Page 1 of 2 12 LastLast
Results 1 to 15 of 16

Thread: Ground Plane Measurement VS Anechoic

  1. #1
    Member
    Join Date
    Dec 2020
    Location
    Oregon
    Posts
    50

    Ground Plane Measurement VS Anechoic

    I was looking for a study showing the difference between ground plane measurement and (VS) anechoic measurement but I was not able to find anything. Obviously I know that the GP measurement will indicate higher SPL compared to the anechoic that will have lower sensitivity. I am referring to the shape of the response curve. How is that affected? I will also assume that the high frequencies will be affected less than (compared to) the low frequencies.

    Let me give an example of what I am hopping to find;
    For sake of discussion lets assume that we have a 3 cu ft tower speaker (in the upright position) that measured (ground plane) flat 20Hz to 20Khz. Then suspend the same speaker 50ft in the air, take another measurement and compare the two curves to get a picture of how the shape of the curve is affected. I am hoping to find a compatible type of study.

    Any thoughts / suggestions appreciated

  2. #2
    Member
    Join Date
    Aug 2019
    Location
    UK
    Posts
    36
    Quote Originally Posted by Steve C 2020 View Post
    I was looking for a study showing the difference between ground plane measurement and (VS) anechoic measurement but I was not able to find anything. Obviously I know that the GP measurement will indicate higher SPL compared to the anechoic that will have lower sensitivity. I am referring to the shape of the response curve. How is that affected? I will also assume that the high frequencies will be affected less than (compared to) the low frequencies.

    Let me give an example of what I am hopping to find;
    For sake of discussion lets assume that we have a 3 cu ft tower speaker (in the upright position) that measured (ground plane) flat 20Hz to 20Khz. Then suspend the same speaker 50ft in the air, take another measurement and compare the two curves to get a picture of how the shape of the curve is affected. I am hoping to find a compatible type of study.

    Any thoughts / suggestions appreciated
    Any frequency at which a transducer radiates in an omnidirectional pattern will be reinforced or reduced depending on the wavelength, the size or the enclosure and distance from the ground plane. Up in the air it's mostly the cabinet dimension that effect which frequencies reach the measuring mic and which partly radiate into free space. As enclosure dimensions are usually small compared to LF wavelengths the low frequency response will be less in the air, but most likely less ragged than measured on the ground. Ground plane measurement probably give a better idea of LF output in a room, but not necessarily how smooth the response will be.

  3. #3
    Member
    Join Date
    Dec 2020
    Location
    Oregon
    Posts
    50
    Quote Originally Posted by Epplerd View Post
    Any frequency at which a transducer radiates in an omnidirectional pattern will be reinforced or reduced depending on the wavelength, the size or the enclosure and distance from the ground plane. Up in the air it's mostly the cabinet dimension that effect which frequencies reach the measuring mic and which partly radiate into free space. As enclosure dimensions are usually small compared to LF wavelengths the low frequency response will be less in the air, but most likely less ragged than measured on the ground. Ground plane measurement probably give a better idea of LF output in a room, but not necessarily how smooth the response will be.
    Yes that is true. The size and shape of the front baffle of the speaker affects the frequency response of the speaker. I believe that is referred to as BD (baffle diffraction). Keep in mind that if in the speaker measures flat on a GP measurement, then the issues of BD have already been compensated for (for the most part) when the speaker is moved into an anechoic environment. For this reason, the GP measurement and anechoic measurement should both measure relatively flat.

  4. #4
    Administrator Robh3606's Avatar
    Join Date
    Apr 2003
    Location
    Rocinante
    Posts
    8,170
    "I could be arguing in my spare time"

  5. #5
    Senior Member RMC's Avatar
    Join Date
    Feb 2017
    Location
    Canada
    Posts
    1,626
    In his Loudspeaker Handbook 2nd ed, Eargle explains these, as well as a method using ground plane measurements to approximate anechoic measurements usually done in full space, including a response graph showing Ground plane, anechoic and half-space (P. 340-1).

    There's also the open field environment mentioned which is similar to ground plane. "It is often built on a roof top, where there are few space limitations." (...) "The open field is often referred to as a 2Pi, or half-space, measuring environment." (p. 335-6)

    I can't reproduce the interesting typical curves shown for the three. However, curve shape is the same for anechoic and ground plane, though there's a level difference between these two, which looks like 6 db or so.

    As for the ground plane vs half-space curves, from 100 hz going down, they're identical.

    Hopefully the above will help you.

    Richard

  6. #6
    Senior Member
    Join Date
    Apr 2003
    Location
    Australia
    Posts
    7,942
    I think this is what you are referring to.


