• The move to the new server is done. There are some software and database maintenance updates in process. This has us passing the hat around to help out. We appreciate any donations. Seriously, even a dollar helps. The payment page may be found here - https://www.audiokarma.org/support.html

AAAACCCCKKK!!! neophyte question: basic enclosure types

oneinchside

Openly Baffled
I put this over in the general forum and got no response, hopefully we have some speaker design type folks around here... I need some edjumacation.

Ok, I'm pretty confused, at least with horns... I'm going to write out what I *think* I know, mainly so those of you who actually do know can correct me. (please do, I'm here to learn)
basicspeakerenclosuretypes.jpg


Open baffle- optimal size seems to be 35.5"x31.5", placed 3' from the wall. Benefit is that there is no box coloration (does that mean that the wrong box can screw things up?) The front is "in" phase, the back while out of phase is separated by the baffle. The dimensions above seem to be enough to make it work.

open back-partial less bass than the open baffle? If placed too close to a wall the reflection from the opening can interfere with the signal from the front and cause comb filtering. The size of the box and the size of the opening seem to greatly influence the sound, optimum would be calculated base on Vas, Qts and Fs?

sealed box completely separated the in phase and out of phase signal. The trapped air may act like a "spring" and allow the driver to play lower than in an open design. Parallel surfaces may cause standing waves, which may effect the rear of the driver and in effect create comb filtering. The size of the box seems to greatly influence the sound, optimum would be calculated base on Vas, Qts and Fs?

ported box Like a sealed box but with a tube or port to allow some of the bass energy to come out. Size of the box, diameter and length or the port are all crucial as the shape of the port determines what frequency the box is tuned for. Again, the size of the box and the size of the opening seem to greatly influence the sound, optimum would be calculated base on Vas, Qts and Fs?

horn (is this a "back loaded" horn?) The horn acts as a guide not to actually amplify but to remove obstruction in the air so the sound can travel more efficiently. Longer horns would emphasize lower frequencies, are shorter horns as drawn used to improve separation and reduce coloration from other drivers? All of this is calculated mathematically, I guess using T/S parameters.
-----Why doesn't anyone use longer grammaphone style horns?

horn ????? The taper gives some support for the rear of the driver but still has an output for bass frequencies?
-----They look kinda cool?

Horn (is this "front loaded?") Seems to be a popular design from the 1950's to now. Provides midrange and bass, used in many one driver designs also. Approximately 638,987 bajilion variations. I guess it works? Do you need to use the T/S parameters to design something like this or is it just "the longer/bigger it is the lower the bass rolloff" Isn't the information coming out of the horn out of phase with the information coming off the cone?

Bass Horn Seems to very similar to the above but the driver is internal, a small sealed box behind the driver and then a very large horn on the other side. Since one side is sealed, all of the information is "in" phase.
-----Is this basically a big honking gramophone bell? Is all high/mid information lost in this design? Or is the high/mid just very dependent on the texture, shape, and reflectivity of the the internal surface making it darn near impossible to tune for HiFi?


Are there other types of enclosures that I'm not familiar with? I'm sure it's been done, but I would love to hear a shoot out using the same driver in the different types of enclosures.

What other criteria are important in designing a speaker enclosure? Obviously I'm getting ready to make some of my first enclosures, a little guidance would be greatly appreciated.

Thanks
Mike
 
Register to hide this ad
The one you have labelled "back loaded horn" is a front loaded horn. The one labelled "horn ?????" is a tapered transmission line. The one labelled "front loaded horn" is a back loaded horn (horn comes off the back of the cone). And the one labelled "bass horn ?????" is a front loaded horn functionally the same as the other front loaded horn (that you have labelled back loaded horn).

Open baffle- optimal size seems to be 35.5"x31.5", placed 3' from the wall. Benefit is that there is no box coloration (does that mean that the wrong box can screw things up?) The front is "in" phase, the back while out of phase is separated by the baffle. The dimensions above seem to be enough to make it work.

