Moving away from the regular this amp, that sub q's.

How many people take into account the effect of baffles on the dispersion of sound when planning your installs? Do they have an effect? How large is it?

What effect do baffles have on the speaker response? Has anyone actually tested differing types of baffles on the same driver to determine what actually changes? Convex, concave?

Cheers

Here is a thread that talks some what along the line of baffling, additional soundeadning etc... its not exactly what your after, but certainly a good read.
http://www.diymobileaudio.com/forum/showthread.php?t=34898
His approach to measure things is great & the results show that extra baffling, door treatments, deadning etc. do pay off. Hope that helps.

sure do!

ok, ppl put so much effort in box design. but the box is only 1/2 the equation, ie, when the speaker drives rearward.

wat happens when it moves forward? a lot!

most importantly are the 'SPL' effects of baffle design. from simple baffle gain to diffraction, step response (which is a transient response), the effects the proximity of a woofer has on a tweeter etc etc

its pretty darn interesting reading if u decide to look into it.

that is a baffling question

so lame i know

1- Not enough
2- Yes
3- Massive
4- Impedance coupling with the air, increasing sensitivity and efficiency. Eliminating phase cancellation due to front wave/back wave interaction. Increasing off-axis response by diffusing directly radiating waves.
5- Yes. "Application of the Geometric Theory of Diffraction (GTD) to Diffraction at the Edges of Loudspeaker Baffles" JAES V34, I10, pp771-779
"Improvement in Dome Loudspeaker Characteristics by Using a Spherical-Wave-Front Horn Baffle" JAES V36, I7/8 pp575-584

Concave - Can VERY effectively increase diffusion by 'opening' the sweet spot.
Convex - Not exactly a great choice, because it will become very difficult to manage in a car, but it can be quite effective (when utilised correctly) to increase the sweet spot as well. The Nautilus utilises this theory.

Is a baffle just an object behind the magnet of the woofer?

No. Baffle is a term used to describe a non-transparent (acoustically) material acting as a boundary between the rear and front waves, as the driver itself is mounted via the basket.

For a baffle to be effective, it must be (at a minimum) greater then 1/3 of the longest wavelength being reproduced, from the centre of the cone to the edge of the baffle. This prevents front and rear-wave interaction.

Depending on the angle of the baffle (whether it be flat, concave, right angled or sharper) will also determine the impedance loading the driver has on the air. You can gain more then 10dB 'for free' simply by designing a good baffle for the speaker.

An MDF spacer is not a baffle. Your door cards/trim/skin is not a baffle. The enclosures you see for subwoofers are the closest to being a baffle that people regularly use. Its just not carried through to other drivers.

So what is appropriate for a midbass driver as a baffle? Playing from 80-250htz?

MDF, around 12mm thick, on the entire door surface, would suffice as a sufficient baffle for door mounted midbass.

For a midrange crossed over at 250Hz, 115/250 = 0.46 metres, an MDF baffle needs to be 0.46meters in radius to be ultimately effective.

Some tweeters already have a great baffle (such as the Esotar), and work exceptionally well above 2.5k.

Keep in mind though, its not like an on/off switch. Smaller baffles will allow full impedance loading at the higher frequencies, but not at the lower ones in the drivers bandwidth. What you'll find is, a 5-6dB drop in response around this frequency as the gains are lost.

For example. You have a midrange, and make a baffle that is .5 metre in radius (lets say its ~ a circle). What will happen is a response due to the loading above 230Hz, but below that, you'll get a shelf and a 6dB drop. This is assuming that nothing else will assist in the baffle effect (such as a firewall, or a partition between the boot and the cabin, etc. etc.).

Is MDF needed? What else can be used?

So we've established that door skins are not baffles, but is a sealed and fully deadened door skin even close to a poor-to-fair version of a baffle, since there's no way most of us are going to be able to jam 12mm thick mdf between our door skins and the trims?

Also, what does this mean for pods? Is the baffle effect only as good as the thin bit of mdf they are screwed into? Or does the baffle that the pod is sealed to still have effect?

And for subs, does this mean that a box with the biggest front baffle possible will sound the best at the right internal volume? Or does there have to be a certain amount of depth beyond the motor for it to work properly?

