Ok here goes.
I've built a 500WPC stereo amplifier specifically to drive two JBL GTO1220D subs. Each speaker gets its own channel of bass-driving juice. These subs were selected because their response in a vented enclosure is relatively flat (easily controlled by a nice 9-band sub-EQ).
As for the enclosure, I know I want them both in the same box (though obviously in separate chambers). Trouble is, woodworking is not my forte and I'm stuck on what construction methods to employ.
I was thinking of using 3/4" MDF, but have my doubts as to how it will hold up with up to 1000W of power coming through the speakers. I foresee a lot of wall flex if braces are not used.
So my questions are:
Should I use a thicker MDF?
What type of joints should I use? Glue and screw or other?
If I should use bracing, what kind?
Any and all suggestions will be appreciated!
JBL Sub Site
Box Specs and Performance
I've built a 500WPC stereo amplifier specifically to drive two JBL GTO1220D subs. Each speaker gets its own channel of bass-driving juice. These subs were selected because their response in a vented enclosure is relatively flat (easily controlled by a nice 9-band sub-EQ).
As for the enclosure, I know I want them both in the same box (though obviously in separate chambers). Trouble is, woodworking is not my forte and I'm stuck on what construction methods to employ.
I was thinking of using 3/4" MDF, but have my doubts as to how it will hold up with up to 1000W of power coming through the speakers. I foresee a lot of wall flex if braces are not used.
So my questions are:
Should I use a thicker MDF?
What type of joints should I use? Glue and screw or other?
If I should use bracing, what kind?
Any and all suggestions will be appreciated!
JBL Sub Site
Box Specs and Performance
3/4 inches thick MDF is a nice start.
Good wood glue is enough, it's even better than screws but you need wood clamps. You can use glue with screws if you don't have wood clamps.
Remember to seal very very well your box, any air leak, as small as you could imagine, will compromise the performance ALOT.
You should use bracing.
I suggest this kind of bracing...
Good wood glue is enough, it's even better than screws but you need wood clamps. You can use glue with screws if you don't have wood clamps.
Remember to seal very very well your box, any air leak, as small as you could imagine, will compromise the performance ALOT.
You should use bracing.
I suggest this kind of bracing...
Attachments
Use 1 inch MDF, 18mm just isn't rigid enough. Make sure it is well braced, preferably with shelf braces.
Draw up plans, take it to a local supplier and get them to cut it all for you. Saves a lot of dust and time.
The drivers don't need to be flush mounted, so you can cut the holes with a jig saw if you don't have a router.
Screws will help hold things together while the glue dries if you don't have clamps. I use cleats in all joints to guarantee an air tight fit.
Cheers
David
Draw up plans, take it to a local supplier and get them to cut it all for you. Saves a lot of dust and time.
The drivers don't need to be flush mounted, so you can cut the holes with a jig saw if you don't have a router.
Screws will help hold things together while the glue dries if you don't have clamps. I use cleats in all joints to guarantee an air tight fit.
Cheers
David
Attachments
Thanks Dave!
I've started construction using 3/4" MDF (1" was not available at any local lumberyard). My neighbor has a nice woodshop so I used his equipment to make the cuts I needed. The outter shell is complete and now I will be starting in on the bracing.
Photos soon. I'll also run a separate thread detailing my DIY sub amp.
~Aaron
I've started construction using 3/4" MDF (1" was not available at any local lumberyard). My neighbor has a nice woodshop so I used his equipment to make the cuts I needed. The outter shell is complete and now I will be starting in on the bracing.
Photos soon. I'll also run a separate thread detailing my DIY sub amp.
~Aaron
If you woodworking skills are a bit shy, then I would try a Sonotube sub, they’re easy and no bracing is necessary.
My Sonosub Project
http://kingdaddy.linaeum.com/Sonosub/
No need to get fancy like mine, there are many others out there much easier.
My Sonosub Project
http://kingdaddy.linaeum.com/Sonosub/
No need to get fancy like mine, there are many others out there much easier.
Like any sub, the alignment and enclosure size is very important to the sound quality. Too big a box yields a soft less punchy sound that may be to mundane for movies, too small an enclosure will make the driver bloated and uneven at certain frequencies. Port size and tuning it also a bit of an art. I made several prototypes and tried them out to get the sound I wanted, I had enough Sonotube to make two at a time, one big and one small, and then A/B compared them side-by-side. This kind of testing will give you a good reference to what the different enclosure sized will sound like with a given driver. I prefer a slightly larger then optimum box for a smoother FR and better extension.
To answer your question, I think these sound great, as good or better then any other box sub I’ve heard, but not as good as my recently completed 4-18" Avalanche IB, which is why they are now sitting in a back room. I'm actually thinking about selling them, but I doubt I could get what I put into them.
To answer your question, I think these sound great, as good or better then any other box sub I’ve heard, but not as good as my recently completed 4-18" Avalanche IB, which is why they are now sitting in a back room. I'm actually thinking about selling them, but I doubt I could get what I put into them.
