The No.1 Website for Pro Audio
 Search This Thread  Search This Forum  Search Reviews  Search Gear Database  Search Gear for sale  Search Gearslutz Go Advanced
My Experiment with a Metal Panel Absorber Condenser Microphones
Old 31st December 2011
  #271
Lives for gear
 
G. E.'s Avatar
 

Enhanced Porous Absober Scheme

Quote:
Originally Posted by boggy View Post
... G. E. made a measurements of this construction:
Perforated Panel with Porous Absorber trap

... as something that may be called "before and after burst decay analysis":
Perforated Panel with Porous Absorber trap

Absorber is built with IsoBond and steel sheet (G. E. may correct me if I'm wrong), and mounted in corner(!), then he improve response in room as you can see in measurement results. Absorber isn't mounted glued to the heavy back wall, it is free standing (G. E.?), and it is even supported (at side of steel plate?) with couple of batons. ...
The construction of this enhanced porous absorber (true broadband absorber) is based on the following scheme:
Attached Thumbnails
My Experiment with a Metal Panel Absorber-enhanced-porous-absorber-scheme.gif  
Old 31st December 2011
  #272
Lives for gear
 
boggy's Avatar
Quote:
Originally Posted by G. E. View Post
The construction of this enhanced porous absorber (true broadband absorber) is based on the following scheme:
Thanks G. E., this is very informative!
It is probably interesting to measure this particular absorber's influence when it is placed on the floor and when it is hung (freely) from the ceiling.

And this is resonant absorber... if you are unsure, try the similar construction (double IsoBond placed at an angle in the corner), but without steel plate inside and you must see the difference in your tone burst response.
Old 2nd January 2012
  #273
Lives for gear
 
G. E.'s Avatar
 

Variations

Quote:
Originally Posted by boggy View Post
... It is probably interesting to measure this particular absorber's influence when it is placed on the floor and when it is hung (freely) from the ceiling. ...
It can't be hung right now because of the "glue-less" construction but it could stand "freely" on the floor away from the corner.

Quote:
Originally Posted by boggy View Post
... And this is resonant absorber... if you are unsure, try the similar construction (double IsoBond placed at an angle in the corner), but without steel plate inside and you must see the difference in your tone burst response.
I still prefer to call it a porous absorber (enhanced by a resonant plate).

The difference between a typical and an enhanced porous absorber of the same size looks basically like this (burstdecay diagram):
Attached Thumbnails
My Experiment with a Metal Panel Absorber-typical.gif   My Experiment with a Metal Panel Absorber-enhanced.gif  
Old 2nd January 2012
  #274
Lives for gear
 
scoring4films's Avatar
 

Question

Quote:
Originally Posted by G. E. View Post
...
The difference between a typical and an enhanced porous absorber of the same size looks basically like this (burstdecay diagram):

Wow. Totally nuked that room mode at ~44Hz. Was the plate 1mm or 2.5mm? Were the absorbers Basotect or Caruso Isobond?
Old 2nd January 2012
  #275
Lives for gear
 
Schaap's Avatar
 

Quote:
Originally Posted by G. E. View Post
The difference between a typical and an enhanced porous absorber of the same size looks basically like this (burstdecay diagram):
Very impressive G. E. . It tamed the difficult <70 Hz area almost completely.
Old 2nd January 2012
  #276
Lives for gear
 
boggy's Avatar
Quote:
Originally Posted by G. E. View Post
It can't be hung right now because of the "glue-less" construction but it could stand "freely" on the floor away from the corner.
Can you describe your "glue-less" construction? IsoBond isn't glued to stell plate?
Quote:
Originally Posted by G. E. View Post


I still prefer to call it a porous absorber (enhanced by a resonant plate).
OK.
Quote:
Originally Posted by G. E. View Post
The difference between a typical and an enhanced porous absorber of the same size looks basically like this (burstdecay diagram):
Yes, difference is more than noticeable...
Old 2nd January 2012
  #277
Lives for gear
 
G. E.'s Avatar
 

Sizes

Quote:
Originally Posted by scoring4films View Post
... Was the plate 1mm or 2.5mm? Were the absorbers Basotect or Caruso Isobond?
I used CIB for sure but I can't remember the exact plate thickness (I have 6 different plates ranging from 0,5mm [~0.02"] up to 2,5mm [~0.1"]).
Old 2nd January 2012
  #278
Lives for gear
 
