with your mastery of wood working i think you can do so with wood.
go half way down this page....
http://www.vandersteen.com/pages/M5lit.html
nest to the title "construction" you will see a cabinet made of 22 layers of MDF
what if each layer was progressively smaller (front to back) forming a taper....
just a thought.
same concpept as your woofer box but the MDF layers are aligned front to back instead of bottom to top.
Love
navin
go half way down this page....
http://www.vandersteen.com/pages/M5lit.html
nest to the title "construction" you will see a cabinet made of 22 layers of MDF
what if each layer was progressively smaller (front to back) forming a taper....
just a thought.
same concpept as your woofer box but the MDF layers are aligned front to back instead of bottom to top.
Love
navin
http://www.vandersteen.com/pages/Pdffiles/RefReview.pdf
another interesting and slightly less related link
another interesting and slightly less related link
Hi Rob
I'm interested to see how you get on with your project, as you're doing quite a few things I'm planning.
At this stage I'm just planning to build a set of bass cabinets, using the same SS drivers you're using (already ordered) but only one per cabinet. The cabinet will be a standard rectangular cabinet from MDF, as I figure that the acoustic benefits (ie reduction in standing waves) from the great shape you're using might not apply for the XO frequency of 300 HZ that I'm planning. To keep the size down I'm planning to use a cabinet volume of around 50l (sealed), which I calculate will give a Qt of 0.8. To get a Qt of 0.7 means about 70 litres in volume (an extra 40%), and I'm hoping that there won't be too much difference. Any opinions on that point?
The bass speakers will be powered by an AKSA 100W amp (kit amp from Melbourne, Australia) - already built. The mid/high will be my current AE1 speakers, powered by a cyrus 3 and PSX-R.
The active XO will be the ESP 12 dB/Oct L-R, which I'm right in the middle of building at the moment.
I have been thinking for the next step of building a new set of mid/high speakers, with a cabinet like the one that you're building from stacked MDF profiles. I was thinking of cutting them out with a jig saw, and as well as the elliptical type curve, having a gradual taper in the cabinet width from the bottom to the top. I'd do this by jigsawing on a slight angle, but I'm not sure what kind of shape might result. I need to play around on the 3D cad to figure that one out.
Did you get the MDF laser cut? I've only just heard that laser cutting wood/mdf is possible, and it might be a good way of making baffle cutouts for those without jig saws/routers.
One of the other things I'll need to do is to separate pre-power section of the Cyrus 3, so that I can feed out from the pre to the active XO and then back to the power amp part. Have you done such a thing on a Cyrus? Any suggestions here?
Keep up the good work- and the discussion too if you have time.
Mick
I'm interested to see how you get on with your project, as you're doing quite a few things I'm planning.
At this stage I'm just planning to build a set of bass cabinets, using the same SS drivers you're using (already ordered) but only one per cabinet. The cabinet will be a standard rectangular cabinet from MDF, as I figure that the acoustic benefits (ie reduction in standing waves) from the great shape you're using might not apply for the XO frequency of 300 HZ that I'm planning. To keep the size down I'm planning to use a cabinet volume of around 50l (sealed), which I calculate will give a Qt of 0.8. To get a Qt of 0.7 means about 70 litres in volume (an extra 40%), and I'm hoping that there won't be too much difference. Any opinions on that point?
The bass speakers will be powered by an AKSA 100W amp (kit amp from Melbourne, Australia) - already built. The mid/high will be my current AE1 speakers, powered by a cyrus 3 and PSX-R.
The active XO will be the ESP 12 dB/Oct L-R, which I'm right in the middle of building at the moment.
I have been thinking for the next step of building a new set of mid/high speakers, with a cabinet like the one that you're building from stacked MDF profiles. I was thinking of cutting them out with a jig saw, and as well as the elliptical type curve, having a gradual taper in the cabinet width from the bottom to the top. I'd do this by jigsawing on a slight angle, but I'm not sure what kind of shape might result. I need to play around on the 3D cad to figure that one out.
Did you get the MDF laser cut? I've only just heard that laser cutting wood/mdf is possible, and it might be a good way of making baffle cutouts for those without jig saws/routers.
