Hi guys,
I posted this initially in the Multi-Way forum, but I've since decided to make this a fullrange speaker, so I'll continue on the thread here. The old thread is located here.
For those who didn't see the original thread, a quick recap:
I'm designing a speaker for an Object Design class at university. I'm an audio engineer, but I've never delved much into acoustics. I want to make something that sounds good, but unfortunately a major component of the course is making a challenging and interesting looking design, so I can't just make it look like a box on the outside, despite the research and science that goes into the architecture. I've got a great wood workshop available at university, but I can't use MDF there, so I'm trying to stay away from that if possible.
I'm tossing up between two ideas at the moment, and I'll decide in the next couple of days.
The first would look something like this
but in a cylindrical (U shaped box) or spherical shape. I'd cut up all the blocks into tiny little segments, and then put them together.
The second idea is to use rings to create the design, but this may waste a bit of material.
________________________________________________________________________________________________________________
Interesting design. I found this pic of the inside:
It does look quite similar to my design. Couple of questions:
-There's no material on the inside of that pic. Is that because they've taken it out to show you the inside, or does the speaker design not require any dampening?
-Could I do a design similar to this using hardwood instead of MDF? Or would the horizontal length warp?
Thanks again.
I'm also looking into drivers. I'm not overly restricted by money, so I've been looking at the EL70's which seem to be quite popular here. Thoughts?
I posted this initially in the Multi-Way forum, but I've since decided to make this a fullrange speaker, so I'll continue on the thread here. The old thread is located here.
For those who didn't see the original thread, a quick recap:
I'm designing a speaker for an Object Design class at university. I'm an audio engineer, but I've never delved much into acoustics. I want to make something that sounds good, but unfortunately a major component of the course is making a challenging and interesting looking design, so I can't just make it look like a box on the outside, despite the research and science that goes into the architecture. I've got a great wood workshop available at university, but I can't use MDF there, so I'm trying to stay away from that if possible.
I'm tossing up between two ideas at the moment, and I'll decide in the next couple of days.
The first would look something like this
An externally hosted image should be here but it was not working when we last tested it.
but in a cylindrical (U shaped box) or spherical shape. I'd cut up all the blocks into tiny little segments, and then put them together.
The second idea is to use rings to create the design, but this may waste a bit of material.
________________________________________________________________________________________________________________
Interesting design. I found this pic of the inside:
An externally hosted image should be here but it was not working when we last tested it.
It does look quite similar to my design. Couple of questions:
-There's no material on the inside of that pic. Is that because they've taken it out to show you the inside, or does the speaker design not require any dampening?
-Could I do a design similar to this using hardwood instead of MDF? Or would the horizontal length warp?
Thanks again.
I'm also looking into drivers. I'm not overly restricted by money, so I've been looking at the EL70's which seem to be quite popular here. Thoughts?
Last edited:
I'm interested in what happens with diffusive surfaces inside cabinets. The inside in that one pic looks like the egg crates people put on their studio walls- lots and lots of diffusion at only one frequency.
I think it's possible that breakup style diffusion- edges creating diffusion- inside cabs will lead to larger enclosure losses. I'm looking at using some convex curves inside a rectangular cab to avoid standing waves, but I've been leery of having edges sticking out inside- seems like it could easily cause turbulence. Turbulence is not good news in your duct, except perhaps for a bit around the edges to "grease" the flow of the main plug of air, according to the Salvatti et al AES paper on ports. Then again, I'm not sure this is a vented cabinet you're working on, so I may be on apples and oranges.
I'm building cabinets out of seasoned (14 year old, in fact) solid maple. I just can't get excited about plywood and mdf.
I think it's possible that breakup style diffusion- edges creating diffusion- inside cabs will lead to larger enclosure losses. I'm looking at using some convex curves inside a rectangular cab to avoid standing waves, but I've been leery of having edges sticking out inside- seems like it could easily cause turbulence. Turbulence is not good news in your duct, except perhaps for a bit around the edges to "grease" the flow of the main plug of air, according to the Salvatti et al AES paper on ports. Then again, I'm not sure this is a vented cabinet you're working on, so I may be on apples and oranges.
I'm building cabinets out of seasoned (14 year old, in fact) solid maple. I just can't get excited about plywood and mdf.
Greetings. May I ask what is the problem you want to solve?
You want to solve problems in the order of greatest impact. The box's -outer- geometry will arguably have greater impact than its -inner- geometry. And that's not even the biggest fish you have to fry!
But anyway, I'm not sure diffusing internal standing waves is of such great importance, nor that you can do it better than simply attenuating them through absorption with e.g., wool.
If you want to explore the (rather massive) impact that the baffle has on frequency response, consider downloading The Edge and get ready for some surprising results, e.g. a massive 3db lump and lots of ripples in the response, all courtesy of a "normal" baffle with center-positioned driver.