I have a related question: are there no rules of thumb for the maximum spans for planning bracing? I would think that, for example, 3/4" MDF, when used for a specific amount of power in a specific size enclosure, would have a specific "maximum recommended span" between braces and/or box corners. But I've never seen anything that gave such guidelines. Do any exist anywhere?
dscline said:
Excellent question. I've searched but found no reports that document the material characteristics of MDF. With vibration analysis (to determine flex, elasticity, rigidity, etc), you could determine the bracing needed to constrain to your individual specifications. Anyone want to volunteer to do some research??
Speaker companies often do such research, but do not typically publish their results/findings. We have to rely on information readily available to the "layman".
It's not really because of the box size or the power used.
The main reason is to push the panel vibration mode, causing distortion, higher in frequency so out of the subwoofer band, so the subwoofer doesn't excite it anymore.
I think that the panel vibration mode is at a quarterwave of the distance between two braces. (Don't remember if it's a quarter or half).
A rule of thumb is to brace at every six(minimum) to twelve(maximum) inches. You also should avoid to always use the same distance between braces because the panel vibration mode will be at a single frequency so you'll get alot of vibration at that frequency instead of spreading out the vibrations over a range.
Speed of sound = 343.4 meters per second.
6 inches = 0.1524 meters
12 inches = 0.3048 meters
6 inches will push panel vibration modes over 563 Hz.
12 inches will push panel vibration modes over 282 Hz.
This is a simple model, because we use rectangular panels which have a series of panel vibration modes, not spherical panels which would have a single panel vibration mode.
The main reason is to push the panel vibration mode, causing distortion, higher in frequency so out of the subwoofer band, so the subwoofer doesn't excite it anymore.
I think that the panel vibration mode is at a quarterwave of the distance between two braces. (Don't remember if it's a quarter or half).
A rule of thumb is to brace at every six(minimum) to twelve(maximum) inches. You also should avoid to always use the same distance between braces because the panel vibration mode will be at a single frequency so you'll get alot of vibration at that frequency instead of spreading out the vibrations over a range.
Speed of sound = 343.4 meters per second.
6 inches = 0.1524 meters
12 inches = 0.3048 meters
6 inches will push panel vibration modes over 563 Hz.
12 inches will push panel vibration modes over 282 Hz.
This is a simple model, because we use rectangular panels which have a series of panel vibration modes, not spherical panels which would have a single panel vibration mode.
Here's my take on things....
The panel resonance has Nothing to do with the wavelength of the frequency being excited. Just like vibrations in a string the panels with have resonances which depend on MASS and Stiffness(tension for stirngs).
Since these panel resonances are typically far out of the passband(unless you cross your sub extremely high or use extremely large unbracedpanels) they will not be excited to a significant degree, not by harmonics, distortion, or whatever you may imagine.
With that in mind, you may ask why brace panels? Well the answer is simple. The interior of the enclosure is a pressure chamber. Flexing in the walls with comprimise some of the pressure.
I think the panels could be modeled as passive radiators. They have a xmax(it is extremely small). They have a resonance(should be very high). They have a mechanical dampening factor. They also have a compliance(stiffness). Since we are operating FAR below resonance the so called passive radiator(panel) will be radiating out of phase with the output of the frontwave therefore it will have a canceling effect.
The output of the panels will then depend on the compliance of the panel. A panel then with higher stiffness(lowerVAS) will have lower output thus it shall cancel out the driver's hard work less and be a louder enclosure.
Now considering that if the panels' output eqauals HALF of the drivers' output it will only result in a 3db loss. It makes you think that most of that bracing is probably overkill....
Thoughts? Criticism? I look forward to response
The panel resonance has Nothing to do with the wavelength of the frequency being excited. Just like vibrations in a string the panels with have resonances which depend on MASS and Stiffness(tension for stirngs).
Since these panel resonances are typically far out of the passband(unless you cross your sub extremely high or use extremely large unbracedpanels) they will not be excited to a significant degree, not by harmonics, distortion, or whatever you may imagine.
With that in mind, you may ask why brace panels? Well the answer is simple. The interior of the enclosure is a pressure chamber. Flexing in the walls with comprimise some of the pressure.
I think the panels could be modeled as passive radiators. They have a xmax(it is extremely small). They have a resonance(should be very high). They have a mechanical dampening factor. They also have a compliance(stiffness). Since we are operating FAR below resonance the so called passive radiator(panel) will be radiating out of phase with the output of the frontwave therefore it will have a canceling effect.
The output of the panels will then depend on the compliance of the panel. A panel then with higher stiffness(lowerVAS) will have lower output thus it shall cancel out the driver's hard work less and be a louder enclosure.
Now considering that if the panels' output eqauals HALF of the drivers' output it will only result in a 3db loss. It makes you think that most of that bracing is probably overkill....
Thoughts? Criticism? I look forward to response
I agree with you on the panels acting like a big passive radiator.