One of the other things I'll need to do is to separate pre-power section of the Cyrus 3, so that I can feed out from the pre to the active XO and then back to the power amp part. Have you done such a thing on a Cyrus? Any suggestions here?
Keep up the good work- and the discussion too if you have time.
Mick
i too have built similar cabinets but using 2 12" Audio Concepts SV12 per cabinet. The cabinets net internal volume in my case is 120 liters. The Qts works out to 0.8.
While there is not much audible difference between a Q of 0.8 and 0.6 (I wanted but could not get as WAF interfered). There are still ways to reduce the effective Q of your system.
1. mass load the woofer. You can do this by applying blu tak or silly putty to the dust cap
2. adding an aperiodic vent
3. facing the woofer to a wall (very close to the wall).
I did the last. hope this helps.
regasds
navin
While there is not much audible difference between a Q of 0.8 and 0.6 (I wanted but could not get as WAF interfered). There are still ways to reduce the effective Q of your system.
1. mass load the woofer. You can do this by applying blu tak or silly putty to the dust cap
2. adding an aperiodic vent
3. facing the woofer to a wall (very close to the wall).
I did the last. hope this helps.
regasds
navin
Hi Mick, and Navin.
Thanks for the interest.
"To get a Qt of 0.7 means about 70 litres in volume (an extra 40%), and I'm hoping that there won't be too much difference. Any opinions on that point? "
If you have some spare mdf you could build the larger cab and fill it with bricks to lessen the volume, and see what you think of the change in the sound. You can then select a volume you're happy with. An extra 20L would not be that bigger if you increase height, width and depth a bit - My cabs are about 840mm deep, but you wont notice the depth as the fronts will be where my current speakers are anyway. - as it's sealed the positioning should be more forgiving.
Yes I did laser cut my cabs - luckily it's part of my job to be able to use a laser, and I have a rough working knowledge of the cad system we use - enough to plod along and get a shape drawn anyway - I usually hassle my mate at work if I get stuck.
The laser cuts to at least 0.1mm precision and is a great way of getting accurate shapes cut. If you do diminishing layers, an option I was considering was to square cut the edges, then fill the 'steps' with automotive body filler to smooth the shape over. If you've got cad you could draw each layer and work out the angle of the cut for the jigsaw.
I did remove the vol. pot from my jvc amp, but Planet10 (Dave) suggested it may be nice to leave out the output buffers on the x-over, and use the vol. pots on my cyrus 1's to attenuate the signal. If you follow the signal path from the inputs it should go to the source selecter, then the vol. pot/pre then into the amp section. (it did in the jvc,anyhow) you're doing something slightly different, but I'm sure if you post the question on its own thread someone on the forum will know - between them they've probably got everything covered!🙂
Cheers
Rob
Thanks for the interest.
"To get a Qt of 0.7 means about 70 litres in volume (an extra 40%), and I'm hoping that there won't be too much difference. Any opinions on that point? "
If you have some spare mdf you could build the larger cab and fill it with bricks to lessen the volume, and see what you think of the change in the sound. You can then select a volume you're happy with. An extra 20L would not be that bigger if you increase height, width and depth a bit - My cabs are about 840mm deep, but you wont notice the depth as the fronts will be where my current speakers are anyway. - as it's sealed the positioning should be more forgiving.
Yes I did laser cut my cabs - luckily it's part of my job to be able to use a laser, and I have a rough working knowledge of the cad system we use - enough to plod along and get a shape drawn anyway - I usually hassle my mate at work if I get stuck.
The laser cuts to at least 0.1mm precision and is a great way of getting accurate shapes cut. If you do diminishing layers, an option I was considering was to square cut the edges, then fill the 'steps' with automotive body filler to smooth the shape over. If you've got cad you could draw each layer and work out the angle of the cut for the jigsaw.