Also, on this forum under "Articles," read the "Arpeggio" article which is a general overview on fullranger design decisions and tradeoffs.
You want to solve problems in the order of greatest impact. The box's -outer- geometry will arguably have greater impact than its -inner- geometry. And that's not even the biggest fish you have to fry!
But anyway, I'm not sure diffusing internal standing waves is of such great importance, nor that you can do it better than simply attenuating them through absorption with e.g., wool.
If you want to explore the (rather massive) impact that the baffle has on frequency response, consider downloading The Edge and get ready for some surprising results, e.g. a massive 3db lump and lots of ripples in the response, all courtesy of a "normal" baffle with center-positioned driver.
Also, on this forum under "Articles," read the "Arpeggio" article which is a general overview on fullranger design decisions and tradeoffs.
Yes, it looks like you're wanting to put diffusors on the inside walls, something like this:
PME Records QRD Diffusor Construction
From the last post, the immediate thought is to put a "diffusor" on the front panel, but after a short moment's thought you don't want anything for the high frequencies to hit on, reflect off of, or diffract around.
Here's the page for that "Edge" software, I've seen it before but haven't used it:
Home of the Edge
Unfortunately it doesn't appear to show the effect of rounding off the front baffle edge around to the sides, top and bottom, which as I understand can significantly soften the bumps and dips along and above the baffle step response. I see this edge rounding in the smaller Infinity cabinets as well as others.
Here's more info:
Baffle Step Compensation
The sphere shape is "ideal" shape for the baffle, but perhaps the worst shape for the inside of the cabinet (maybe fighting it out with a cube, with all X Y and Z dimensions the same, giving a big resonance at the corresponding frequency). Maybe the front half be a hemisphere and the back half be something else to give appropriate internal volume?
PME Records QRD Diffusor Construction
From the last post, the immediate thought is to put a "diffusor" on the front panel, but after a short moment's thought you don't want anything for the high frequencies to hit on, reflect off of, or diffract around.
Here's the page for that "Edge" software, I've seen it before but haven't used it:
Home of the Edge
Unfortunately it doesn't appear to show the effect of rounding off the front baffle edge around to the sides, top and bottom, which as I understand can significantly soften the bumps and dips along and above the baffle step response. I see this edge rounding in the smaller Infinity cabinets as well as others.
Here's more info:
Baffle Step Compensation
The sphere shape is "ideal" shape for the baffle, but perhaps the worst shape for the inside of the cabinet (maybe fighting it out with a cube, with all X Y and Z dimensions the same, giving a big resonance at the corresponding frequency). Maybe the front half be a hemisphere and the back half be something else to give appropriate internal volume?
Thanks guys,
Just a quick update, I've spent all day checking out various drivers, playing around with the "Edge" software. I think at this point in time I'm going to go with a U shaped box, as thin as possible (these as desktop speakers would be nice), so maybe 150mm max. As tall and as deep as is necessary. Still not sure whether I should go for the rings or the blocks, I'm thinking blocks.
I'm also tossing up between the following drivers:
Alpair 6p, Alpair 5, CHP-70, Audience A3.
If anyone's got any further input that'd be great, otherwise I'll just keep researching and updating this as I go.
Just a quick update, I've spent all day checking out various drivers, playing around with the "Edge" software. I think at this point in time I'm going to go with a U shaped box, as thin as possible (these as desktop speakers would be nice), so maybe 150mm max. As tall and as deep as is necessary. Still not sure whether I should go for the rings or the blocks, I'm thinking blocks.
I'm also tossing up between the following drivers:
Alpair 6p, Alpair 5, CHP-70, Audience A3.
If anyone's got any further input that'd be great, otherwise I'll just keep researching and updating this as I go.
Okay, so the problem you're addressing is the need to build something with unconventional construction. I had thought you were addressing specific design goals of the final result.
So what you're doing sounds like fun, and you're exploring what's possible. But it's not really speaker design (yet) because there is no defined objective, e.g., (random example) a flat frequency response for Driver X, in dimensions no bigger than Y, down to bass frequency Z, given the desired positioning,etc.
So what you're doing sounds like fun, and you're exploring what's possible. But it's not really speaker design (yet) because there is no defined objective, e.g., (random example) a flat frequency response for Driver X, in dimensions no bigger than Y, down to bass frequency Z, given the desired positioning,etc.
It can be a theoreticaly myth, because many guys that built spheres haven't notice any resonance...
Yeah sorry I should have made that clearer. I do have some specific goals in mind at the moment, I'll do a bit for research tomorrow and hopefully post my requirements like the ones you just listed.
I don't have a huge amount of time to mess around with various designs unfortunately, so I've decided to take the BR/ML-TL Hybrid build here and change it into a U shaped design with my blocks. Does this sound feasible? Are there any particular things to keep in mind when I change the shape of the box, or should I just make sure to keep the volume the same?