G. E.'s Avatar
 

Glueing

Quote:
Originally Posted by boggy View Post
Can you describe your "glue-less" construction? IsoBond isn't glued to stell plate? ...
No glue at all, simply one part leaning against the other.
Old 2nd January 2012
  #279
Lives for gear
 
andrebrito's Avatar
 

Having a rigid backing changes the impedance of the system so you get different results with and without rigid backing
Old 4th January 2012
  #280
Lives for gear
 
Dange's Avatar
 

Quote:
Originally Posted by Nuuk View Post
There has been a quote from Fuchs somewhere, which basically said that, if you remove the rigid backing, you create a shortcut between the back and the front of the VPR, resulting in lower efficiency.
I'm thinking this might similar to the reason a loudspeaker works better in a baffle than in free air. In a baffle, the front and rear radiated pressures are separated and cannot interfere and cancel with each other.

With a VPR panel in free air (completely away from any boundaries) there is a 'shortcut' front to back, as if a loudspeaker was without a baffle. When it's on a rigid backing (or close to a boundary) this shortcut may not occur due to the proximity of the rigid boundary, the size of the wavelengths being absorbed and porous absorber.
Old 4th January 2012
  #281
Lives for gear
 
Dange's Avatar
 

Quote:
Originally Posted by Victor_Stoian View Post
i think the issue here is if the back of the porous absorber has to be tied (glued) on the rigid backing.
Ah I see. From what I've read on how the VPR works, then I don't think the porous absorber needs to be glued to a rigid backing either. I believe it still needs to be close to a boundary though for the best result
Old 4th January 2012
  #282
Lives for gear
 
DrFrankencopter's Avatar
Quote:
Originally Posted by Dange View Post
Ah I see. From what I've read on how the VPR works, then I don't think the porous absorber needs to be glued to a rigid backing either. I believe it still needs to be close to a boundary though for the best result
I think it needs to be more than close to a boundary. I suspect it needs contact with the boundary (though should not need to be rigidly fixed to it). The absorber is acting like a spring and damper...if it doesn't touch the boundary, then the spring constant is very different (requiring the air between the absorber and the boundary to serve as part of a spring in series with that of the absorber itself).

Cheers

Kris
Old 4th January 2012
  #283
Lives for gear
 
Dange's Avatar
 

Quote:
Originally Posted by DrFrankencopter View Post
I think it needs to be more than close to a boundary. I suspect it needs contact with the boundary (though should not need to be rigidly fixed to it). The absorber is acting like a spring and damper...if it doesn't touch the boundary, then the spring constant is very different (requiring the air between the absorber and the boundary to serve as part of a spring in series with that of the absorber itself).
Not necessarily but I see your point. The mass spring action only accounts for absorption at one resonant frequency. Absorption at other frequencies is due to plate modal deformation, this is damped by the adhesive and foam constraining the plate*


*Edit: a bit like Green Glue constraining one layer with another in a wall construction
Old 5th January 2012
  #284
Gear maniac
 
aackthpt's Avatar
 

Quote:
Originally Posted by Dange View Post
I'm thinking this might similar to the reason a loudspeaker works better in a baffle than in free air. In a baffle, the front and rear radiated pressures are separated and cannot interfere and cancel with each other.

With a VPR panel in free air (completely away from any boundaries) there is a 'shortcut' front to back, as if a loudspeaker was without a baffle. When it's on a rigid backing (or close to a boundary) this shortcut may not occur due to the proximity of the rigid boundary, the size of the wavelengths being absorbed and porous absorber.
I believe the loudspeaker putting out nearly nothing at LF if in free air is due to its size relative to wavelength. Consider a speaker with a cone of infinite size with the same motion, wouldn't that put out some SPL? Whereas clearly a small one would not. The larger the diaphragm the less "spillover" or "short circuiting" due to lower ratio of edge length to area. So there is a continuum of maximum short circuiting with a small "diaphragm" part to no short circuiting with an infinite diaphragm. I think this is the reason for the minimum size of ~1x1.5m as stated in the thread by Hannes here as well as from the Fraunhofer developers.

My expectation is that putting the porous absorber / damping material against a rigid boundary may provide extra benefit but isn't necessary for the device to operate as intended in at least some aspect. I'm not sure if the relative strength of the effects is predictable. It looks like you (Dange) already concluded much of this but I thought a different line of reasoning might be of interest.

