etc etc I was very VERY pleased with their performance, surely nothing could make them even better?

Well yes actually, and for a darn sight less than they cost to buy. In fact an amazing tweak was had for just 1.4% of the price of the speakers. YES! For £28 including taxes and shipping, my esteemed acquaintance Jim from diffractionbegone created a pair of baffle covers to perfectly fit my speakers.

So the science is not really up my street, but what I can hear are the effects of a scientific principal working in my favour, and boy can I hear a difference with these things! Have a look and see how they fit, and I'll tell you how they effect (read as 'improve') the sound.


(Sorry, couldn't bring the pictures forward but an example can be seen in my avatar)


Now you may balk a little at the looks, but in the flesh they look superb and if I'm honest the looks have really grown on me. However, it's not about the looks I hear you say. So the sound...

Well my friends, the difference these things have made is nothing short of staggering. I was expecting it to effect the treble, but in actual fact it's quite the opposite.

The first thing that really struck me was the midrange clarity - think 'crack of the snare', 'clap of the hand' and 'strum of the guitar' and you're on exactly the right lines. The focus in this audible region has to heard to be believed, but that's not all - oh no! The other amazing element is in the bass lines, particularly bass guitar, which takes on a start/stop nature that is as tight as the deft handwork as any talented guitarist. Again I'm going to use the superlative 'focus', clarity in this region is so improved, amazingly in fact.

So is that it? Nope, there's two other areas which are even more impressive, namely imaging and vocals. Imaging on my Revels was always a little bit of a sticking point after owning some Martin Logan electrostats, but these baffle covers have augmented the imaging to near-electrostat levels, and that's saying something. I guess this is why vocals sound sooo good, they just stand so clear from the mix with such audibility and yet never become overblown.

If you have thirty quid that you fancy putting towards a positive tweak in your system, you'll NEVER go wrong with these."

[jimdgoulding]

Le petite bump for the preservation of sound waveforms and the information they carry, all your system is working in earnest to produce.

[sq225917]

Check out my ES14's in my previous house, now there was a well damped room.

[goraman]

I found the use of wool refraction rings to remove slight sibilance and reduce off axis dispersion from the corners that can distort the sound stage.
I used some wool felt circular ones from Madisound.
I wrote the review. http://www.madisoundspeakerstore.com/acoustic-damping/dampers-felt-defraction-ring-for-tweeters/#product-page-reviews
There pretty thick wool felt and fit perfectly over the metal body of my tweeters and don't look out of place,you don't even know there there unless you look closely.
They are cut from the wool felt they sell in bulk and have an adhesive back, I talked to the Tech that cuts them out with a die tool.
It's well worth the effort.