    Half space is like a soffit monitor mounted flush with no air leakage around the enclosure to prevent cancellation of frequencies equal to wavelength the perimeter of the enclosure (as l recall?)

    Someone can think that though and debate the point so you all withdrawn from alzheimer's.

    Ultimately if your interest is in real low frequency output and bass extension then a bass reflex simulator will give you real world usable low frequency output below 100 hertz at various power levels.

    Most acoustic measurements are small signal or at 1 watt. A loudspeaker transducer can typically run out of Xmax or thermal power limit before real usable bass response can be realised.

    If you are looking at response above 100 hertz this is typically dominated by room mode characteristics that can be measured with REW.

    Large baffle loudspeaker designs such ac Jbl floor standing designs are not typically effected by baffle diffraction like your British narrow baffle floor standing systems that use multiple 6.5 inch woofers to equalise the baffle compensation effect.
    Attached Images Attached Images  

  7. #7
    Administrator Robh3606's Avatar
    Join Date
    Apr 2003
    Location
    Rocinante
    Posts
    8,170
    Take a look at post 32. Greg warns about issues using GP measurements. Not that you can't get good measurements you just have to be aware of what the differences are compared to other methods.

    Rob

    http://www.audioheritage.org/vbullet...id-I-get/page3
    Attached Images Attached Images
    "I could be arguing in my spare time"

  8. #8
    Senior Member RMC's Avatar
    Join Date
    Feb 2017
    Location
    Canada
    Posts
    1,626
    RE "I think this is what you are referring to."

    Graph shown somewhat similar to Eargle's (Data after D'Appolito). However, contrary to the graph shown, on Eargle's the top curve (ground plane) and the second curve (half-space) are EXACTLY the same from 100hz going down...

    Since it was mentioned before to search Mark Gander, well Eargle indicates "See Gander (1982) and D'Appolito (1998) for added details of this measurement method." (p. 340) The latter's work on this being more recent, maybe explains the small differences.

  9. #9
    Senior Member
    Join Date
    Apr 2003
    Location
    Australia
    Posts
    7,942
    Another approach is to put the enclosure face up on the ground and place a temporary mdf baffle extension or apron around the baffle of at least one metre.

    The thing is of course a domestic room setting is going to make any of these idealised measurements superfluous. Appropriate room equalisation with Dirac would resolve most of the issues.

    Edit

    The other thing often overlooked because it’s difficult to measure is the influence of room boundary proximity below 150 hertz. It’s a case by case thing. Then you have Room modes that usually swamp native response variations in relatively small domestic listening rooms. In my family room there was a broad hump from 150 hertz down to 40 hertz. The Dirac cleaned up the midrange after that was removed.

  10. #10
    Senior Member RMC's Avatar
    Join Date
    Feb 2017
    Location
    Canada
    Posts
    1,626
    Like the 1rst para. idea, resemblance to the roof top thing, but being on the ground.

    As for "difficult to measure is the influence of room boundary proximity below 150 hertz"

    "Roy Allison (1974) (RMC: Allison Acoustics speakers) is one of only a handful of loudspeaker designer-manufacturers who have tackled this problem headlong in making specific loudspeaker placement recommendations for the consumer." (Eargle p. 368)

    Unfortunately Allison's room boundary paper explaning all this isn't simple & available to all, plus not free (1974 JAES).

    However, in Listening room boundary conditions section, Eargle gives a summary of this, as well as a graph showing LF acoustical output for three different boundary situations. So a peek of what one may expect. Still not as simple as we would expect though...

    Might be easier to get, and understand, an old Allison Acoustic speaker Owner Manual explaning this! LOL

    As lucky as he is with finds, maybe Seawolf would dig out one to post here...

    Out for now.

  11. #11
    Senior Member
    Join Date
    Apr 2003
    Location
    Australia
    Posts
    7,942
    There are some simulation programs using Excel to assess the low end using XYZ locations, room materials and so on. In practice it’s simply a matter on using a Dirac equipped preamp or amplifier and taking 5-7 measurements in different locations. Dirac then computes an error correction curve that is then uploaded to the component with Dirac. This kind of convolution process was very complicated and costly until Dirac was launched. A number of custom curves can be saved. You can also edit a curve on the fly. Dirac has minimal effect on the audio quality. If you have problematic bass then l would recommend a demonstration at a dealers showroom. It’s a lot more straightforward than screwing around with non WAF compliant changes to your room.

    If you have a square room with a solid concrete structure don’t expect miracles as the reverberation time at low frequencies will cause the room to ring. Dirac will notch out those resonant high Q modal frequencies but you might loose some of the recorded frequencies.