I did remove the vol. pot from my jvc amp, but Planet10 (Dave) suggested it may be nice to leave out the output buffers on the x-over, and use the vol. pots on my cyrus 1's to attenuate the signal. If you follow the signal path from the inputs it should go to the source selecter, then the vol. pot/pre then into the amp section. (it did in the jvc,anyhow) you're doing something slightly different, but I'm sure if you post the question on its own thread someone on the forum will know - between them they've probably got everything covered!🙂
Cheers
Rob
Navin,
I checked the links out - very interesting with the layers thing🙂 - My problem with the mid enclosure is the volume 7L. If I taper it the length gets bigger and bigger! I'm going for parrallel sides, eye shaped enclosure for mid, and poss. the same but smaller for the tweet. The mid is drawn up, and I'll get on the tweet on monday.
I may change the inside diameter of the cutouts on the mid, so that the sides are not parrallel inside the enclosure. I'm not sure whether lots of steps inside the enclosure would be a help or a hindrance😕
Cheers
Rob
I checked the links out - very interesting with the layers thing🙂 - My problem with the mid enclosure is the volume 7L. If I taper it the length gets bigger and bigger! I'm going for parrallel sides, eye shaped enclosure for mid, and poss. the same but smaller for the tweet. The mid is drawn up, and I'll get on the tweet on monday.
I may change the inside diameter of the cutouts on the mid, so that the sides are not parrallel inside the enclosure. I'm not sure whether lots of steps inside the enclosure would be a help or a hindrance😕
Cheers
Rob
Q and cabinet size
Thanks Rob and Navin for your thoughts
"An extra 20L would not be that bigger if you increase height, width and depth a bit"
I don't really want to increase the planned cabinet height, as I want to park the AE1's on the top of my bass cabinets, and I'd like to leave the width attractively narrow (am I being a victim of fashion here?😎 ) and so that just leaves depth, which is already 500mm. Another option for changing the Q and lowering the fc is to use a Linkwitz transform (see the ESP site), which would be fairly easy to do with my active XO, with only one more Op amp in the chain, using a board from ESP.
I'd like to try experimenting with cabinet size, but given limited time, I think I'll have to decide on one and go for it.
Its hard to know what effect steps caused by stepped MDF construction would have on the sound. Because the steps would only be say 20mm wide and say 5mm, deep, you would expect only the higher frequencies to be potentially affected, as lower frequency waves would not be affected by surface irregularities of that size. If your Xo frequency were say 2500Hz, then the wavelength at this frequency would be 150mm. I can't imagine that an irregularity of 20mm x 5mm would make too much difference to the reflections of a wave of that length. Assuming that the tweeter is acoustically sealed from the enclosure, then it does not have a rear wave. As you suggest Rob, you could always body fill the interior smooth, or even use a damping compound. Doing the outside might be a bit more work.
Mick
Thanks Rob and Navin for your thoughts
"An extra 20L would not be that bigger if you increase height, width and depth a bit"
I don't really want to increase the planned cabinet height, as I want to park the AE1's on the top of my bass cabinets, and I'd like to leave the width attractively narrow (am I being a victim of fashion here?😎 ) and so that just leaves depth, which is already 500mm. Another option for changing the Q and lowering the fc is to use a Linkwitz transform (see the ESP site), which would be fairly easy to do with my active XO, with only one more Op amp in the chain, using a board from ESP.
I'd like to try experimenting with cabinet size, but given limited time, I think I'll have to decide on one and go for it.
Its hard to know what effect steps caused by stepped MDF construction would have on the sound. Because the steps would only be say 20mm wide and say 5mm, deep, you would expect only the higher frequencies to be potentially affected, as lower frequency waves would not be affected by surface irregularities of that size. If your Xo frequency were say 2500Hz, then the wavelength at this frequency would be 150mm. I can't imagine that an irregularity of 20mm x 5mm would make too much difference to the reflections of a wave of that length. Assuming that the tweeter is acoustically sealed from the enclosure, then it does not have a rear wave. As you suggest Rob, you could always body fill the interior smooth, or even use a damping compound. Doing the outside might be a bit more work.