The optimal size really depends on what you want out of it. The bass will drop off where the wavelength is 2x the baffle width (aka the baffle width is 1/2 the wavelength). That's where the signal from the back of the driver wraps around and cancels the wave from the front. The size you quoted is an effective size but you could get more bass with a bigger baffle or less with a smaller one. It depends on the driver and your application.

open back-partial less bass than the open baffle? If placed too close to a wall the reflection from the opening can interfere with the signal from the front and cause comb filtering. The size of the box and the size of the opening seem to greatly influence the sound, optimum would be calculated base on Vas, Qts and Fs?

Functionally this is the same as an open baffle. The path length is just figured differently as it has to wrap around the sides as well as the front baffle. For this (or open baffle) you usually want to have a fairly stiff suspension (high Qts) where the cone motion will be controlled and not just flapping wildly.

sealed box completely separated the in phase and out of phase signal. The trapped air may act like a "spring" and allow the driver to play lower than in an open design. Parallel surfaces may cause standing waves, which may effect the rear of the driver and in effect create comb filtering. The size of the box seems to greatly influence the sound, optimum would be calculated base on Vas, Qts and Fs?

Right on with the exception of "comb filtering". That's not realy what happens, it's just a distortion as the standing wave will add to the speaker signal.

ported box Like a sealed box but with a tube or port to allow some of the bass energy to come out. Size of the box, diameter and length or the port are all crucial as the shape of the port determines what frequency the box is tuned for. Again, the size of the box and the size of the opening seem to greatly influence the sound, optimum would be calculated base on Vas, Qts and Fs?

A ported box really isn't like a sealed box. It's a 4th order system where the driver and the air mass in the port are acting together while being coupled by the air mass (spring) inside the box.

horn (is this a "back loaded" horn?) The horn acts as a guide not to actually amplify but to remove obstruction in the air so the sound can travel more efficiently. Longer horns would emphasize lower frequencies, are shorter horns as drawn used to improve separation and reduce coloration from other drivers? All of this is calculated mathematically, I guess using T/S parameters.
-----Why doesn't anyone use longer grammaphone style horns?

A horn is an impedance matching system between the driver diaphragm and the air. As to longer horns...they're still there, they just get folded up.

horn ????? The taper gives some support for the rear of the driver but still has an output for bass frequencies?
-----They look kinda cool?

It's a transmission line. The length of the line will set a resonance at 4x the length that will augment the bottom end of the driver...if done right. It's a lot more complicated than that but that gets the general idea across.

Horn (is this "front loaded?") Seems to be a popular design from the 1950's to now. Provides midrange and bass, used in many one driver designs also. Approximately 638,987 bajilion variations. I guess it works? Do you need to use the T/S parameters to design something like this or is it just "the longer/bigger it is the lower the bass rolloff" Isn't the information coming out of the horn out of phase with the information coming off the cone?

There are lots of calculations that go into the design of a good BLH. They will be based on the T/S parameters, the response of the driver, and the desired performance of the horn.

Bass Horn Seems to very similar to the above but the driver is internal, a small sealed box behind the driver and then a very large horn on the other side. Since one side is sealed, all of the information is "in" phase.
-----Is this basically a big honking gramophone bell? Is all high/mid information lost in this design? Or is the high/mid just very dependent on the texture, shape, and reflectivity of the the internal surface making it darn near impossible to tune for HiFi?

These are used all the time. Functionally, this is the same as the Avantgarde speakers that cost a bazillion dollars. The high frequency is limited by the throat of the horn, the low end by the length and mouth. The response shape will depend on the shape in between the throat and mouth.


Are there other types of enclosures that I'm not familiar with?

You left out bandpass boxes.

What other criteria are important in designing a speaker enclosure? Obviously I'm getting ready to make some of my first enclosures, a little guidance would be greatly appreciated.

The shape of the enclosure (especially in horns) is important, the materials used, damping within the enclosure, bracing, and if you're playing around with things, ability to change.

So, where are you going to start? :)
 
Last edited:
Yeah, and TQW ( Tappered Quarter Wave ) as well, though some people will include those in Transmission lines.

I was going to try a TQW as my first enclosure but I'm not sure if the any of the drivers I have would be a good match for it, but from a carpentry standpoint that's just about my upper level right now.

It looks like I'm going to be starting with an OB and just listen to every driver I can in it as a way to learn.
 