Test results of a device with the porous against the wall and at various spacing off the wall would be most enlightening.

Consider the data we have so far. G.E.'s device away from the wall worked very well. Audiothings' device appears to work slightly less well in the "diagonal" configuration than in the "against wall" configuration, but most of the device's effect is gained in the "diagonal" configuration. Considering that it probably works in two directions in "diagonal" rather than mostly one in "against wall", it could be seen as a choice to be made in how best to deploy the device in a specific situation.

Last edited by aackthpt; 5th January 2012 at 05:35 AM.. Reason: Clarify final paragraph.
Old 5th January 2012
  #285
Lives for gear
 
G. E.'s Avatar
 

Touching Walls

Quote:
Originally Posted by aackthpt View Post
... G.E.'s device away from the wall worked very well. ...
The back straddled the vertical corner with both back edges still touching walls as shown in the "Enhanced Porous Absorber Scheme".
Old 6th January 2012
  #286
Gear maniac
 
aackthpt's Avatar
 

Quote:
Originally Posted by G. E. View Post
The back straddled the vertical corner with both back edges still touching walls as shown in the "Enhanced Porous Absorber Scheme".
Understood. I was considering this to be "not on the wall" since the back of the device was not against the wall. But you're right, since that mount may limit the "short circuit" to only the top, it might be considered a different class of mounting than simply "off the wall" while parallel to it.
Old 6th January 2012
  #287
Gear interested
 

all about the VPR

hey guys, id like to chime in on the thread as i am currently myself trying to build two VPR absorbers.

i tried to read through the whole but thats quite a mission so please dont get mad if i am repeating some things other members already stated.


i am reading the fuchs book and i try to clear out some things that people are disscussing,

the reason why thin metal sheets are used is simply because of the density, the higher the mass of the plate the lower the resonance,

if you wanted to have a wooden plate with a high mass to make it work in under 100hz it would have to be alot thicker than the thin metal sheet which in turn lowers its ability to resonate and vibrate.

fuch calls for a material with a low inner friction (that being steel) and an absorber with a high inner friction.

comparing basotect with other foams based on their density is not the whole story, its complicated to explain a bit,

they key is a foam which takes the vibration from the plate because its glued to the whole surface,(acting as the spring) without in first place killing the vibration on the surface of the plate! where the foam meets the metal the foam has to be rigid enough so it can take the vibration from the plate, at the same time not be too dense so it would just dampen the plate.

that is different from what u would normally call to damp a plate. just damping would mean to put heavy foam on it to stop it from vibrating.

i hope somebody here understands what i mean lol,
sorry i am not a native speaker.


in the book they never put a VPR to a corner but always on-to a wall. because the closer to the wall the higher is the sound pressure which creates resonance in the plate. if you put it across a corner you loose a fair bit of performance, at that distance from the wall the ideal absorber is foam with a high flow resistance to slow down the soundwaves and reduce their engergy, close to the wall the pressure rises as the speed drops which calls for a membran absorber.

rigidly mounting the vpr to a wall makes for a proper spring/mass system, without that the foam would move with the plate to some xtent, thus lowering performance.

the reason for leaving the surround open is because the VPR are by definition broadband absorbers, the open foam behind the plate kills the mids and highs that enter it.

its late and iam tired but i think acoustic short circouting should not matter alot in that design. the very large waves would have to bend around the membrane within a space of a few inches or cm and remain at the same phase required to dampen the resonance, alltogether unlikely also because of the absorber foam on the back of the plate slowing down the wave etc etc..

certain mid-high frequencies (around 800hz as shown in the book from helmut fuchs) diffract alot around edges, thus entering the foam.

i know alot of those things have already been correctly pointed out (and more fluently explained for sure), the thing that interests me most for now is, how big the difference really is between just putting a thick panel on the back of the foam and then just leaning the VPR against a wall for example, compared to rigidly fixing the back panel of the foam (or the foam itself) to the wall with screws/glue.

so many things to consider.. and concentration is long gone..

lets solve that puzzle together and kill those evel room modes!! looking forward to mount them in my music room YAY

btw: i try to use Plexiglas aka perspex myself which is not as ideal as metal, at least its heavier than wood but... too late now...
Old 6th January 2012
  #288
Lives for gear
 
scoring4films's Avatar
 

Quote:
Originally Posted by sebourbon View Post
...
comparing basotect with other foams based on their density is not the whole story, its complicated to explain a bit,

they key is a foam which takes the vibration from the plate because its glued to the whole surface,(acting as the spring) without in first place killing the vibration on the surface of the plate! where the foam meets the metal the foam has to be rigid enough so it can take the vibration from the plate, at the same time not be too dense so it would just dampen the plate.
thanks sebourbon. your post was very helpful for me (and I had no problem understanding your english at all!).

