Announcement

Collapse
No announcement yet.

PR and air leaks

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • Scarface1
    replied
    Hi guys, i found out that only the bass at tuning frequency is lacking. I took out the passives and sealed the enclosure to make it a sealed enclosure and put a 50 hz test tone tru it. It get me around 76db wich is close to the simulated output ( 80db at 50hz with 4mm xmax ). The weird thing is that with the original enclosure with 2 nd105 passives, i only get around 78 db with the same power, while tuning point is at 54 hz. The passives are moving enough, how is this possible? I schould AT LEAST get 10db more! I really don't know what i'm doing wrong. I measured multiple speakers at the same spot and get the spl it supposed to have, even with the sealed design..

    Leave a comment:


  • Scarface1
    replied
    I tried to get a flat responce with the dsp and highpass a bit below tuning point and even a bit higher then tuning point to get rid of over excursion but it didn't work and i still was missing around 4 db. But i think the last graph showed exactly what is happening now. I gonna use the nd105 PR to get max spl.

    Leave a comment:


  • Scarface1
    replied
    Thanks guys, i gues i know what i have doing wrong. I sealed everything and run a 20 hz test tone and there where no leaks that i can hear. I did some spl measurements and get like 89db sometimes even with distortion so bad that it can't be the driver wich is maybe past xmax. I looked for the peerless 3.5 passive radiator parameters and i saw a Qms of 10 wich is very very high, while the PR is very very stiff. So i looked up the parameters on the forum and the Qms was 1.44 and i get different results. I don't know why i put a Qms of 10 but mostly i do this with cheap PR because they are not stiff at all.
    Attached Files

    Leave a comment:


  • billfitzmaurice
    replied
    You'd use the sim to diagnose the measured result. If you don't have the sim chart for comparison you have no way of knowing if the measured result is good or not.

    Leave a comment:


  • donradick
    replied
    Originally posted by billfitzmaurice View Post
    The effect on impedance of changing Ql is seen in WinISD. It has the most impact on the value of the upper peak, which goes down as Ql goes down.
    Of course, but we are talking about measuring the actual possible leakage on a box that's been built.

    Leave a comment:


  • billfitzmaurice
    replied
    The effect on impedance of changing Ql is seen in WinISD. It has the most impact on the value of the upper peak, which goes down as Ql goes down.

    Leave a comment:


  • donradick
    replied
    I'm surprised no one has mentioned the value of an impedance sweep. A PR box, since it's essentially a ported alignment, should have 2 sharp impedance spikes in the
    bass region, with the actual box Fb in the minimum between them. If you have significant leaks, the peaks will be way lower than normal, and they will be rounded,
    So first, make sure that the box is sealed up, go over all the seams on the inside with caulk.
    Then, as Tom Z has mentioned, use some weatherproofing spongy tape under the drivers and PRs.

    Leave a comment:


  • billfitzmaurice
    replied
    Originally posted by Scarface1 View Post

    Thanks, i will look this up tomorrow in winisd. If passive designs are not 100% airtight, does this mean that a PR is designed with a 'breathing' surround or something?
    It's not just a PR thing.You will see the same effect with a sealed MTM. Push in one cone, the other should extend. If there's minimal leakage the second cone will slowly retract. If there's zero leakage it won't retract. If there's a lot of leakage it won't extend.

    Leave a comment:


  • Scarface1
    replied
    Originally posted by tomzarbo View Post
    Larger speaker drivers seem to be easier to get airtight.

    Push in quickly on a typical 8" to 15" subwoofer with a PR and you should see the PR's slowly come back to resting position... it should take maybe between 4 to 8 seconds as an estimate to get back to normal resting position.

    Then let go of the driver... it should take another 4 to 8 seconds to reach equilibrium again. That is pretty much normal for PR enclosures I've constructed. You'd think they would be more airtight than that but that's been my experience so far, and I don't think there were really any 'leaks' in any of these so to speak... really just some 'leaky' materials that allow the changes in air pressure.

    Now, I've done a few projects with ND91 size drivers and smaller, and they can be a bit of a bugger to seal completely. It's common for those little ones to 'bounce back' to resting position a lot quicker than that... partly because there is less air compressed or rarefied, but also because they are often TOUGH to seal up perfectly... flanges a few MM thick and terminals that stick out farther than the frame of the driver are two pretty good reasons this happens. It's actually pretty critical to seal up the small ones as an air leak can just KILL any bass response on those. Just glump up the openings with some gasket caulk or black foam mounting tape and do the best you can.