Last edited:
Mike, I highly recommend you purchase a copy of Vance Dickason's Loudspeaker Cookbook. It's got just about all the enclosure info. you could want. But, more importantly, a wealth of other loudspeaker info. you will undoubtedly need as you progress with your project.
A purchase you'll never regret.....
 
The one you have labelled "back loaded horn" is a front loaded horn. The one labelled "horn ?????" is a tapered transmission line. The one labelled "front loaded horn" is a back loaded horn (horn comes off the back of the cone). And the one labelled "bass horn ?????" is a front loaded horn functionally the same as the other front loaded horn (that you have labelled back loaded horn).

Ah, I had it bass ackwards. got it.


re:open baffle said:
The optimal size really depends on what you want out of it. The bass will drop off where the wavelength is 2x the baffle width (aka the baffle width is 1/2 the wavelength). That's where the signal from the back of the driver wraps around and cancels the wave from the front. The size you quoted is an effective size but you could get more bass with a bigger baffle or less with a smaller one. It depends on the driver and your application.
So if I increased the dimension vertically there wouldn't be much of a change? I was thinking about 35.5w x 42h. If I understand it correctly I may want to raise my woofer a proportional amount... but then I would lose some of the loading from the floor wouldn't I?


re:partially open back said:
Functionally this is the same as an open baffle. The path length is just figured differently as it has to wrap around the sides as well as the front baffle. For this (or open baffle) you usually want to have a fairly stiff suspension (high Qts) where the cone motion will be controlled and not just flapping wildly.
AHA!!! Being a guitar player the lack of thought that goes into guitar cabs is astounding, everybody basically does whatever Leo did. (of course I admit, I really like a lot of the things Leo did). I've always wondered if fairly small back openings (like in Dumbles) acted as a ported box (with a 6"x8" port that was 3/4 of an inch long, or what. I'm still trying to figure it out.


re:sealed box said:
Right on with the exception of "comb filtering". That's not realy what happens, it's just a distortion as the standing wave will add to the speaker signal.
So comb filtering refers to air from the back side mixing with air from the front and since the two are out of phase then some of the frequencies cancel each other out? (and that wouldn't happen with a closed box)


re:ported box said:
A ported box really isn't like a sealed box. It's a 4th order system where the driver and the air mass in the port are acting together while being coupled by the air mass (spring) inside the box.
and that sounds like a lot more than I'm ready to tackle.

A horn is an impedance matching system between the driver diaphragm and the air. As to longer horns...they're still there, they just get folded up.
Are they used for anything other than bass horns?

It's a transmission line. The length of the line will set a resonance at 4x the length that will augment the bottom end of the driver...if done right. It's a lot more complicated than that but that gets the general idea across.
So I guess it is possible then to mix a front loaded horn and a transmission line. I imagine it like a driver somewhere in the middle of a tuba.

There are lots of calculations that go into the design of a good BLH. They will be based on the T/S parameters, the response of the driver, and the desired performance of the horn.

These are used all the time. Functionally, this is the same as the Avantgarde speakers that cost a bazillion dollars. The high frequency is limited by the throat of the horn, the low end by the length and mouth. The response shape will depend on the shape in between the throat and mouth.

It seems like the math part of this has been figured out a long time ago, given the right parameters it can be plugged into an equation and the optimum shape would be generated. It get the feeling that in real life it 's not so simple...



You left out bandpass boxes.
Yes, I did. I guess I had only seen them in cars subwoofers.

The shape of the enclosure (especially in horns) is important, the materials used, damping within the enclosure, bracing, and if you're playing around with things, ability to change.

I really like the look of horns. That's where I want to head, but I don't even know where to start yet. Driver selection?

So, where are you going to start? :)
Tonight. I'm laying out my shopping list for my first set of OB's. :banana:

Thanks again
Mike
 

Right now I've got the drivers from 3 circa '65 consoles, plus the drivers in a Fisher console (that are sacrosanct and must not be f'd with) and some single drivers in guitar cabs, oh... and this set of RCA component speakers with "powered subwoofers" the power came from the BPC unit that they shipped with... it is dead and gone. It looks like something that came from WalMart or Best Buy 5 or 6 years ago. I could live with ripping those apart.