This uses the Fountek FR88ex driver.
If you guys are able to confirm what I'm thinking and give me a "Yes this is a good idea" or "No stop horrible idea", then I'll go ahead and build a 3D Model so that I can put it together nice and easily.
This uses the Fountek FR88ex driver.
If you guys are able to confirm what I'm thinking and give me a "Yes this is a good idea" or "No stop horrible idea", then I'll go ahead and build a 3D Model so that I can put it together nice and easily.
It is easy to do a nice drawing. If it is an established design and you build EXACTLY what is drawn, then it is just woodwork. The physics of a TL are very complex. They are hard to tune, sensitive to stuffing densities, and from a woodworking side a lot more parts.
As this is your class project, not just building a nice pair of speakers, building someone else's published design is not what yo are looking for. I would think it would limit you in your block idea. Overly complex is kind of the opposite goal of industrial engineering. Aesthetics blended with creativity with the engineering to get a simple, pleasing yet effective, producible design.
I can build a very well engineered box-type speaker in an afternoon. They fail on the aesthetics part. You are looking to re-implement the second most complex box with a labor intensive assembly. When graded, think about the question of what does it do better?
Don't get me wrong. I have been thinking about what kinds of advantages in managing panel resonances your idea could help with. I may even make some test panels to do some measuring. I only suggest using the idea to make a unique shape from a simpler acoustic design first. There is a lot to learn about building speakers.
Bottom line is, you have lost a week on this forum where you could have done a couple of prototypes. Time to get sommething build.
As this is your class project, not just building a nice pair of speakers, building someone else's published design is not what yo are looking for. I would think it would limit you in your block idea. Overly complex is kind of the opposite goal of industrial engineering. Aesthetics blended with creativity with the engineering to get a simple, pleasing yet effective, producible design.
I can build a very well engineered box-type speaker in an afternoon. They fail on the aesthetics part. You are looking to re-implement the second most complex box with a labor intensive assembly. When graded, think about the question of what does it do better?
Don't get me wrong. I have been thinking about what kinds of advantages in managing panel resonances your idea could help with. I may even make some test panels to do some measuring. I only suggest using the idea to make a unique shape from a simpler acoustic design first. There is a lot to learn about building speakers.
Bottom line is, you have lost a week on this forum where you could have done a couple of prototypes. Time to get sommething build.
Ok great. Thanks.
I actually started putting a prototype together this week (some of the pieces aren't flush - I did about 6 with a different and inaccurate method, but put them in there anyway.. all the newer ones are perfect). Pics:
It's got the same volume as the design I posted, but semi-circular instead of rectangular. The issue I've got now is screwing every little piece together.. very difficult and time consuming. Is just glueing them together enough? Or could I thread long rods from top to bottom to keep them together?
I actually started putting a prototype together this week (some of the pieces aren't flush - I did about 6 with a different and inaccurate method, but put them in there anyway.. all the newer ones are perfect). Pics:
An externally hosted image should be here but it was not working when we last tested it.
It's got the same volume as the design I posted, but semi-circular instead of rectangular. The issue I've got now is screwing every little piece together.. very difficult and time consuming. Is just glueing them together enough? Or could I thread long rods from top to bottom to keep them together?
Gluing,
Since you are using bits that are aprox. the same size, and look to be same angle to each cut, glue the individual layers in circles ( only gluing the 2 half circles, not the entire circle, conversely you could glue the entire circle and cut into half circles after the adhesive has cured, this would give you a flat surface to attach the baffle to. ). When the layers have cured, then stack and glue the layers together.
The individual circles can be clamped with something as simple as a bit of string, or large elastic bands. The clamping force need not be excessive, just hold the bits together till the glue dries. Modern adhesives are generally stronger than the wood you will be joining anyway.
I would also probably put something on the top and bottom of the circles to keep the layers flat as the glue cured, this will make the stacking much simpler. If you have problems with the blocks moving about when the glue is still in the stage where its not tacky yet, an internal form could be constructed out of a simple cylinder of any material that will keep the blocks from moving 'till the glue sets.
I would also do the finish sanding of the outside surface of the blocks, and any staining, prior to glue-up. The "squeezle" ( Technical term for the glue that comes out of the joints, when clamped ) at the joints can be dealt with with careful scraping after it's cured, but glue's generally will not take stain.
Interesting concept, I'll look forward to seeing your final results.
John
Since you are using bits that are aprox. the same size, and look to be same angle to each cut, glue the individual layers in circles ( only gluing the 2 half circles, not the entire circle, conversely you could glue the entire circle and cut into half circles after the adhesive has cured, this would give you a flat surface to attach the baffle to. ). When the layers have cured, then stack and glue the layers together.
