Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: and???
The concept of excitation of a resonance is complex - as usual.
No resonance occurs in issolation it has a response from DC to light, but it peaks at its resonance frequency. And no excitation is pure either, it has to start sometime and it has to end sometime and the startup and stoping are broadband excitations.
Hence any real world resonance will respond to any real world excitation. How much is the issue.
Things get really tough when there are multiple resonances. Consider a system with lets say two resonances and that I excite them in between the two. After the transients have died down the response is purely at the driven frequency, but when I turn off the excitation, the decay does not even contain the driven frequency. It decays with two tones at the resonances with the original source driven frequency completely missing.
This later facts is why it is thought that bass can sometimes not sound right in a undamped room. Because the bass notes will be detuned as they decay and this will not sound consonant.
These are only SIMPLE examples. Plates, like in loudspeaker cabinets, are dispersive - meaning that the sound speed is frequency dependent adding even more complications to the problem.
ShinOBIWAN said:
The concept of excitation of a resonance is complex - as usual.
No resonance occurs in issolation it has a response from DC to light, but it peaks at its resonance frequency. And no excitation is pure either, it has to start sometime and it has to end sometime and the startup and stoping are broadband excitations.
Hence any real world resonance will respond to any real world excitation. How much is the issue.
Things get really tough when there are multiple resonances. Consider a system with lets say two resonances and that I excite them in between the two. After the transients have died down the response is purely at the driven frequency, but when I turn off the excitation, the decay does not even contain the driven frequency. It decays with two tones at the resonances with the original source driven frequency completely missing.
This later facts is why it is thought that bass can sometimes not sound right in a undamped room. Because the bass notes will be detuned as they decay and this will not sound consonant.
These are only SIMPLE examples. Plates, like in loudspeaker cabinets, are dispersive - meaning that the sound speed is frequency dependent adding even more complications to the problem.
GM said:
These sources always get everything right? They don't have their own misconceptions to spread?
planet10 said:
Dave, I'm not getting that. I wrote the first quote, why would I think it's untrue?
Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: and???
Thanks for the elaboration Earl.
So my question is how audible are these? I'm guessing, for the situation of a typical loudspeaker cabinet, this distortion mechanism is rather less of a worry than the direct 1:1 coinciding of panel resonance and stimulus wavelength.
By the sounds of it the broadband rise and fall you mentioned will on average be of much lower energy the main steady state signal. Or because music content is anything but steady does this mean otherwise?
gedlee said:
Thanks for the elaboration Earl.
So my question is how audible are these? I'm guessing, for the situation of a typical loudspeaker cabinet, this distortion mechanism is rather less of a worry than the direct 1:1 coinciding of panel resonance and stimulus wavelength.
By the sounds of it the broadband rise and fall you mentioned will on average be of much lower energy the main steady state signal. Or because music content is anything but steady does this mean otherwise?
MJL21193 said:
I think it was meant as an answer to anyone who generally doubted what you stated.
Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: and???
How much relating to the Q of the resonance?
dave
gedlee said:
How much relating to the Q of the resonance?
dave
planet10 said:
A fairly decent cook I am. BBQ ribs are especially good.
My Lebanese chicken wings are spectacular. 😎
Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: and???
Audibilty has never been thouroughly tested. My bet is on the direct excitation of the panels by the loudspeaker as opposed to the coupling through the air. Air coupling falls off as 1/f, but direct coupling of the vibrations is almost constant. The direct vibration is best combated by damping the front panel after which the vibrations of the box will not be very significant at all. A resonant baffle board is thus the worst case scenario.
I always well damp my baffles and the waveguides, but I only use a piece of oak as an internal cross brace. I buy oak flooring for $50 a bundle and I get enough Oak to last for 100's of cabinets. My previous enclosures were composite plastic - very well damped, good paint surface, looked great, very expensive. I became convinced that box resonances were not a big deal and so I am scraping these expensive cabinets and going to MDF (or anything else the customer is willing to pay for, I don't really care).