I should point out that the whole mdf'ing the door thing is GREAT news for the install I'm doing in my uncle's holden fx ute. The entire door trim is a dead flat piece of wood with some vinyl on it. This is absolutely ideal, and the rest of the door is incredibly easy to deaden. Now I just have to convince him to spend a little more on drivers and a good HU and he can have a serious crack at the MEASQ comp!

Yeah but if it was between your door skin and trim then it wouldn't be acting as a baffle, the baffle is the front face of wherever your driver is mounted, so in that case the door trim would become the baffle and as we already ascertained not a very good one due.

In the case of the Fx, if you did make the door skin out of a flat piece of mdf then you would have yourself a very good baffle.

another way of thinking of what a baffle is, is it is the first reflective surface adjacent to the sound source (in our case, usually a point source)

so, if u were to mount a midbass driver in the usual place (bottom left hand corner) then the dash and firewall also become part of the baffle.

hell, even surfaces that aren't even physically connected to the driver can be used as baffles. remember ye 'ole corner loading that some HT setups used where the sub or midbass driver is placed in the corner? yup, the corner acts as a baffle (even tho they are seperated from the driver and what would be considered by many ppl as the initial 'reflective' surface) which leads us to an increase in SPL.

its not as simple as making a door a 12mm thick piece of timber, particularly given the usual proximity of other close reflecting surfaces. IF we were to place the driver in the middle of the door and have a full wavelength of material around the driver, then yes, we can then assume that the SPL variations are minimised (but not eliminated). however, we still have other baffle design considerations such as the proximity of other drivers (which results in combing and sometimes even a bizzare impedence loading, say if u have a midbass/sub next to the tweeter), baffle edge diffraction and of course, the very early reflections.

SO, in a car, we have to resort to waveguides in order to minimise SPL variations (using horns for example) or other, more subtle designs given the amount of room we have and other considerations

I'm in agreeance with Scorpion and Luke. Some good valid points raised.

As mentioned previously, its not a black and white issue. Its not on then off. Baffles dont need to be in direct contact with the driver, but become more effective when they are. Baffles dont need to be 12mm MDF, but are more effective when they are (or something else of similar/better reflective property).

I'll be mounting my midrange in the corner of the footwell, and my tweeters in the corners of my dash. The tweeter will use the A-pillar, the windscreen and the dash (deadened of course) as its baffle. At such high frequencies, the baffle doesnt need to be thick to be effective. My midrange will be using the kick, the floor and the firewall as its baffle, which obviously need to be thicker (then the dash) in order to be effective.

The interior of a car has so many different types of surfaces (most reflective in some way/shape/form) that can be utilised as baffles for our benefit. And yet none of them are (most of the time). The dash, the floor, the kicks, the parcel shelf (when reinforced, of course). When used properly, we can get a lot of gains from utilising these surfaces.

Hmm. direct contact with the driver? I thought that was bad?

what they mean is that the driver needs to be flush mounted, not surface or mounted at the top of a cone-type enclosure

im going to say a baffle is when a different reading is measured from a speaker that is tested in mid air hanging by a peice of string compared to a speaker with some form of mounting apparatus around it. As soon as a structure is placed around it and a different reading is noticed , then the speaker has begun to be baffled.

I'm Baffled.

Zion, re-read Luke, Steve, and Matt's post. They have explained pretty well what a baffle is.

It has nothing to do with measurements, although the result is you want a baffle that controls the dispersion of the driver in a certain way (generally for flat power response).

no
http://www.google.com.au/search?hl=en&...on&ct=title

Just to add to a great thread...

Speaker = An electro/magentic/mechanical diver mounted on a baffle. eg A DD sub mounted on MDF.

Driver = bare driver (everything that makes up what we call the speaker) as in suspended in air or by a piece of string as Zion's example. Eg the DD Sub alone.

i dont like that explanation as it also encompasses an enclosure. ie, u mount it in a box. however, if u have 1 box with damping on the rear wall and another box without damping, u can measure the difference in the response of the driver BUT u haven't changed the baffle.

its a bit confusing that explanation.

well u need to read the dictionary on what a baffle is.I can install a 4" speaker into a crumpet and it will have a baffle effect , but the wine and cheese brigade of audiophiles wont call it a baffle.

well the dictionary is wrong, or u read it wrong

I like zion's explanation...

made plenty of sense to me...

bloody cheese and wine audiophiles...