So in short,
2 6" midrange Motorola + 2 3" Moto tweets + 1 10" Moto woofer
2 2 zenith horns + 2 3" sealed back zenith mids + 2 10" zenith woofers
4 4" Columbia mid/tweets + 1 10" Columbia woofer

2 1" RCA BPC tweets + 2 2.5" mids + 2 mystery "subwoofers"

and after a few yard sales... who knows?
 
So if I increased the dimension vertically there wouldn't be much of a change? I was thinking about 35.5w x 42h. If I understand it correctly I may want to raise my woofer a proportional amount... but then I would lose some of the loading from the floor wouldn't I?

You will always get reflection from the floor (not really loading unless you're very close). The frequency depends on how far from the floor you are.

AHA!!! Being a guitar player the lack of thought that goes into guitar cabs is astounding, everybody basically does whatever Leo did. (of course I admit, I really like a lot of the things Leo did).

Getting the size right takes a little thinking and tinkering but most of the "sound" of a guitar amp comes from the driver itself and the amp driving it.

I've always wondered if fairly small back openings (like in Dumbles) acted as a ported box (with a 6"x8" port that was 3/4 of an inch long, or what. I'm still trying to figure it out.

It could. It depends on the driver. I seem to say that a lot, huh?

So comb filtering refers to air from the back side mixing with air from the front and since the two are out of phase then some of the frequencies cancel each other out? (and that wouldn't happen with a closed box)

Comb filtering comes from 2 sources playing the same (or very close) frequency. If you look at a polar plot you will see the level go up (constructive interference) and go down (destructive interference). If things are set up right, it looks like a comb.

and that sounds like a lot more than I'm ready to tackle.

There are many calculators for ported enclosures available. Quite a few free ones around the internet. The math is done for you.

Are they used for anything other than bass horns?

Front loaded horns? Yep, they're all over the place. The Altec model 19's that folks love have horn mid/tweeters. Paul Klipsch built a company out of horn speakers. Avantgarde. Magico Ultima. Econowave (granted, it's a waveguide but a similar mechanism).

So I guess it is possible then to mix a front loaded horn and a transmission line. I imagine it like a driver somewhere in the middle of a tuba.

Can't think of one off the top of my head but I'm sure somebody has tried it. I would think you would get very bizarre loading effects that way.

Yes, I did. I guess I had only seen them in cars subwoofers.

They get used indoors too sometimes. They can be picky.

I really like the look of horns. That's where I want to head, but I don't even know where to start yet. Driver selection?

Generally you start with a low Qts. That's a generalization, but a place to start.

The Cookbook is a good place to start learning more.
 
Good way to go for OBs. I think I had a pair of the 6" Motorola drivers -- are they twincones with a goldish-colored basket? If so, they were nice looking but pretty uninspiring sounding drivers. Those would still be worth a try on OBs I'd say.

Have you seen this thread?
http://www.audiokarma.org/forums/showthread.php?t=114192

This is what I came up with for the cutoff frequency (-3 dB point) for an OB based on
http://www.iol.ie/~waltonaudio/ob.html

attachment.php


Ray can probably comment meaningfully on the accuracy of the model.
 
Don't know about meaningfully, but it looks good to me. :thmbsp:
 
I responded at length in you general audio post just a short while ago so check there too. BTW OBs generally like Qts greater than about .7 or so, at least if you want bass response without EQ.

mike
 
I responded at length in you general audio post just a short while ago so check there too. BTW OBs generally like Qts greater than about .7 or so, at least if you want bass response without EQ.

mike


for those of you who haven't seen it, this was mashaffer's response:

Let me humbly try to make a few corrections and additions, then others can correct my corrections.

Open Baffle: One of the benefits is lack of box colorations. One of the biggest advantages is the effect of (mostly midrange) reflections of the back-wave on the sound stage and imaging. In my experience OB gives extremely good presence (in the sense of in the room with you sound not frequency balance) and is very forgiving of listening position short of directly 90 degrees off axis.

The optimum dimensions will depend on the driver in question. Some drivers with lowish Qts will not typically perform very well in any reasonably sized baffle. 3' is a good starting point but each room and speaker will be a little different and experimentation is in order.