Music as a source is widely variant. Classical music is almost noise-like in its statistics, but lots of rock music is highly peaked, not at all Gaussian. I ahd some success with correlating perception of distortion to the signal statistics, but I found that music was so wide in its content that even this was kind of futile.
ShinOBIWAN said:
Audibilty has never been thouroughly tested. My bet is on the direct excitation of the panels by the loudspeaker as opposed to the coupling through the air. Air coupling falls off as 1/f, but direct coupling of the vibrations is almost constant. The direct vibration is best combated by damping the front panel after which the vibrations of the box will not be very significant at all. A resonant baffle board is thus the worst case scenario.
I always well damp my baffles and the waveguides, but I only use a piece of oak as an internal cross brace. I buy oak flooring for $50 a bundle and I get enough Oak to last for 100's of cabinets. My previous enclosures were composite plastic - very well damped, good paint surface, looked great, very expensive. I became convinced that box resonances were not a big deal and so I am scraping these expensive cabinets and going to MDF (or anything else the customer is willing to pay for, I don't really care).
Music as a source is widely variant. Classical music is almost noise-like in its statistics, but lots of rock music is highly peaked, not at all Gaussian. I ahd some success with correlating perception of distortion to the signal statistics, but I found that music was so wide in its content that even this was kind of futile.
No More Re's!!
You said this in post #122:
Either it's significant or it's not. Pick one.
Not according to Linkwitz. He shows mechanical from the driver decreases with frequency rise.
Also advises the the best way to reduce this vibration is to isolate the driver flange from the baffle with gasket and couple the magnet directly to the box structure.
gedlee said:
You said this in post #122:
gedlee said:
Either it's significant or it's not. Pick one.
gedlee said:
Not according to Linkwitz. He shows mechanical from the driver decreases with frequency rise.
Also advises the the best way to reduce this vibration is to isolate the driver flange from the baffle with gasket and couple the magnet directly to the box structure.
Re: No More Re's!!
Might not have to pick one. No one said these were significant things. It doesn't sound like either the broadband stimulus exciting panel resonances nor the mechanical forces transferred between the driver-cabinet coupling are significant.
MJL21193 said:
Might not have to pick one. No one said these were significant things. It doesn't sound like either the broadband stimulus exciting panel resonances nor the mechanical forces transferred between the driver-cabinet coupling are significant.
Pied Piper??tinitus said:
pinkmouse said:
No politics please Tinitus. You know the rules.
geez... someone finally demonstrates some common sense here... and the mods jump down their throat...😀 😀
I agree with Ant... MDF works like a champ... as long as you pay attention to the details and understand its properties and limitations. Anyone ever used particle board for a cabinet...
I even built a prismatic pentagonal isobaric sub out of leftover 1/2" OSB 10 yrs. ago... still works great, but I'm gonna scavenge the drivers for the last two speakers in my (soon to be) 7.1 HT setup, seeing as I only have 20 drivers running as a 5.1 setup right now😉 😉
John L
auplater said:
I agree with Ant... MDF works like a champ... as long as you pay attention to the details and understand its properties and limitations. Anyone ever used particle board for a cabinet...
What are the important details we all should be paying attention to? Can you do a brief summary of the properties and limitations of MDF?
OSB:
Attachments
details
What's really fun is cutting miters, dados, compound angles in OSB... eh?
🙄
Funny you should ask. I' was trying to track down any significant details on plywood, MDF, Baltic Birch, etc. googling this afternoon, but came to the unfortunate conclusion that the technical details are difficult to pin down. I found all sorts of conflicting data re: strength in tension (MDF much lower than plywood parallel to the lams, about the same across laminations or even lower, due to joint failure). Strength in compression wasn't much different (kinda makes sense since they're both composits of wood and glue), but here again the data was highly variable depending on the source(s)... ah well, not like the aerospace flight tested data on metals and such which I'm all too familiar with...🙄
AFAIK, seems the material is much less important than the method re: bracing and construction. Ant uses the "heavy duty" approach and appently has exceptional results. What I've found using MDF is that as long as you pay attention to the joints by using both mechanical connections (dados, face fasteners, etc.) and high quality glue with adequate surface roughening and pre-treatment (re: pre-apply glue, allow to rest, then reapply b4 clamping) MDF seems as good as, or better than plywood I've used. It also doesn't suffer from blowouts/voids I've seen even in high quality stuff.