Zion, no offence, but its best used to utilise some brain matter in understanding something. If it doesnt sound right, it probably isnt.

Ask yourself, how on God's Green earth will you be able to get accurate measurements from a speaker suspended by string? Is it just me, or is that the most ridiculous thing you've seen all day? I saw the google link, and also abide by the rule "just because its on the net, doesnt mean its true".

Drivers are measured FIXED down to prevent movement, either open air or infinite baffle. String aint doing anyone here any favours. Please restrain from arguing this point, and lets keep the level of discussion here intelligent.

To answer some of the other questions, for something to qualify as a baffle, as Steve has said, it needs to be reflective. Keep in mind that reflective properties change with wavelength, so different requirements are needed for baffles of different frequency bandwidths.

A Speaker is a whole-part. Baffle, enclosure, driver/s. The whole kitten-caboodle. A driver is simply a single electro-magnetic transducer.

You can have baffles of all shapes and sizes, but you want to aim for the most EFFECTIVE baffle, that is PRACTICAL in your situation, within your bandwidth. A proper baffle is FREE output. It costs you nothing. No power cost, no mechanical cost, no suspension cost. Free. With the same power (and hence same thermal properties) and the same cone displacement (hence no extra movement) you can get upto 12+dB MORE output from your driver. YOU CAN GET LOUDER WITH LESS!!!

If you have a perfect attachment, you'll get the entire amount. Any 'imperfection' will assist is robbing you of some of the benefits, and some may contribute to other problems. Baffle step response and driver coupling are just two examples of such.

ahhh, im sorry to push aside the wine and cheese tasting here ,but if there is an object that causes a change in the pathway of any energy whether is sewerage ,light or sound then that energy has begun to be baffled.to me that object can be called a baffle.Its true that to avoid misunderstanding that baffle proberly should be called a crumpet baffle.

So, ummm... if I mount my mids in to the stock speaker locations using MDF spacers, are the baffles the spacers + the door frames?

Where does the stock trim that goes over the speakers come in to it...

but thats not what u originally said. it may be what was going through ur mind. I understand what u were TRYING to say, but it may not be very clear to others what u were trying to say. hence my reply saying that wat u said was confusing.

then u rambled off on something about something coz i lost interest in trying to decipher ur posts.

I'm just going to ignore the stupid comments zion. You can either smarten up, grow up and contribute properly, or not at all.

No. A spacer is not a baffle. This is because it isnt wide enough. Its only the width of the basket. It needs to encompass pretty much the entire door. The door frame itself is aluminium/steel, and is a useless baffle (because it doesnt reflect sound. Sound passes straight through it).

Again, the trim is useless if it doesnt reflect the sound. It might as well not be there at all for 'all the good it will do'.

I think some people are taking the dictionary discription of a baffle vs what in practical terms is actualy a baffle that we can plan and use to our advantage and what we are referring too, vs something that is really just another surface to reflect off.

"thematt" I may be of target here but would one way to differtiate between what is acting as a reflective surface and what is really acting as a practical baffle that is working correctly, would be with a RTA vs Impulse/ILS etc measurement, since an RTA will measure everything including reflections whereas a Impulse measurement will only measure the first wave, which would a baffle advantage show up as part of the first wave or would it still show up as a reflection?

Luke

This makes for interesting, but confusing reading.


How does a baffle make the sound louder? The rear sound waves bounce off the baffle, joining the front waves and add to it? Therefore making it louder?

No, not in the way that we are talking about a baffle, the terms in which we are talking about using a baffle is trying to stop exactly that, essentially a sound wave tries to travel outwards in perfect sphere, so alot of the waves that are emitted from the forward face of your drivers cone when they reach the drivers edge try to spill backwards away from the listener, and what we are trying to do is stop those waves from going in that direction and directing them back towards the listener. So in other words we are redirecting/recovering what would noramlly be lost sound or just reflect of random objects to produce a cluttered sound.

So you can imagine a concave baffle design will be very efficient in that it will direct everything back to the listener.

