1. Things you need to know about the new ‘Conversations’ PM system:

    a) DO NOT REPLY TO THE NOTIFICATION EMAIL! I get them, not the intended recipient. I get a lot of them and I do not want them! It is just a notification, log into the site and reply from there.

    b) To delete old conversations use the ‘Leave conversation’ option. This is just delete by another name.
    Dismiss Notice

C´mon you diffraction experts.

Discussion in 'audio' started by Chris, Jan 12, 2019.

  1. Chris

    Chris pfm Member

    When I built my WD25T speakers (see my avatar) I remember consulting Peter Comeau, the designer about modifying their dimensions because, accustomed to my previous Totem Arros, they seemed very wide. His reply was not to touch their width because it would affect their baffle step. In the end I changed their height and depth a bit but respected the baffle width. Or so I thought. At that time I decided to copy a Troels and Gravesen model by putting go faster stripes on each side of the tweeter from top to bottom inb an attempt to make them look more svelte. Several years of 95% satisfaction but always with a grumbling doubt about their ability to project a believable image, I now realise that my go faster stripes, which are in fact grooves cut along the length (height) of my boxes and hi-lit by a pro lacquer job might well be also tantamount to having reduced the baffle width in fact or at least having disturbed the proposed diffraction reducing effect of the rounded extreme adges of my boxes.
    Before I go about trying to do some marquetry to fill in the grooves with dark wood strips, am I "whittling about bugger áll " or might this just be affecting certain higher frequencies and wrecking the imaging.

    Before you ask, yes, I am a nerd but I don´t own an anorak. Not much use where I live.

    Not wishing to cast aspersions on a T&G model, I do believe his stripes were made with inset wood but I didn´t twig why at the time.
     
  2. h.g.

    h.g. pfm Member

    Baffle step correction concerns the frequencies over which sound changes from radiating largely forward to radiating equally in all directions. Your grooves are irrelevant to this but they will be a source of diffracted high frequency sound but the level is almost certainly at high enough frequency and low enough level to be inaudible.

    A tweeter that isn't flush with the baffle creates diffraction ripples that are audible as shown here. However, the distance from the tweeter to the diffracting edge is the same along the whole length of the edge and so at a listening position on the axis they will sum together enhancing the peaks and dips. Off axis they will sum differently with a fair amount of cancellation resulting in a wigglier response but with substantially reduced peaks and dips. A pity this isn't shown in the link.

    In your case the distance from tweeter to groove varies along the length of the groove and so there will be plenty of cancellation of what is likely a lower level of diffraction anyway compared to an edge. If you have a microphone the strength is easy to determine by simply measuring with and without something like blutak in the groves.
     
    tuga likes this.
  3. awkwardbydesign

    awkwardbydesign Officially Awesome

    Tape a sheet of something over the grooves and listen again. Card might do.
    Oh, and I put 2mm diameter brass bars across my tweeters to protect them, and had to remove them as I could hear them!
    [​IMG]
     
  4. Chris

    Chris pfm Member

    What about some anti diffraction rings in thick wooden felt on the Millennium faceplates ? Also I’m going to try filling in the groove from woofer edge to the top with some non hardening black plastic wood that I have used in places on my wooden flooring to hide woodworm damage. It’s water based so easy to get a good finish without sandpaper and virtually undetectable unless you get really close up. More anon.
    Thanks for the info.
    Chris
     
  5. YNWOAN

    YNWOAN 100% Analogue

    If these are the speakers in your avatar then the edges of the cabinet to the bass driver are already too close to properly avoid diffraction effects. Rounded cabinet edges have to be of a surprisingly large radius to avoid s condary difraction and even the drive roll surround and basket edge introduce annomolies. The tweeter is another matter although the symmetrical position means that edge artefacts are potentially doubled. Experimentation is easy to achieve though.
     
  6. Chris

    Chris pfm Member

    God, I feel humbled. 5 years since I built my WD25Tex speakers and five years enjoying them to 90% but always knowing there was something weird going on and blaming myself, the room, the weather, the Gods, crappy vinyl copies etc. etc. It was the presence of my go faster grooves; I have just stuck 4 strips of soft isolating tape to cover them up and the whole ball game has changed. That bloody diffraction comb effect is really nasty. Now all is well and smooth and complete and with a much greater impression of depth. God, I feel a right berk.
     