Plywood, OTOH, holds screws perpindicular to the plys much better than MDF, but not as good end on (I wouldn't recommend using end-on screwing for either, certainly needs pre-drilling at least as well as metal screw inserts). When I internally brace with MDF, I double the thickness b4 cutouts for larger spans to stiffen the flexural modes. I do the same with plywood (but I stopped using plywood after Hugo trashed the $$$$ market).
Finishing is different for sure, but I haven't had any real problems with either material, just needs differnt treatments (stain vs. paint, that sort of thing)
IMHO, the damping with MDF can be superior to plywood due to incresed density and higher acoustic losses at some frequencies. Of course, ymmv. These are just my experiences/preferences after building speakers for over 40 years.
John L.:
MJL21193 said:
What's really fun is cutting miters, dados, compound angles in OSB... eh?
🙄Funny you should ask. I' was trying to track down any significant details on plywood, MDF, Baltic Birch, etc. googling this afternoon, but came to the unfortunate conclusion that the technical details are difficult to pin down. I found all sorts of conflicting data re: strength in tension (MDF much lower than plywood parallel to the lams, about the same across laminations or even lower, due to joint failure). Strength in compression wasn't much different (kinda makes sense since they're both composits of wood and glue), but here again the data was highly variable depending on the source(s)... ah well, not like the aerospace flight tested data on metals and such which I'm all too familiar with...🙄
AFAIK, seems the material is much less important than the method re: bracing and construction. Ant uses the "heavy duty" approach and appently has exceptional results. What I've found using MDF is that as long as you pay attention to the joints by using both mechanical connections (dados, face fasteners, etc.) and high quality glue with adequate surface roughening and pre-treatment (re: pre-apply glue, allow to rest, then reapply b4 clamping) MDF seems as good as, or better than plywood I've used. It also doesn't suffer from blowouts/voids I've seen even in high quality stuff.
Plywood, OTOH, holds screws perpindicular to the plys much better than MDF, but not as good end on (I wouldn't recommend using end-on screwing for either, certainly needs pre-drilling at least as well as metal screw inserts). When I internally brace with MDF, I double the thickness b4 cutouts for larger spans to stiffen the flexural modes. I do the same with plywood (but I stopped using plywood after Hugo trashed the $$$$ market).
Finishing is different for sure, but I haven't had any real problems with either material, just needs differnt treatments (stain vs. paint, that sort of thing)
IMHO, the damping with MDF can be superior to plywood due to incresed density and higher acoustic losses at some frequencies. Of course, ymmv. These are just my experiences/preferences after building speakers for over 40 years.
John L.:
Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: and???
I'm aware of such vibrations. This video here able demonstrates the forces coming from a 10" sub driver (video courtesy of Vikash):
http://www.vikash.info/audio/xls10/830452_run_in.avi
Because of this I made a conscious effort to minimise such forces being transferred throughout the cabinet.
I choose to build separate baffles and place a sheet of sorbothane to decouple the baffle from the enclosure. The only problem then is that the baffle has to soak up up the energy so I choose brute force here once again and laminated 4 layers of 18mm MDF + one 9mm and then milled the details out of this block. Subjectively speaking it works well. I feel this approach renders the problem pretty much inert but have no data to back that up.
Few people have the equipment to quantify just how inert various construction approaches are. I know I do a great deal of guess work based loosely on idea's and accepted knowledge, I suspect much of the talk in this thread is likewise. What's not always clear is just how that guess work relates to and determines how dead any particular approach may be because, as you say, this is a complex subject. A safer bet from my point of view is go brute force.