I'm not sure how technially correct some of my terms are but hopefully the discriptions helped clear it up a bit.

my words will stand the test of time .
Im not even gonna cut and paste them here .
???heh.
Do u know what u book worms do? you complicate things ,u make out like everything has to be designed with the greatest of knowledge.
well guess what in simple terms the whole inner door is a baffle devise.

Like cheese and yougurt are both dairy products.
Inner door and complicated design are both baffles.

So a baffle just redirects sound waves back toward the listener? And the size of the baffle dictates which frequencies are able to be bounced back?

Corey, you're in the SQ forum, in a topic that is focused on advanced techniques. If you can't understand or dont like the level of topic discussed, go hang around the Beginners Forum.

Daniel, what you've linked to is a Baffle too. We use them all the time to 'deflect' sound energy, usually away from a listener (have a look along the freeways or motorways - they're used along there all the time).

A baffle in the purest sense simply directs energy through reflection. The use of such a baffle in our cars allows us to take the rear-acting part of the front-firing wave (ignoring the wave off the rear of the cone) and reflect it back into our environment. This gives as a boost in energy, and when done correctly, the rear-acting wave will be inphase with the front-acting wave, and will produce a simple boost. ~6dB for a basic flat surface (near-field).

If you're on the other side of this baffle (like people living near motorways/freeways) you'll notice a drop in the level of sound because the energy is being reflected back into the vehicle.

And you've hit the nail on the head in regards to size of baffle related to frequency (or actually, wavelength). To find your baffle size, divide 115 by your frequency (hz). A full 'baffle' is large enough to reflect the entire wavelength, measured from the centre of the cone to the edge of the baffle.

If the baffle is smaller, those larger frequencies will have the last part of their wavelength 'curve' around the side of the baffle. This is Baffle Step Diffraction Response. The result is not nice, will give you a noticeable drop in your frequency response (imagine seeing a flat line, and then a >3dB step down) and can produce nasty artifacts in your sound.

If the baffle is the correct side for full baffling of the target frequency, what about when smaller wavelengths are generated ie a midrange speaker?

Does this mean that the next wavelength down will have less benefit and so on until there are waves that are negatively affected and so on until a wavelength size that gets the entire benefit again? How might that be calculated?

I showed that without identifying a base foundation or a common model,in which all further advancements can come from will be going against the discussion.I showed the common model in that once the free air speakers dispursed energy pattern is changed then the structure that changes it is called a baffle .
U need to understand that i have done this and really u are just typing replies i dont even listen to. Im not the one that says a baffle is this and this and no thats not a baffle.The word baffle was around before a speaker was invented so please dont try and use it in the context of some type of elaborate high end society book only the rich folk could understand.Either create your own words to describe the understandings of "theMAtt", or take your own advise.

bam.

I'm with you, zion.

Now, can we PELASE move on from semantics, and more towards using our collective knowledge to design successful baffles in our cars? I need to know both theory and practicalities, and so do most people here.

nuf' said

peace yo

lol. no, all u explained was a partially correct definition of a baffle. but ur description also accounts for other reflections that aren't part of the baffle.

so for instance, using ur definition, if i put the free air speaker into a room (without an enclosure) then the change in the transient response and SPL of the driver is due to the baffle. this is NOT correct when we are talking acoustic engineering. the baffle MUST be the first reflective surface around the driver. not ANY surface that changes the response of the driver like u are trying to say. i understand what u are TRYING to say, but what u are ACTUALLY saying is not what u are trying to say. the reason why we have to be careful with this definition is because the concept of a baffle is very easy to understand, however, u dont want to mix it up with other phenomenae such as reflective surfaces that create reverb fields.

yes the word baffle was around before speakers, but u are completely missing the context. thats something a 4 year old does. why are u trying to alienate someone who is obviously much more intelligent than the average? (aka, thematt) do u not like a mature and interesting conversation about acoustics, or are u just trying to big note urself?



we aren't talking semantics, we are trying to define what a baffle is. if u dont know what a baffle is, then how can u design a successful baffle other than by accident? what corey is saying is not actually 100% true. because once u delve into baffles, u also have to be very careful not to confuse the effects of the baffle with a reverb field. for instance, i could put a tweeter that plays 20kHz in a corner and change the way it sounds. BUT, because of the very short wavelength, the corner doesn't act as a baffle, so the change that occurs is due to a fundamental change in the reverb field.

this is why we are talking semantics. its important to correctly define something so that everyone can understand what it is, because it has very important implications later on.