You are correct that the wrong box can screw things up (this includes box design construction and stuffing etc) and to a certain degree even the right box causes negative effects. There are trade offs of course.

Open Back: The open box may have more or less bass for any given driver in comparison with the Open Baffle depending on the shortest effective path from the rear of the driver to the front (or probably more correctly to the ear). The length of this path determines the wavelengths (and thus frequencies) at which cancelation occurs. As you mention size of the box is important but so is the shape. Deep boxes with parallel sides tend to produce noticeable resonances which can cause strange colorations for example.

Closed Box: Whether the trapped air in the box acts as a spring or not depends on the size of the box relative to Vas. If the box is large in relation to Vas (say 10xVas) then there is very little "spring" loading. The sealed box improves bass response not because of a spring effect but because the back wave can no longer reach the front and cancel. In fact if the volume of the box is small enough to have a spring effect it actually raises the resonant frequency of the system making the low frequency roll off start earlier (and if small enough causing an actual bump in response in the mid/upper bass region). This is why up to a point a larger box gives more bass extension. Of course Qts has a major impact on when these things occur and to what extent.

Ported (or passive radiator) box: The port supports bass response because it is a resonator at a frequency determined by box size, port length and port cross sectional area. At and near a certain frequency the bass is reinforced by the resonating of the port (like blowing over the mouth of an empty bottle). Above and below this frequency the port becomes inactive in essence. Far enough below the tuned frequency the port becomes just a big leak and the response drops off rapidly (much faster than a sealed box). This is why sealed systems with a higher -3dB point may actually produce more of the true low bass than a ported system that specs out better. Also note that in many rooms the room gain nearly equals the roll off of a sealed system and can result in surprisingly good bass performance compared to the results in an open field.

Qts again comes into play in determining not only the appropriate tuning but whether the driver will work very well at all in this type of enclosure.

An interesting variation on the ported enclosure is the Extended Bass Shelf where a ported box is tuned well below the driver's natural resonance (fs) and the box increased in size to produce lower ultimate roll off but with an extended shelf slightly below the midrange output.

Horns: You have the names turned around. Your first example is a front loaded horn. That is to say, the horn which loads the driver is on the front. The one marked Horn ???? is a back loaded horn. The horn is on the back of the driver and reinforces the direct radiation off of the front of the driver.

The front loaded horn is generally limited to two or three octaves of effective range. Since the back loaded horn is usually just used to reinforce the direct radiation it is often for single driver systems.

The one you have listed as a Bass horn is a type of front loaded horn.

Actually the horn does amplify in a way. It increases efficiency some say by matching the acoustical impedance of the speaker with the air. There is some argument about the actual mechanisms but it does allow the driver to get more acoustical output from a given input signal.

The low frequency cut off is IIRC more a function of the mouth area and taper than the length per se. There are complex interactions that make predictions difficult without a software tool like "HornResp".

Another thing you might want to do is google tapped horn. This is a different type of horn (or some say transmission line) which is essentially a front loaded horn with the rear of the driver exposed in the mouth of the horn for some interesting results.

The main types you didn't mention that I know of are:

- passive radiator which is essentially a resonator like a ported box but where a drone cone is tuned to resonate at the desired frequency. It acts a lot like a ported box.

- transmission line which is much like a back loaded horn but without the large increasing taper of a horn. In fact many (most?) have a reverse tapper. It is a resonating type somewhat similar to a ported box. These are often referred to as quarter wave designs because the length of the transmission line is 1/4 of the wavelength to be reinforced.

- band pass which uses multiple resonance chambers. All of these that I have seen have the driver completely enclosed (no part is visible on the outside). Those cheap home theater subs with just a hole where the sound comes out are of this type. The response on both ends of the spectrum is determined by the tuning so that it passes a band of frequencies (hence the name). Quality systems can be made this way but it is very popular for cheap systems. These also seem to be popular in cars. I suppose due to size considerations.

Well that should be enough opportunity for me to embarrass myself. At least it ought to get the ball rolling.

mike

Thanks again!
 
Ray, you are just truck load of speaker knowledge! Got a website for a lecture series? (I am serious! one of those days I have to pick your brain!)
 