Agreed but its always interesting to try such things. Having said that I've got all the pieces for a concrete mold kit here that I've had for the past couple of years. I don't feel compelled to go through the frankly huge effort to see if it improves things. Good enough seems OK for now.
I'm aware of such vibrations. This video here able demonstrates the forces coming from a 10" sub driver (video courtesy of Vikash):
http://www.vikash.info/audio/xls10/830452_run_in.avi
Because of this I made a conscious effort to minimise such forces being transferred throughout the cabinet.
An externally hosted image should be here but it was not working when we last tested it.
I choose to build separate baffles and place a sheet of sorbothane to decouple the baffle from the enclosure. The only problem then is that the baffle has to soak up up the energy so I choose brute force here once again and laminated 4 layers of 18mm MDF + one 9mm and then milled the details out of this block. Subjectively speaking it works well. I feel this approach renders the problem pretty much inert but have no data to back that up.
Few people have the equipment to quantify just how inert various construction approaches are. I know I do a great deal of guess work based loosely on idea's and accepted knowledge, I suspect much of the talk in this thread is likewise. What's not always clear is just how that guess work relates to and determines how dead any particular approach may be because, as you say, this is a complex subject. A safer bet from my point of view is go brute force.
Agreed but its always interesting to try such things. Having said that I've got all the pieces for a concrete mold kit here that I've had for the past couple of years. I don't feel compelled to go through the frankly huge effort to see if it improves things. Good enough seems OK for now.
Re: No More Re's!!
Good point. I'd have to say that all things considered its not significant. So why do I dampen the baffle at all. I might find in the future that even this is a waste of time. A lot of modern Pro cabinets are injection molded plastic and as weak as they can be. They ring like crazy too and yet they sell just as well as wood ones.
I'd have to think about the freq dependence of the driver reaction force, but at first guess since the motor force is constant with freq, I'd have to say that the reaction force would be too. It would take some analysis to really iron this one out.
I would seriously question trying to isolate the driver from the baffle as this allows the frame to move relative to the baffle and reduces the net cone motion. Clamping the magent just transfers this force somewhere else and it still has to be isolated or dealt with. The frame isolation would not be so bad except that it would be freq dependent and this would cause a non-flat net response. This too could be insignificant. But then it seems to me that this whole discussion is insignificant in the big picture of what we hear in the room.
MJL21193 said:
Good point. I'd have to say that all things considered its not significant. So why do I dampen the baffle at all. I might find in the future that even this is a waste of time. A lot of modern Pro cabinets are injection molded plastic and as weak as they can be. They ring like crazy too and yet they sell just as well as wood ones.
I'd have to think about the freq dependence of the driver reaction force, but at first guess since the motor force is constant with freq, I'd have to say that the reaction force would be too. It would take some analysis to really iron this one out.
I would seriously question trying to isolate the driver from the baffle as this allows the frame to move relative to the baffle and reduces the net cone motion. Clamping the magent just transfers this force somewhere else and it still has to be isolated or dealt with. The frame isolation would not be so bad except that it would be freq dependent and this would cause a non-flat net response. This too could be insignificant. But then it seems to me that this whole discussion is insignificant in the big picture of what we hear in the room.
dlr said:
There may be some significant issue, and it will almost certainly be measurable.. But can we measure it, or more likely can we determine a difference with the measurements we currently employ?
I've actually done this (quite a while back) and could NOT find any real measurement difference for the "suck the life out" phenom, and yet it was quite apparent - IN SOME INSTANCES. It was almost exclusively related to the baffle and what would commonly be referred to as more "efficient" drivers. The exception would be rear loaded horns - but that you can actually determine a measurable difference ..BUT perhaps *not* a difference that equates to full measure of what was perceived as a difference.