    This is all on the www and supersedes at lot of the early theory, trial and error work by researchers back in the day. There’s nothing precious about this topic with the technology now available to the consumer in 2021.

  12. #12
    Member
    Join Date
    Dec 2020
    Location
    Oregon
    Posts
    50
    Quote Originally Posted by RMC View Post
    In his Loudspeaker Handbook 2nd ed, Eargle explains these, as well as a method using ground plane measurements to approximate anechoic measurements usually done in full space, including a response graph showing Ground plane, anechoic and half-space (P. 340-1). There's also the open field environment mentioned which is similar to ground plane. "It is often built on a roof top, where there are few space limitations." (...) "The open field is often referred to as a 2Pi, or half-space, measuring environment." (p. 335-6) I can't reproduce the interesting typical curves shown for the three. However, curve shape is the same for anechoic and ground plane, though there's a level difference between these two, which looks like 6 db or so. As for the ground plane vs half-space curves, from 100 hz going down, they're identical. Hopefully the above will help you. Richard
    I apologize for the late reply and thanks to all for this information. I don't have a copy of Eargle's loudspeaker handbook. Do you have an easy way to share that graph as an attachment? Thanks !!!

  13. #13
    Member
    Join Date
    Aug 2019
    Location
    UK
    Posts
    36
    Quote Originally Posted by Steve C 2020 View Post
    I apologize for the late reply and thanks to all for this information. I don't have a copy of Eargle's loudspeaker handbook. Do you have an easy way to share that graph as an attachment? Thanks !!!
    Room resonances exist in two domains, frequency and time. Accoustic treatment will reduce the room resonances in both domains. Depending on the severity of the problem, sometimes simple things like the furnishings can assist in combatting room resonances. At the low frequency end its much more difficult. The sound level can vary dramatically at different locations within the room. And the more sound pumped into the room the more the resonances will be excited.

  14. #14
    Member
    Join Date
    Dec 2020
    Location
    Oregon
    Posts
    50
    Quote Originally Posted by Epplerd View Post
    Room resonances exist in two domains, frequency and time. Accoustic treatment will reduce the room resonances in both domains. Depending on the severity of the problem, sometimes simple things like the furnishings can assist in combatting room resonances. At the low frequency end its much more difficult. The sound level can vary dramatically at different locations within the room. And the more sound pumped into the room the more the resonances will be excited.
    Thanks for that information. My ground plane measurements are typically conducted outdoors not in close proximity to objects.

  15. #15
    Member
    Join Date
    Dec 2020
    Location
    Oregon
    Posts
    50
    Quote Originally Posted by Ian Mackenzie View Post
    I think this is what you are referring to. Half space is like a soffit monitor mounted flush with no air leakage around the enclosure to prevent cancellation of frequencies equal to wavelength the perimeter of the enclosure (as l recall?) Someone can think that though and debate the point so you all withdrawn from alzheimer's. Ultimately if your interest is in real low frequency output and bass extension then a bass reflex simulator will give you real world usable low frequency output below 100 hertz at various power levels. Most acoustic measurements are small signal or at 1 watt. A loudspeaker transducer can typically run out of Xmax or thermal power limit before real usable bass response can be realised. If you are looking at response above 100 hertz this is typically dominated by room mode characteristics that can be measured with REW. Large baffle loudspeaker designs such ac Jbl floor standing designs are not typically effected by baffle diffraction like your British narrow baffle floor standing systems that use multiple 6.5 inch woofers to equalise the baffle compensation effect.
    Yes correct. Thank you !!! Is that graph 5 or 10db per division? It is a little hard to read

Thread Information

Users Browsing this Thread

There are currently 1 users browsing this thread. (0 members and 1 guests)

Similar Threads

  1. JBL 2226H - Outside Ground Plane Measurements
    By dkalsi in forum Lansing Product DIY Forum
    Replies: 4
    Last Post: 04-07-2023, 04:04 PM
  2. Apple Anechoic Chamber
    By Titanium Dome in forum General Audio Discussion
    Replies: 6
    Last Post: 02-08-2018, 10:13 AM
  3. Who built 2 anechoic chambers in Los Angeles recently?
    By 10 Watt Street in forum General Audio Discussion
    Replies: 19
    Last Post: 06-29-2016, 09:57 AM
  4. Plane Wave Tube Construction
    By 1audiohack in forum General Audio Discussion
    Replies: 41
    Last Post: 11-17-2011, 01:43 PM
  5. LSR in anechoic Chamber
    By Valentin in forum Lansing Product DIY Forum
    Replies: 7
    Last Post: 11-08-2007, 08:51 AM

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •