I'm still chugging away on my W371A inspired design, and I am on to my 5th design for the subwoofer section. All the previous designs worked, but as I am refining the design, I am having to come up with more and more specific solutions.
I previously built a cabinet with 2x 4" ports. It works great, but I decided I need to shrink the cabinet as much as possible. This cabinet pictured is 18"x18"x48", but I want to shrink it down to 16"x 16". This will constrict the vertical area where the ports are currently installed, so I'm convinced this port design won't work right when I shrink it down.
So I experimented by building a radius port from some scrap wood and aluminum sheet. It is 8"x2.75"x30". It has a similar cross sectional area as the dual ports, but with the advantage that it exits straight out the back of the cabinet, eliminating the funky port exhaust chamber in the back of the old design.
This first test was quick and dirty as you can see. It does work quite well. There is some noticeable resonance, but likely due to the unsupported aluminum sheet. I also included a turning vane to help ensure as much of the air is traveling the same distance. (Obviously the air on the inside of the curve will travel slower and take a shorter path. This causes turbulence in pipes.) I don't know if this helps or hurts.
Next step is making the fiberglass version. To make it, I used my router to machine molds out of MDF. I made each from 3 slices, glued them together, sealed, primed, cleared, etc.
And the finished port next to the prototype. (Yes, I hot glued the port into the test cabinet. It worked. And I did add a couple screws afterward....)
I'm sure I made this a bit too good, but I am using it as an excuse to test materials and processes I will need for other projects. So it is overkill, but it is worth doing to me. Mistakes were made, but I got there in the end!
And data! These show both subs individually, and a comparison of the old and new designs. Measurement mic was placed in front of the port, so the low frequency peak is the port resonance. The actual in room response is much flatter than this looks.
Comparison between aluminum tubes and fiberglass port with no EQs or filters:
Measurements SPL and distortion for the aluminum ports with EQs and filters applied:
Measurements of SPL and distortion for the fiberglass port with EQs and filters applied:
(The tiny dip at ~23hz corresponds to the fiberglass port wanting to vibrate. It is largely unsupported right now, so a little bracing should fix that issue.)
That's good enough for me! I'm happy that the fiberglass port is giving me almost the same results as the dual port design, and it fixes an issue at 100hz. Best of all it will let me build my speakers without having to sacrifice by using a smaller port.
I previously built a cabinet with 2x 4" ports. It works great, but I decided I need to shrink the cabinet as much as possible. This cabinet pictured is 18"x18"x48", but I want to shrink it down to 16"x 16". This will constrict the vertical area where the ports are currently installed, so I'm convinced this port design won't work right when I shrink it down.
So I experimented by building a radius port from some scrap wood and aluminum sheet. It is 8"x2.75"x30". It has a similar cross sectional area as the dual ports, but with the advantage that it exits straight out the back of the cabinet, eliminating the funky port exhaust chamber in the back of the old design.
This first test was quick and dirty as you can see. It does work quite well. There is some noticeable resonance, but likely due to the unsupported aluminum sheet. I also included a turning vane to help ensure as much of the air is traveling the same distance. (Obviously the air on the inside of the curve will travel slower and take a shorter path. This causes turbulence in pipes.) I don't know if this helps or hurts.
Next step is making the fiberglass version. To make it, I used my router to machine molds out of MDF. I made each from 3 slices, glued them together, sealed, primed, cleared, etc.
And the finished port next to the prototype. (Yes, I hot glued the port into the test cabinet. It worked. And I did add a couple screws afterward....)
I'm sure I made this a bit too good, but I am using it as an excuse to test materials and processes I will need for other projects. So it is overkill, but it is worth doing to me. Mistakes were made, but I got there in the end!
And data! These show both subs individually, and a comparison of the old and new designs. Measurement mic was placed in front of the port, so the low frequency peak is the port resonance. The actual in room response is much flatter than this looks.
Comparison between aluminum tubes and fiberglass port with no EQs or filters:
Measurements SPL and distortion for the aluminum ports with EQs and filters applied:
Measurements of SPL and distortion for the fiberglass port with EQs and filters applied:
(The tiny dip at ~23hz corresponds to the fiberglass port wanting to vibrate. It is largely unsupported right now, so a little bracing should fix that issue.)
That's good enough for me! I'm happy that the fiberglass port is giving me almost the same results as the dual port design, and it fixes an issue at 100hz. Best of all it will let me build my speakers without having to sacrifice by using a smaller port.
I love creative design and use of novel materials or techniques to achieve an outcome. My first fiberglassing project was a car audio sub that needed custom moulding to fit my full size driver with only 2mm of spare space. I used boat building TAFE lessons to teach myself the correct technique for the fiberglassing as so many of the Car Audio people were doing it very badly.
Hello Fredy;
That is amazing work. I assume it was vacuum bagged? If so will you share some details? I have never vacced a complex shape before.
Looking at the first picture I am not quite sure what I am looking at. Maybe I cant see some passages.
Thank you Fredy.
Barry.
That is amazing work. I assume it was vacuum bagged? If so will you share some details? I have never vacced a complex shape before.
Looking at the first picture I am not quite sure what I am looking at. Maybe I cant see some passages.
Thank you Fredy.
Barry.
That first photo is the dry fitting of the design I am comparing the fiberglass port against. The speaker has 2- 12" front facing drivers (sealed), and 1- 10" rear subwoofer (ported). I set up this photo so you can see the 2- 4" aluminum ports running vertically, on either side of the driver cut-out. The big oval cut-out is a combined port exit...the ports exit under the rear facing driver.