Even if the xmax is 1mm overall, the panel area is quite large so you can lose alot of sound.
I agree that alot of bracing is overkill.
You'll get better decay with overkill bracing on the other hand, and that can improve the sound quality alot.
Mass of the panels is a factor too, you're right. Stiffness is also a factor.
I found an interesting article about this :
Part I :
http://www.audioholics.com/techtips/audioprinciples/loudspeakers/mechanicalnoisefloor.php
Part II :
http://www.audioholics.com/techtips/audioprinciples/loudspeakers/MechanicalNoiseLoudspeaker.php
Even if the xmax is 1mm overall, the panel area is quite large so you can lose alot of sound.
I agree that alot of bracing is overkill.
You'll get better decay with overkill bracing on the other hand, and that can improve the sound quality alot.
Mass of the panels is a factor too, you're right. Stiffness is also a factor.
I found an interesting article about this :
Part I :
http://www.audioholics.com/techtips/audioprinciples/loudspeakers/mechanicalnoisefloor.php
Part II :
http://www.audioholics.com/techtips/audioprinciples/loudspeakers/MechanicalNoiseLoudspeaker.php
BassAwdyO said:
simon5 said:
Thank you both for your posts. It definitely puts things into perspective. I'll most likely start bracing in the next couple of days, so I'll let you know what comes of it. Also, thank you for the article links.
Always curious, always learning-it's the human way.
~Aaron
And this is precisely why I thought that the displacement and interior volume would have an effect on how much bracing is required. If you have high displacement in a small volume, the pressures involved should be higher.BassAwdyO said:
I'm also at the stage of bracing my box, and it's really difficult to get a handle on how much is needed, without doing more than necessary.
"With vibration analysis (to determine flex, elasticity, rigidity, etc), you could determine the bracing needed to constrain to your individual specifications. Anyone want to volunteer to do some research??"
It's your lucky day
A few months ago I did FEA (finite element analysis) of a 16 cu ft sealed box made from 3/4" plywood, for a pair of Avalanche 18 drivers.
I was thinking along the same lines as BassAwdyO, that there's a lot of overbuilding of sub boxes for the reasons he stated.
I was hoping to satisfy myself that an unbraced box would be fine.
At xmax, the internal air prssure is 0.32 psi (as an interesting aside, this means the driver is exerting 60 lb of force). I applied this pressure load on all the panels, and calculated the panel Vd from the area times the average displacement (fortunately the s/w makes this easy).
This gives panel output of about -25 dB relative to the drivers'.
The max panel displacement was about 0.012". This is easily felt and would no doubt cause people to pronounce it poorly constructed.
I decided to crossbrace for a couple of reasons.
I wanted the first mode to be higher, IIRC it was about 140 Hz.
While well above 80 Hz, XO's are not brickwall, and even an octave above resonance the amplification factor is 2 for the normal range of damping of mechanical systems (meaning the panel deflection is double what it would be for static pressure).
"If the enclosure is vented, where does the pressure build up come from?"
It's effectively a vent only below Fb.
At Fb the vent air is rushing in and pressurizing the enclosure as the driver is trying to move in as wll, creating pressures quite a bit higher than a sealed box.
It's your lucky day
A few months ago I did FEA (finite element analysis) of a 16 cu ft sealed box made from 3/4" plywood, for a pair of Avalanche 18 drivers.
I was thinking along the same lines as BassAwdyO, that there's a lot of overbuilding of sub boxes for the reasons he stated.
I was hoping to satisfy myself that an unbraced box would be fine.
At xmax, the internal air prssure is 0.32 psi (as an interesting aside, this means the driver is exerting 60 lb of force). I applied this pressure load on all the panels, and calculated the panel Vd from the area times the average displacement (fortunately the s/w makes this easy).
This gives panel output of about -25 dB relative to the drivers'.
The max panel displacement was about 0.012". This is easily felt and would no doubt cause people to pronounce it poorly constructed.
I decided to crossbrace for a couple of reasons.
I wanted the first mode to be higher, IIRC it was about 140 Hz.
While well above 80 Hz, XO's are not brickwall, and even an octave above resonance the amplification factor is 2 for the normal range of damping of mechanical systems (meaning the panel deflection is double what it would be for static pressure).
"If the enclosure is vented, where does the pressure build up come from?"
It's effectively a vent only below Fb.
At Fb the vent air is rushing in and pressurizing the enclosure as the driver is trying to move in as wll, creating pressures quite a bit higher than a sealed box.
Freq will be increased the most with braces at midpoints of unsupported areas.
The aim is decrease further from -25 dB, say to -40 dB.
This is the panel output, which is of opposite polarity to the driver output.
Which raises another interesting point - the panel output is not distortion, it just reduces the total system output.
The aim is decrease further from -25 dB, say to -40 dB.
This is the panel output, which is of opposite polarity to the driver output.
Which raises another interesting point - the panel output is not distortion, it just reduces the total system output.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.