I'm very eager to build a couple of VPRs too, but I'm still stuck on the the foam component.

Caruso-ISO-Bond seems to perform the best (and I'd prefer to use it because it's so stable and doesn't off-gas), but I can't find a supplier here in Canada. I've tried to contact the manfacturer in Germany several times, and I even asked one of my German friends to call them on my behalf, but still no response. The international shipping and duties might end up being too costly anyway.

I did find a source for basotect (here in Canada), but it's expensive. The supplier wants ~US$200 incl. shipping for one 150cmX100cmX10cm piece.

I wish there was some way to figure out if there are any viable foam alternatives.

Trial and error with furniture/mattress foams doesn't seem like the best option.
Old 7th January 2012
  #289
Lives for gear
 
Bjorn Omholt's Avatar
 

Here's my VPR that I "built". Good luck guys! heh
Attached Thumbnails
My Experiment with a Metal Panel Absorber-006-custom-.jpg  
Old 7th January 2012
  #290
Lives for gear
 
Hannes_F's Avatar
 

Quote:
Originally Posted by sebourbon View Post
they key is a foam which takes the vibration from the plate because its glued to the whole surface,(acting as the spring) without in first place killing the vibration on the surface of the plate! where the foam meets the metal the foam has to be rigid enough so it can take the vibration from the plate, at the same time not be too dense so it would just dampen the plate.
The key is taking energy out of the system. The optimum would be what is called critical dampening.

In a conventional porous absorber the energy is used up by air basically moving back and forth through the pores. In the VPR the air is also moving parallel to the steel plate, it is pumped from the dents in the steel to the hills and vice versa. That makes the energy consumption process much more effective.

That is one part of the story. The other part is that IF the foam has viscoelastic capabilities (which I don't know but it might be the case) this eats up additional energy. Viscoelastic means that it sort of flows like a liquid, with low resetting tendency.
Old 7th January 2012
  #291
Gear Nut
 
SRDigital's Avatar
 

Quote:
Originally Posted by Hannes_F View Post
The key is taking energy out of the system. The optimum would be what is called critical dampening.

In a conventional porous absorber the energy is used up by air basically moving back and forth through the pores. In the VPR the air is also moving parallel to the steel plate, it is pumped from the dents in the steel to the hills and vice versa. That makes the energy consumption process much more effective.

That is one part of the story. The other part is that IF the foam has viscoelastic capabilities (which I don't know but it might be the case) this eats up additional energy. Viscoelastic means that it sort of flows like a liquid, with low resetting tendency.
so what about a memory foam top for a bed? i see they come in 25kg/m3, similar to OP's build.
Old 7th January 2012
  #292
Lives for gear
 
Hannes_F's Avatar
 

Quote:
Originally Posted by SRDigital View Post
so what about a memory foam top for a bed? i see they come in 25kg/m3, similar to OP's build.
I might be mistaken here but I think viscoelastic foam is quite the opposite of memory foam, no? As memory foam always wants to go back to the original state but viscoelastic materials streams into a new form (using up energy) and basically stay there. So when the backswing of the wave comes the viscoelastic material will eat up energy again instead of re-pumping it into the system like any elastic material would do.

But maybe I am not understanding memory foam well enough.
Old 7th January 2012
  #293
Gear Nut
 
SRDigital's Avatar
 

Quote:
Originally Posted by Hannes_F View Post
I might be mistaken here but I think viscoelastic foam is quite the opposite of memory foam, no? As memory foam always wants to go back to the original state but viscoelastic materials streams into a new form (using up energy) and basically stay there. So when the backswing of the wave comes the viscoelastic material will eat up energy again instead of re-pumping it into the system like any elastic material would do.