    One way to find an actual leak though is to run some bass test tones though the speaker at varying frequencies, using a frequency generator app or something similar, then you should be able to hear the 'leak' a little better.

    Hope this helps.

    TomZ
    Thanks Tom, i will run a 20hz sinewave tru it. If i push the passives and holt it, the nd91 takes 1 or 2 seconds at max to gain back to normal position. So not really bad i gues??

    I glued everything and i'm 99% sure it isn't the problem, so i gonna make a second(stereo) enclosure and if i have the same reaction i will put some extra glue behind the tweeter and use some isolation strip for the nd91 to see if there is some difference.

    Leave a comment:


  • Scarface1
    replied
    Originally posted by billfitzmaurice View Post
    The OP is correct. The cone should move back to its neutral position, but very slowly. The leak with a PR will have the same effect as with a sealed or ported cab, it will upset the tuning. The source of the leak could be in a number of places, from the cabinet joints to the driver and PR gaskets. The usual method to find leaks is with a test tone too low in frequency to hear, allowing one to locate the leak via the whistling it creates.

    To see the effect of a leak on response in WinISD look at the Ql in the Advanced tab of the Box window. The default value is 10, which is an average reasonably tight enclosure. A value of 100 would be absolutely airtight, a value of 5 moderately tight, a value of 1 very leaky.
    Thanks, i will look this up tomorrow in winisd. If passive designs are not 100% airtight, does this mean that a PR is designed with a 'breathing' surround or something?

    Leave a comment:


  • Scarface1
    replied
    Originally posted by devnull View Post
    A speaker should never be 100% sealed. It needs to be able to equalize the internal cabinet pressure with the outside atmospheric pressure.

    How does your model compare to measured? Where is the speaker -6 dB down?
    I just measure how loud music can be played. In previous builds i run like 10 songs and pick the highest spl wich is archieved after that. I get pretty close to a save maximum spl with this way. It's not that accurate, especially with dsp added, so i have to look at this as well.

    Leave a comment:


  • Billet
    replied
    Think of the rate that the PR is moving. If it were converted to a frequency, it would be probably something like 1 Hz or less. Not really a problem for actual bass response.

    Leave a comment:


  • tomzarbo
    replied
    Larger speaker drivers seem to be easier to get airtight.

    Push in quickly on a typical 8" to 15" subwoofer with a PR and you should see the PR's slowly come back to resting position... it should take maybe between 4 to 8 seconds as an estimate to get back to normal resting position.

    Then let go of the driver... it should take another 4 to 8 seconds to reach equilibrium again. That is pretty much normal for PR enclosures I've constructed. You'd think they would be more airtight than that but that's been my experience so far, and I don't think there were really any 'leaks' in any of these so to speak... really just some 'leaky' materials that allow the changes in air pressure.

    Now, I've done a few projects with ND91 size drivers and smaller, and they can be a bit of a bugger to seal completely. It's common for those little ones to 'bounce back' to resting position a lot quicker than that... partly because there is less air compressed or rarefied, but also because they are often TOUGH to seal up perfectly... flanges a few MM thick and terminals that stick out farther than the frame of the driver are two pretty good reasons this happens. It's actually pretty critical to seal up the small ones as an air leak can just KILL any bass response on those. Just glump up the openings with some gasket caulk or black foam mounting tape and do the best you can.

    One way to find an actual leak though is to run some bass test tones though the speaker at varying frequencies, using a frequency generator app or something similar, then you should be able to hear the 'leak' a little better.

    Hope this helps.

    TomZ

    Leave a comment:


  • billfitzmaurice
    replied
    The OP is correct. The cone should move back to its neutral position, but very slowly. The leak with a PR will have the same effect as with a sealed or ported cab, it will upset the tuning. The source of the leak could be in a number of places, from the cabinet joints to the driver and PR gaskets. The usual method to find leaks is with a test tone too low in frequency to hear, allowing one to locate the leak via the whistling it creates.

    To see the effect of a leak on response in WinISD look at the Ql in the Advanced tab of the Box window. The default value is 10, which is an average reasonably tight enclosure. A value of 100 would be absolutely airtight, a value of 5 moderately tight, a value of 1 very leaky.
    Last edited by billfitzmaurice; 07-10-2021, 01:53 PM.

    Leave a comment:


  • fpitas
    replied
    What devnull said. Unless the leak is simply enormous, the main danger is whistling noises.

    Leave a comment:

Working...
X