If we could move back to the practicalities of what constitutes a reasonable baffle in the car...



Okay, so the door frame is useless as a baffle, as is an MDF spacer.

Although a sheet of MDF is a good baffle.

Questions:

With flush mounted speakers, does sound deadening the doors improve it's properties as a baffle?

What is the best way to simulate a concave baffle in a car? e.g. Mounting in the kick panels rather than the doors? Could you do something interesting like surrounding your drivers with strips of sound deadener (or MDF) of progressively increasing densities?

nope i feel it is acceptable to have a simple standard base to start from.
and when u build your complex structure and test it then u can call it a reverb field baffle or what ever u wanna call it.

Why would u think i would alienate matt, who are u???
Big note myself? ummmm i purposely never book study anything
U better look at the link again and dicover that the word baffle is used for many many things and for good reason , its cause all those things create the baffle effect!

http://www.google.com.au/search?hl=en&...on&ct=title

Take a look at my install thread and see how many times i went about trying to build an effective baffle effect, purposly making various designs to try and produce the effect that i thought would be good.Each side of the internal door i think is a baffle structure much like each side of a boats walls hold the water out, yet each boat wall does have a specific name.
When u get people comming in and asking what a baffle is and there u are with your wine and cheese and say , no u cant use the word baffle in that context cause we are audiophiles and we like to have complex meanings, plus we have taken the simple meaning of the word baffle which was always simple and made it something different.

But Zion by your definition any surface around a driver that will change the response of a driver you are calling a baffle, when that is simply not true.

If I put a AP mat over my driver it changes the response, but it's not a baffle, yet by your definition it is, a grill could make a minor response change, again it's a grill not a baffle, or look at CDT with that little arm that holds the tweeter over the driver cone, you could probably measure a response difference from that, but it's not a baffle it's a stupid little arm that holds a tweeter.

There's an acoustic advantage baffle, and then there is just random reflecting surfaces, that may or may not work as an acoustic advantage. I think that, that is alot more simpler explanation then what your trying to call Baffle's (essentially any physical object in the drivers environment).

spacer is useless because its no wider than the driver. door frame isn't useless as a baffle, it just causes a lot of issues. it allows part of the midbass soundwaves to move through it. (assuming it is bare metal and no trim)


the MDF creates a very stiff surface. this is good because it means the midbass isn't going to pass through it (partially) by moving the baffle (unlike sheet metal)

only if it makes it stiffer or increases the mass of the door. (i go stiffer AND increase mass of the door for midbass)



why do u want a concave baffle? wat are u trying to achieve? if u just want a straight up SPL boost, then thats something u might aim for. wat u really want is a baffle that produces no SPL variations. ie, u dont want the baffle to be creating a 6db boost @ 300 Hz and a -6dB boost @ 600Hz. that would be to hard and gay. unfortunately the best baffle for the least (all baffles change the SPL variation) variation is a sphere. gg cars. otherwise u could try a horn (to hard for me to try to explain)

u *could* try making a concave baffle, but it would also depend on the frequency being played. u dont want the midbass to be re-radiating once it exists the 'mouth' of the baffle.

concave is to hard in a door. placing the driver in a corner is going to be much much easier to design. ie, the kickpanel where u can use the dash as a waveguide to increase the stage height, but at the same time, could potentially do strange things to the response due to unwanted reflections. (ie reflections off the sidewall or top can then result in combing, lobing or steering). this is where we use polar plots.

u also have to consider other drivers and the relative location to each other. this can be particularly tricky as u can gain benefits such as smoother responses and increased tranisent response but this is dependent on the driver. so to work out which works best would require experimenting and using polar plots to predict certain drivers responses, paying particular attention top issues such as lobing.

baffle design is NOT about simply obtaining an SPL boost if u want to gain tangible audible benefits. baffle design is about smoothing the SPL variations. we have to be very careful with baffle edge diffraction and the shape of the baffle creating unwanted reflections. a concave, whilst inscreasing SPL, suffers terribly from unwanted reflections UNLESS you design for the reflections (same as a corner loaded dirver). a good example of this is a well designed horn. every aspect of a horn should be designed to prevent unwanted reflections.