The individual circles can be clamped with something as simple as a bit of string, or large elastic bands. The clamping force need not be excessive, just hold the bits together till the glue dries. Modern adhesives are generally stronger than the wood you will be joining anyway.
I would also probably put something on the top and bottom of the circles to keep the layers flat as the glue cured, this will make the stacking much simpler. If you have problems with the blocks moving about when the glue is still in the stage where its not tacky yet, an internal form could be constructed out of a simple cylinder of any material that will keep the blocks from moving 'till the glue sets.
I would also do the finish sanding of the outside surface of the blocks, and any staining, prior to glue-up. The "squeezle" ( Technical term for the glue that comes out of the joints, when clamped ) at the joints can be dealt with with careful scraping after it's cured, but glue's generally will not take stain.
Interesting concept, I'll look forward to seeing your final results.
John
The project is due on the 30th of October.
I went out and bought some second-hand wood (a requirement of the project is to use recyclable/second-hand materials) yesterday, massive rip-off but oh well. Not sure what wood it is, but its similar to Cedar, according to my Dad. Light, no warping, absorbent.. looks to fit my needs pretty well.
I've glued the pieces shown above just now, so we'll see how that turns out. Using a piece of string to hold the pieces in a perfect circle while it glues. Then I'll cut it in half after it dries.
I'm unable to find the FR88EX to buy anywhere online. I can find the FR89EX, and it looks very similar, can I slot it in this same design without many issues? I have to buy it soon - it'll take two weeks to get here and I only have 3 weeks left.
I went out and bought some second-hand wood (a requirement of the project is to use recyclable/second-hand materials) yesterday, massive rip-off but oh well. Not sure what wood it is, but its similar to Cedar, according to my Dad. Light, no warping, absorbent.. looks to fit my needs pretty well.
I've glued the pieces shown above just now, so we'll see how that turns out. Using a piece of string to hold the pieces in a perfect circle while it glues. Then I'll cut it in half after it dries.
I'm unable to find the FR88EX to buy anywhere online. I can find the FR89EX, and it looks very similar, can I slot it in this same design without many issues? I have to buy it soon - it'll take two weeks to get here and I only have 3 weeks left.
With the available time left I would get what is available right now. The FR89EX is very similar to the FR88EX, you can use the same volume box, just change the vent length if you are going for the bass reflex design.
Some advice for 'free'. A good plan today is better than a perfect plan tommorow'!
I think for the object design class the emphasis is on the uniqness and design process and execution of the design. Correct me if I am wrong but the driver that you will be putting in the box is irrelevant, you could just use pc desktop speakers in the box, sound quality will only form a fraction of the total point allocation?
Keep us up to date with plenty of pics!
Some advice for 'free'. A good plan today is better than a perfect plan tommorow'!
I think for the object design class the emphasis is on the uniqness and design process and execution of the design. Correct me if I am wrong but the driver that you will be putting in the box is irrelevant, you could just use pc desktop speakers in the box, sound quality will only form a fraction of the total point allocation?
Keep us up to date with plenty of pics!
I've bought the FR89EX, it should be here within about two weeks. Not sure how I'll change the vent length, or how I calculate that? Don't need to worry about it for now.
Correct that the sound quality mightn't form much of the marks, but I also want something for my own personal use that'll sound as good as possible. I get to use this when its done!
I glued together the pieces I showed above, but the resulting structure was not very strong - I was able to break it just by squeezing the opposing sides of the circles. Is normal PVA glue enough? I can probably make it stronger by clamping it down and ensuring the blocks are forced into circle with a bit of elastic or something. Alternately, I could have little biscuits in between each blocks. I'd appreciate any advice here.
I've cut up 80 blocks out of my new wood, gotta cut another 20 tomorrow, then start glueing them together. I'll be glueing them into circles, then cutting them into semicircles (half through the glue line, half halfway through the block). 7 circles in total for 2 speakers. I'll then use the leftover wood I have to make the top/bottom/baffle out of planks. Sound like a good plan? Pics coming!
Correct that the sound quality mightn't form much of the marks, but I also want something for my own personal use that'll sound as good as possible. I get to use this when its done!
I glued together the pieces I showed above, but the resulting structure was not very strong - I was able to break it just by squeezing the opposing sides of the circles. Is normal PVA glue enough? I can probably make it stronger by clamping it down and ensuring the blocks are forced into circle with a bit of elastic or something. Alternately, I could have little biscuits in between each blocks. I'd appreciate any advice here.
I've cut up 80 blocks out of my new wood, gotta cut another 20 tomorrow, then start glueing them together. I'll be glueing them into circles, then cutting them into semicircles (half through the glue line, half halfway through the block). 7 circles in total for 2 speakers. I'll then use the leftover wood I have to make the top/bottom/baffle out of planks. Sound like a good plan? Pics coming!
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.