But maybe I am not understanding memory foam well enough.

some memory foams are slated as 'viso-elastic' is why i made the thought.
Old 7th January 2012
  #294
Gear Nut
 

Quote:
Originally Posted by sebourbon View Post
i know alot of those things have already been correctly pointed out (and more fluently explained for sure), the thing that interests me most for now is, how big the difference really is between just putting a thick panel on the back of the foam and then just leaning the VPR against a wall for example, compared to rigidly fixing the back panel of the foam (or the foam itself) to the wall with screws/glue.
I would guess that wouldn't make much of a difference.

Quote:
Originally Posted by sebourbon View Post
btw: i try to use Plexiglas aka perspex myself which is not as ideal as metal, at least its heavier than wood but... too late now...
Plexiglass will have a totally different effect. The reason the VPR works over such a broad range is not only the mass-spring-resonance of the steel plate and the foam (that would be in the range of a few hertz) but the self resonances of the steel plate (that work like in a room with multiples of the plate dimensions). Those resonances are much lower than in a room with the same dimensions because of the much higher speed of sound in steel (5920 m/s) than in air (343 m/s). So speaking easily the steel plate resonates at all it's own modes that are comparaple to a (2D)-"room" with 17 times the dimensions of the steel plate. So the lowest modes are still far below the frequency of interest, but from about 40 Hz there is an even spread of the harmonics of the modes, that make the VPR a broadband bass absorber. (Similarities to the plate reverbs have already been pointed out).

In Plexiglas the speed of sound is 2670 m/s, less than half of the speed in steel. So to get into the same ranges considering the modes alone I think you might need to more than double the dimensions, which makes for 4 times the material. So you won't really save much space or money, if it works at all. But as I said I don't think it would work because the Plexiglass won't bend internally as much and allow to build up the high order modes that are needed.
Old 7th January 2012
  #295
Lives for gear
 
audiothings's Avatar
 

Thread Starter
T1MON - thank you for that post. Since you seem to know about the speed of sound in various materials, could you tell us what the speed of sound is in basotect and isobond? Or even low density mineral wool, which can potentially be an alternative....
Old 7th January 2012
  #296
Lives for gear
 
scoring4films's Avatar
 

...and while we're revisiting the foam issue...
hey Audiothings, what were the specs on the foam you used on the VPR build you posted at the beginning of this thread?

Esp. interested in whether it was open or closed cell. Thanks!
Old 7th January 2012
  #297
Lives for gear
 
G. E.'s Avatar
 

Mixing It Up

Quote:
Originally Posted by T1M0N View Post
... The reason the VPR works over such a broad range is not only the mass-spring-resonance of the steel plate and the foam (that would be in the range of a few hertz) ...
I don't agree, the single mass-spring-resonance is typically in the 40Hz-80Hz range.

Quote:
Originally Posted by T1M0N View Post
... the self resonances of the steel plate (that work like in a room with multiples of the plate dimensions). Those resonances are much lower than in a room with the same dimensions because of the much higher speed of sound in steel (5920 m/s) than in air (343 m/s). So speaking easily the steel plate resonates at all it's own modes that are comparaple to a (2D)-"room" with 17 times the dimensions of the steel plate. So the lowest modes are still far below the frequency of interest, but from about 40 Hz there is an even spread of the harmonics of the modes, that make the VPR a broadband bass absorber. (Similarities to the plate reverbs have already been pointed out).
I don't agree, given the formula c_0 {speed of sound} = f {frequency} x lambda {wavelenght} it followes that the (resonant) frequency of a "room" is rising not falling in a (probably gasiform) medium with a higher c_0. Although sound does travel within a steel plate those modes are not what a VPR is about (they're starting way above a few kHz in a typical VPR), its about the transversal bending modes of the plate. Same goes for plate reverbs.
Old 7th January 2012
  #298
Gear interested
 

regarding the foam issues, you have to look for open cell foams because it needs to let air through it, it should be rigid in a way that you can put some strength on it (not too soft), it should be compressable to under 40% of its original shape without too much pressure, but go back to its initial form immediately. (40% compression rate is one of many standards there are in describing foam, but not all the suplliers know them or will give the numbers away)

i think what most people think of as memory foam in regards of beds, or visco elastic foam, is foam which can take alot of pressure (the weight of a human body over a long time) and still go back to its shape after "some" time, the foam we are looking for has to go back immediately like a spring.

anything in the range of basotect should work somewhat ok, for basotect they give a density of about 9kg/m², air flow resistivity between 8000rayls-20000 (depending on where in the huge blocks they make your piece has been cut out, in the lower parts the bubbles are smaller, hence higher flow resistivity.. -at least thats how i was told, maybe someone can confirm this?)