Hmm. So in a door, do people attempt to use the sealed up door trim ( with sounddeadener and mdf) as a reflective surface for SPL variation smoothing? lol semantics

You build the baffle to the longest wavelength, to ensure that all the smaller wavelengths (of higher frequencies) receive the same benefits. After a full wavelength is formed, the baffle size becomes less important, and you don't get the same problems you would if the wavelengths were larger than the baffle. A wavelength is the reciprocal of frequency. As the frequency gets lower, the wavelength gets larger.

To calculate wavelength size, divide 115 by frequency. That will give you baffle length in metres. You then plan a baffle to be that size from the cone to the edge. This will be an effective baffle from that frequency, on up.

For example. I'm installing a mid at home, and want to know the size of the baffle to achieve 2pi steradians (half space) across its entire bandwidth. The mid plays from 5kHz down to 1kHz. Lower frequencies wavelength is 115/1000 = 0.115 metres, or 11cm. Therefore, my baffle needs to be at least 11cm in distance from the cone to the edge. So I make a square baffle, 30cm in length/width. This will be effective for that bandwidth.

Lets now say I'm in the car. I'm now planning to install my midrange in the kick. Same frequency range though. My floor is a good baffle (because its reinforced steel, very thick). My firewall is a good baffle (again, very thick steel). Only problem is the kick itself, to which I'll add my flat MDF to.

The underside of the dash is now my only problem. I'll need to start investigating ways to fix that too.....

Yes, but its not the SD itself. Its the aluminum backing. Not fantastic by any means, because its soooo thin, but its better than nothing. It will cause problems because its not flat though (keeping in mind my benefits arise by reflecting the wave back into the front wave, in phase - and this is very difficult if my surface isnt flat). Better off putting either MDF down, or a sheet of aluminium (seriously, 1mm will make a difference).

Best way to simulate a concave baffle is to build one. Fibreglass or MDF is cheapest. I think by that second question, you mean thickness not density yeah? In which case, thats a clever option, but a slow and arduous one!!! Build it like a pyramid with MDF, and then go nuts with a sander (nice and smooth sides to the shape) with your driver sitting in the middle. Unfortunately, like Scorp mentioned, its not appropriate for wide bandwidth drivers because of all the other wavelength problems you'll get. It should be ok for something like midbass, if designed correctly, but not much else.

Or you can go the next step, and start utilizing the shape of the interior of your car (with a few tweaks) to start behaving like a waveguide (which is exactly what Scop is saying!!!). A waveguide is basically a 'tuned' baffle that directs the wave in a very sharp direction, and allows high impedance loading of the cone (and hence a massive bump in 'free' SPL). <a href="http://www.trueaudio.com/st_spcs1.htm" target="_blank">http://www.trueaudio.com/st_spcs1.htm</a>

To give you an example of how hard this is to REALLY achieve, Scorpion mentions horns. The waveguide used on the hornbody is designed to eliminate as many internal reflections as possible, but a lot still fail (those with a 'honky' sound). Its not easy to incorporate into a car, but knowing about it is the first step!!!

My understanding is we only need the baffle to bea qtr wavelength in size not a full wavelength, otherwise a waveguide suitable below 1000hz is way too big to fit in a car.

Quarter wavelength rule is definitely for dispersion, I'll have to check whether its the same for baffling. 99% sure its not though. Most noticeable with tweeter baffle step diffraction.

And the whole point is to encourage you to think about whats in your car already, that you can use to assist you to create a baffle in the car.


***Edit***
Found a calculation that states f3 = 115 / WB, where Wb is the baffle width. So its about 1/3 of what I was saying....

Ill edit my previous remarks to suit!!

At 250hz we get a wavelength of almost 1.4 meters.
If we can get the mid in the kick, then this wavelength would be longer then the path to our ears.
The end of the cycle would effectively be behind us... What detrimental effects does this cause?
Not to mention it would be near impossible to make a suitable baffle for this (in home too)...

I have always thought baffle step starts at 1/4 wavelength of the frequency...

**EDIT**
Interesting...

Read above. Some guy called Olson discovered it, and supposedly mapped it too....