I try not to throw it around as I'd rather have the information stand on its own merit, but I have the coolest job in the world. I design speakers. :D

Ask away and I'll do what I can. FWIW, there are some guys around here with way more knowledge than me. Keep an eye open and you'll figure them out pretty quickly.
 
I try not to throw it around as I'd rather have the information stand on its own merit, but I have the coolest job in the world. I design speakers. :D

Ask away and I'll do what I can. FWIW, there are some guys around here with way more knowledge than me. Keep an eye open and you'll figure them out pretty quickly.

After looking at some of the links provided (and an hour with google) it seems that one option is to hinge the baffle, say 24" with 6" wings. From what I have read the bass response will drop as the wings are folded back but my ability to fit them into the house will go up exponentially. The height really doesn't seem to change things much, there seems to be no real benefit or detriment as far as I can see.

Other than the cost of hinges can you see any other drawbacks?
 
Oneinch; Do you have specs on any of the drivers you have? Pics? e may be able to suggest best OB candidates from what you have.

The graph above shows only the effect of cancelation on frequency response. The drivers fs and Qes will also affect the actual response. One useeful thing is to use the parameters of the driver in a program like boxplot and calculate the response in a huge (effectively infinite) box then superimpose the roll off from the baffle cancelation.

For example if your driver has a Qts of say 1.3 then there will be a substantial bump right at fs. If you design the baffle to start rolling off right where the response of the driver begins to rise you can extend flat response well below fs. Use a driver with Qts of .4 but otherwise similar and the response in the same baffle will start falling off way above fs.

mike
 
Oneinch; Do you have specs on any of the drivers you have? Pics? e may be able to suggest best OB candidates from what you have.

The graph above shows only the effect of cancelation on frequency response. The drivers fs and Qes will also affect the actual response. One useeful thing is to use the parameters of the driver in a program like boxplot and calculate the response in a huge (effectively infinite) box then superimpose the roll off from the baffle cancelation.

For example if your driver has a Qts of say 1.3 then there will be a substantial bump right at fs. If you design the baffle to start rolling off right where the response of the driver begins to rise you can extend flat response well below fs. Use a driver with Qts of .4 but otherwise similar and the response in the same baffle will start falling off way above fs.

mike

That's one of my problems, I haven't been able to find specs on any of them, and I don't have the woofer 3 program (or whatever it's called) so I can't derive them myself. It seems that some of the drivers in American consoles weren't American, and they didn't keep a lot of records on them. I *think* all of the drivers I have are actually Japanese, re-branded as Motorola, Zenith, or Columbia.

Part of the reason I wanted to go OB is because I didn't have any info on them, and I wanted to try several drivers so it seemed like the most "neutral" way of hearing them, and then if I like something I was hoping to lean toward "I think it needs a little more this..." and then think about a box or horn or something. (or maybe just the dustbin!)

Mike
 
I see. Drivers from consoles with open backs (or perforrated backs) should be good candidates. Some of those old drivers can be very nice. If you can smap a few pics I would be interested in seeing them. My first OB speakers used 8" full range drivers from a solid state console. I made 3' by 2' baffles with side panels and a top panel all of slightly different size (4-7") and drivers at ear level. They had usable but not powerful bass to about 60Hz and imaging that blew my mind. I have been hooked on OB ever since. Adding a sub made them wonderful.

mike
 
I see. Drivers from consoles with open backs (or perforrated backs) should be good candidates. Some of those old drivers can be very nice. If you can smap a few pics I would be interested in seeing them. My first OB speakers used 8" full range drivers from a solid state console. I made 3' by 2' baffles with side panels and a top panel all of slightly different size (4-7") and drivers at ear level. They had usable but not powerful bass to about 60Hz and imaging that blew my mind. I have been hooked on OB ever since. Adding a sub made them wonderful.

mike

Drivers from the Zenith:
ZenithZ932-10.jpg

ZenithZ932-12.jpg

ZenithZ932-13.jpg

ZenithZ932-17.jpg


Drivers from the Motorola:
Motorola-SK109-inside-half.jpg

Motorola-SK109-preamp-tubes.jpg


And I have a few singles that I'll try, but they probably are destined to be low watt (sub 5w) guitar amp speakers. It's the Zeniths that I'm really intrigued with.

Mike
 
Back
Top Bottom