for plexi vs steel, yes i know steel would be better but i still think that its going to work ok with the perspex, i have two big sheets (about 1m by 1.5m and a bit) with 2mm for mid and 4mm for the deep frequencies,

while speed of sound might be lower by half than in steel, i think what also has to be considered is the much lower bending stiffness in acryl glass, which is higher in steel by a factor of some-ten times i think than in plexi or acryl glass, if the lowest eigenfreq of steel for a 2mm is 6.6hz, it should still be in a usable range for the plexi panel.

i make instrumental music amplified over instrumental speakers, the efficiency of most guitar speakers drops ALOT under 80hz (even under 100hz for some), and i am not a big bass enthusiast anyway, also not in listening music , i use subwoofers mostly for effects in movies.., for good hifi listening i enjoy big old celestion speakers that should go down to about 80hz also (enough for me), the lowest room eigenmode in my case is about 30hz (calculated, i have yet to verify it when i am ready with actually measuring that with a mic...) the next mode is about 70hz and then some around 90...and so on..

so my targets start at 70hz roughly, and i think (and yes i hope ) its gonna work.


i dont argue about the math behind the eigenmodes, its just so much hassle, i look at the formulas that fuchs describes and also as he states more important the measured effect of certain materials etc.. and thats very easily visible and impressive in the book.

eigenmode frequencies are so very difficult to calculate in a "free standing object" like a VPR (membran "swimming" on absorber) and its not just frequencies with a certain direct relationship to the lengths in a room, that would maybe be tru the 1st eigenmode for every dimension... any other wave might start an eigenmode in the panel -no mine are not perfectly rectangular, considering reflected and thus multiplicating or reducing each other waves within the boundaries of the material)

i have to agree with you t1mon on the eigenmodes beeing important overall in relationship to the room, but there is just so much more to it than whats obvious. and yes G.E. the typical spring mass system we see in panel absorbers is higher than in the few hertz, but i think t1mon meant the resonance in a "free standing" steel plate with foam (VPR) which still is a spring mass system, and in that case the lowest resonance is much lower than the same plate used in a airtight closed panel absorber (with reasonable dimensions)

i am still learning about it all too, lets not get in fight over very theoretical issues, as the whole thing is so complex it cannot be calculated really anyway,

trying to finish building installing and measuring with room eq, and then show results, might still take a few weeks tho

very interested in what acoustic impressions (also measured if possible) you all guys here who are already in the testing range have..!!!

chris
Old 8th January 2012
  #299
Lives for gear
 
audiothings's Avatar
 

Thread Starter
scoring4films, i used 23 kg/m3 open cell PU foam.
Old 8th January 2012
  #300
Gear Nut
 

Quote:
Originally Posted by G. E. View Post
I don't agree, the single mass-spring-resonance is typically in the 40Hz-80Hz range.



I don't agree, given the formula c_0 {speed of sound} = f {frequency} x lambda {wavelenght} it followes that the (resonant) frequency of a "room" is rising not falling in a (probably gasiform) medium with a higher c_0. Although sound does travel within a steel plate those modes are not what a VPR is about (they're starting way above a few kHz in a typical VPR), its about the transversal bending modes of the plate. Same goes for plate reverbs.
Listen to G.E. He definitely knows better than I do. Sorry for errors on my side.
New Reply Submit Thread to Facebook Facebook  Submit Thread to Twitter Twitter  Submit Thread to LinkedIn LinkedIn  Submit Thread to Google+ Google+  Submit Thread to Reddit Reddit 
 
Post Reply

Welcome to the Gearslutz Pro Audio Community!

Registration benefits include:
  • The ability to reply to and create new discussions
  • Access to members-only giveaways & competitions
  • Interact with VIP industry experts in our guest Q&As
  • Access to members-only sub forum discussions
  • Access to members-only Chat Room
  • Get INSTANT ACCESS to the world's best private pro audio Classifieds for only USD $20/year
  • Promote your eBay auctions and Reverb.com listings for free
  • Remove this message!
You need an account to post a reply. Create a username and password below and an account will be created and your post entered.


 
 
Slide to join now Processing…
Thread Tools
Search this Thread
Search this Thread:

Advanced Search
Forum Jump
Forum Jump