The same "sort of" holds true for stuffing near a driver. You can often see what would appear to be a BETTER decay and sometimes a smother FR, and yet the sound is often worse. (..and note, that many of the pricer models of "normal" tweeters from certain manufacturers seem to have noted the vary same thing and are willing to "give up" some degree of decay performance for a better sounding tweeter.)
For the most part though I think its a matter of execution - mdf in general is NOT inherently worse for loudspeaker designs.
As an example..
In the case of a back-loaded horn made of mdf you can remove most of the "suck the life out" issues by hard-coating the interior surface with a "high strength" paint, several layers. To a much lesser degree this will also work with driver's coupling to the baffle. Better still for the baffle, use a fairly weighty steel or brass "gasket". (..but again, this does seem to be dependent on the driver.)
Re: details
I don't have a problem with it.
Yes, the reliable data regarding MDF's properties seems to be elusive. Here is what I know:
MDF seems to be better damped than plywood.
MDF is not nearly as strong as plywood. A test: cut a strip of MDF and a strip of plywood. Measure how much force is needed before they each break when on edge. I can easily break pieces of MDF with my bare hands, whereas I could not break the same size piece of plywood.
MDF is not as rigid or flexible as plywood. Plywood will bend further than MDF without breaking.
MDF readily absorbs moisture and in doing so becomes weaker.
MDF will always split when screws are driven into it's edge, even with the correct size pilot hole. Heck, I have seen evidence of splitting from airgun finish nails on pieces that have been broken apart. When screws are driven close to the edge on the face of the MDF panel, tear-out may occur.
MDF is not as dimensional stable as plywood - it is alive with moisture content. The surface will shift and show texture, even under several layers of paint. I have this problem with my 3-way mains, the once mirror like reflection is now slightly mottled. It still feels perfectly smooth, but in the right light it's obvious what's happening. Needless to say, this is a little discouraging.
MDF is not as resilient as plywood. It will distort and not return to it's original shape. Used for a work table, spanned across a couple of saw horses, it will bow down under it's own weight.
I have heard it said that MDF is not as stiff as plywood across it's face when used as a panel. If this were true, wouldn't the resonant frequency of these two equal size and thickness panels be different?
auplater said:
What's really fun is cutting miters, dados, compound angles in OSB... eh?
I don't have a problem with it.
Yes, the reliable data regarding MDF's properties seems to be elusive. Here is what I know:
MDF seems to be better damped than plywood.
MDF is not nearly as strong as plywood. A test: cut a strip of MDF and a strip of plywood. Measure how much force is needed before they each break when on edge. I can easily break pieces of MDF with my bare hands, whereas I could not break the same size piece of plywood.
MDF is not as rigid or flexible as plywood. Plywood will bend further than MDF without breaking.
MDF readily absorbs moisture and in doing so becomes weaker.
MDF will always split when screws are driven into it's edge, even with the correct size pilot hole. Heck, I have seen evidence of splitting from airgun finish nails on pieces that have been broken apart. When screws are driven close to the edge on the face of the MDF panel, tear-out may occur.
MDF is not as dimensional stable as plywood - it is alive with moisture content. The surface will shift and show texture, even under several layers of paint. I have this problem with my 3-way mains, the once mirror like reflection is now slightly mottled. It still feels perfectly smooth, but in the right light it's obvious what's happening. Needless to say, this is a little discouraging.
MDF is not as resilient as plywood. It will distort and not return to it's original shape. Used for a work table, spanned across a couple of saw horses, it will bow down under it's own weight.
I have heard it said that MDF is not as stiff as plywood across it's face when used as a panel. If this were true, wouldn't the resonant frequency of these two equal size and thickness panels be different?
http://www.sparkfun.com/commerce/categories.php?cPath=23_80
accelerometers. For those that do think the forum is work and not play.
Another feature of MDF that I hate is that it tears and dents easily.
and my god, we do have a lot of smiles.
accelerometers. For those that do think the forum is work and not play.
Another feature of MDF that I hate is that it tears and dents easily.
and my god, we do have a lot of smiles.
- Status
- Not open for further replies.