2 more photos:
And yes, I did vacuum bag the port. It is actually pretty simple to do. The trick is getting the bag big enough, and you can do something like lay a string over the part, following the contours, and the bag should be slightly longer than that. And I applied the tap to the bag first, and then stuck it down to the mold, folding pleats as needed.
I'm using wax and PVA for a release agent.
Yes, the molds are re-usable.
Last edited:
I'm looking at the one underway behind the bagged one under vacuum... Is that a polyethylene sheet to act as a separator from the mould or some other step in the process?
It looks like you are using fabric to reinforce your resin and I'm unsure if you are doing somehting different to normal fibreglassing? I have never bagged my parts for this type of activity yet my work in lab producing resin and composite parts we always use vacuum to enhance the result.
Chipboard parts on this are cost saving for prototyping stage?
What software do you use to simulate your designs? I'm needing a lot of software support to overcome my lack of recent experience with the nuance of technical behaviour.
It looks like you are using fabric to reinforce your resin and I'm unsure if you are doing somehting different to normal fibreglassing? I have never bagged my parts for this type of activity yet my work in lab producing resin and composite parts we always use vacuum to enhance the result.
Chipboard parts on this are cost saving for prototyping stage?
What software do you use to simulate your designs? I'm needing a lot of software support to overcome my lack of recent experience with the nuance of technical behaviour.
Mgrobins:
Yes, that is polyethylene sheet over the other mold. I covered it up while spraying the PVA release coating.
I'm using a plain weave fiberglass fabric with epoxy resin. The white perforated material is a core material that adds stiffness without adding much weight. I used 2 layers of 8oz plain weave fiberglass, then the core, and then 2 more layers. And I added a couple extra layers on either end for increased rigidity. I think it is pretty "normal". I didn't have to vacuum bag it. Vacuum bagging mostly improves the fiber to resin ratio, but in this case I just wanted to be sure everything stayed where I put it.
When you ask about the chipboard parts--do you mean the prototype port with aluminum sheet? I was just burning up scrap material I had laying around.
Software. Do you mean speaker design software? That is VituixCAD. I'm using it primarily for the cabinet design feature. My speaker uses active crossovers, so I'm not building crossovers.
Yes, that is polyethylene sheet over the other mold. I covered it up while spraying the PVA release coating.
I'm using a plain weave fiberglass fabric with epoxy resin. The white perforated material is a core material that adds stiffness without adding much weight. I used 2 layers of 8oz plain weave fiberglass, then the core, and then 2 more layers. And I added a couple extra layers on either end for increased rigidity. I think it is pretty "normal". I didn't have to vacuum bag it. Vacuum bagging mostly improves the fiber to resin ratio, but in this case I just wanted to be sure everything stayed where I put it.
When you ask about the chipboard parts--do you mean the prototype port with aluminum sheet? I was just burning up scrap material I had laying around.
Software. Do you mean speaker design software? That is VituixCAD. I'm using it primarily for the cabinet design feature. My speaker uses active crossovers, so I'm not building crossovers.
The cabinets being made from particle board is to keep cost down, yes. I just realized that must be what you meant. I've built a few different variations of cabinet designs with different dimensions and port designs. I do plan to make the real thing a little fancier!
It does seem normal process with the glass ... it just looked a bit different to my own materials I guess 🙂.
Indeed yes I mean the cabinet design software 🙂. What is VituixCAD? A cabinet design feature sounds helpful... worthy of me looking at or does it require prior experience and a lot of learning to get to grips with? I planned to use Sketchup or perhaps Fusion 360 as I need to learn both for 3d printing I do.
I have created my subwoofer design in WinISD, checked it's resonances in Boxnotes and used a port calculator to confirm my rectangular port is ok (Im told WinISD Pro is good for circle ports and makes an error on rectangle ones).
I now need to create the physical design for manufacture and had thought to use Sketchup ... yet also like to know what others more experienced are using!
I would plan to have all of the major cuts done on CNC so I can collect the parts in my small car and then do the assembly at home. I intend to use some beautiful burl veneer, brass rod inlay and perhaps a full Aluminium milled front plate with engraving that I can paint and finish (I'm an airbrush artist).
They will be pretty big so I figure they should look good haha.
I don't have any prior experience in speaker design. VituixCAD was very intuitive for me to use. I liked it better than WinISD. My CAD is Rhino3D, but it doesn't matter what you use.
So far it seems that subwoofer design is fairly forgiving. I wouldn't worry about it too much. But slot ports are a little funny. If a rectangular port is narrower than 8:1, standard programs have trouble with them. But a rectangular or slot port with a ratio of 1:1 to 8:1 is going to be just fine in any program.
Make a cheap prototype and listen to it for awhile. If you want a wide slot port, build it in a way that allows you to adjust it.
So far it seems that subwoofer design is fairly forgiving. I wouldn't worry about it too much. But slot ports are a little funny. If a rectangular port is narrower than 8:1, standard programs have trouble with them. But a rectangular or slot port with a ratio of 1:1 to 8:1 is going to be just fine in any program.
Make a cheap prototype and listen to it for awhile. If you want a wide slot port, build it in a way that allows you to adjust it.
THankfully I have some prior prototyping for tyhis design to rely on and will build it with a little more confidence than a total fresh build... ack that minor changes can sometimes have more than minor outcomes.
I shall look at Vituix cad.
I'm kind of surprised there is no full suite that does initial modelling, actual CAD design using a specific toolset for how speakers are commonly constructed, and then simulation tools to examine that speakers behaviour.
I shall look at Vituix cad.
I'm kind of surprised there is no full suite that does initial modelling, actual CAD design using a specific toolset for how speakers are commonly constructed, and then simulation tools to examine that speakers behaviour.