    Rug Doc likes this.
  7. h.g.

    h.g. pfm Member

    If you are happy with the improvement that is obviously the main thing but did you take any measurements to quantify the effect or was this all done by ear?
     
  8. Chris

    Chris pfm Member

    By ear. I’m not of a scientific bent but the difference is unmistakeable. I don’t have much use for a microphone but imaging is more stable and somehow each note of each instrument seems to have a firmer foundation to it, more complete as if easier to understand why. Singers’ diction is much improved e.g. Patricia Barber on Night Club at the moment and she actually finishes her words, pronouncing past tenses and not slurring. In a word the error correction circuits in my brain have less work to do.
     
    fatmarley and h.g. like this.
  9. awkwardbydesign

    awkwardbydesign Officially Awesome

    Beautifully put. For most people, I expect that type of analysis is of much greater value than any amount of measurement info. :cool:
     
  10. h.g.

    h.g. pfm Member

    Failing to identify the origins of what is heard and hence distinguishing between what will remain constant over time and what is likely to change does not look good value to me. The problem is that false conclusions are likely to be drawn which then form the basis for poor decisions in the future. The audiophile phenomena of course relies on this in order to exist.

    Having said that, it is perfectly reasonable to take the approach of enjoying the journey while remaining ignorant of what is going on in a technical sense so long as you are aware that you are ignorant. I am happy to do this in a wide range areas but it seems rare for audiophiles to do it with home audio (or at least those that chat on audiophile forums and go to shows). They tend to want assign all perceived sound to what is waggling their eardrums and zero to other sources. So when what is waggling their eardrums isn't changed by enough to be perceived they substitute a belief in magic for a belief in science. The magic goes by a variety of aliases including "science" (audiophile redefinition of the meaning like many words), "stuff unknown to science", "don't know but it is real and exists in the physical sound because I heard it", etc...
     
    tuga and Tw99 like this.
  11. AndyU

    AndyU pfm Member

    If the speakers are arranged as in your avatar, then the huge bureau next to them and the wall behind them and quite possibly the pot next to them will be having a non-trivial effect on how they sound also.
     
  12. sq225917

    sq225917 situation engineer

    Better picture, mid bass model number and cab width around mids please. There's online calcs for this very task. A good measurement is worth a 1000 insights though.
     
  13. Chris

    Chris pfm Member

    the speakers were designed to go as close as poss to the rear wall, the bureau and it’s place in the room came with the missus and the pot usually has a palm tree in it. Nothing’s perfect.
    Regarding the scientific slant, I prefer to let others assimilate the theory and then translate it to something I can relate to. For example, hg’s summary of bafflestep in the first reply to my posts is perfect afaiac. Indeed it was on reading this that the penny dropped and I thought of the possible negative effect of my grooves. Let’s if I remain as impressed after a week or so.
     
  14. Chris

    Chris pfm Member

    Seas Prestige H1411 A26re4 paper cone 26cms . Width of box 29 cms. Tweeter Seas Millennium. Grooves, width of table saw blade 3mm approx width and depth.
     
  15. awkwardbydesign

    awkwardbydesign Officially Awesome

    No, measurements have to be interpreted. But of course engineers (and measurists) are uncomfortable with this.
     
  16. AndyU

    AndyU pfm Member

    And the bureau?
     
  17. orangeart

    orangeart KJF Audio Ltd.

    I think the lack of clarity likely just comes down to the design using a 26 cm woofer. There aren't many designs out there doing a 2 way with a woofer past 6 inches that can do imaging that well.
     
  18. Chris

    Chris pfm Member

    The bureau is not negotiable and as far as I can tell sound is affected more by my listening position, head against the opposite wall, is much more toxic.
    Anyway, I was talking about my grooves which no longer exist thanks to my isolating tape,
     
  19. AndyU

    AndyU pfm Member

    If two wee grooves make such a huge difference, why wouldn’t sound hitting the corners of a huge bureau cause far greater diffraction unpleasantness. You owe it to yourself to move the bureau. Maybe when she’s out.
     
  20. Chris

    Chris pfm Member

    But they are much nearer the tweeter. I don’t doubt the bureau is doing its bit but I reckon my boxes are pretty impervious to their surroundings. I have never heard them boom in their lives.
     

Share This Page

  1. This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
    By continuing to use this site, you are consenting to our use of cookies.
    Dismiss Notice