Mick
Re: Q and cabinet size
You could always make the little box aperiodic and lower the Q a bit that way.
dave
Kanga said:
You could always make the little box aperiodic and lower the Q a bit that way.
dave
RobWells said:
An irregular interior should help... for a mid non-parallel sides are important. A folded terminated TL works well (and makes a box that is regular on the outside irregular on the inside)
dave
laser cutting - that is rich! I use a carpenters saw and a hammer to solve all my problems. I also have a drill but rarely use that.
on another topic what di yo think of the vandersteen links as far as building a tapered box for teh midrange/tweeter is concerned.
one way I am considering is maing a template that follows the shape of a mid and tweeter aligned one above the other. Each layer as we get to the back will be smaller and hence theyre will be a taper till we get to the back of the box.
this is for another speaker project. Not yet chosen the drivers or I might rewirk my SS 8546-9900 into this later.
what do you guys think?
on another topic what di yo think of the vandersteen links as far as building a tapered box for teh midrange/tweeter is concerned.
one way I am considering is maing a template that follows the shape of a mid and tweeter aligned one above the other. Each layer as we get to the back will be smaller and hence theyre will be a taper till we get to the back of the box.
this is for another speaker project. Not yet chosen the drivers or I might rewirk my SS 8546-9900 into this later.
what do you guys think?
Hi again,chaps.
Navin - I think the enclosure you are suggesting would work well, and shouldn't be too hard to make with a set of templates - draw them up in cad, and use psp etc to scale them to the stepped sizes. Print and cut out of thick card/or wood and you're there. I'd personally prefer a router to a jigsaw. They seem more accurate.
Dave - I do like the idea of the folded TL, I never even considered that. Would it still work well if inside the enclosure stepped every 18mm rather than a smooth transition. Reason I ask is I was playing on the cad today, and it looks possible doing the layers thing! The tube would be an elipse shape (how do I find the accurate area of an elipse - my cad at work won't do it) I could try using the router to make the transition as smooth as poss.
I e-mailed some cad shots to my home pc today, but the attatchment didnt attatch. I'll try again tomorrow.
Got the first of my 2 cyrus1's delivered today -at last😀
Cheers
Rob
Navin - I think the enclosure you are suggesting would work well, and shouldn't be too hard to make with a set of templates - draw them up in cad, and use psp etc to scale them to the stepped sizes. Print and cut out of thick card/or wood and you're there. I'd personally prefer a router to a jigsaw. They seem more accurate.
Dave - I do like the idea of the folded TL, I never even considered that. Would it still work well if inside the enclosure stepped every 18mm rather than a smooth transition. Reason I ask is I was playing on the cad today, and it looks possible doing the layers thing! The tube would be an elipse shape (how do I find the accurate area of an elipse - my cad at work won't do it) I could try using the router to make the transition as smooth as poss.
I e-mailed some cad shots to my home pc today, but the attatchment didnt attatch. I'll try again tomorrow.
Got the first of my 2 cyrus1's delivered today -at last😀
Cheers
Rob
RobWells said:
18 mm steps shouldn't be much of a problem except perhaps right behind the driver. The line doesn't have to have a terminus. Take the volume you would need to get a low Q box (say 0.5). Make it long and then put a partition in it that divides the box into a folded line with a 3 or 4:1 start/end ratio. Then damp it with progressively denser stuffing as you approach the end.
The elliptical shape will make it interesting, but the idea is to give the backwave as much chance as possible to get absorbed and not reflected back thru the cone.
dave
Hi Dave,
Unless winISD is playing me up, a box with my mid and a Q of 0.5 is about 90L
It will do 0.6 at around 10L so I'll go for this size
If I used a cylinder as the main box, and put in the divider I'd end up with a half moon tapered line as such - should sound ok!
I'll draw it all as a circle, then make it into an elipse which will increase the vol. by an unknown amount, any will help as it will never be too big.
Cheers
Rob
Unless winISD is playing me up, a box with my mid and a Q of 0.5 is about 90L
It will do 0.6 at around 10L so I'll go for this size
If I used a cylinder as the main box, and put in the divider I'd end up with a half moon tapered line as such - should sound ok!
I'll draw it all as a circle, then make it into an elipse which will increase the vol. by an unknown amount, any will help as it will never be too big.
Cheers
Rob
rob wells wrote...
"Unless winISD is playing me up, a box with my mid and a Q of 0.5 is about 90L
It will do 0.6 at around 10L so I'll go for this size"
Qts 0.5 = 90 L and Qts 0.6 = 10L
If that is what winISD is saying something is wrong.
I like the tapered TL too. How long should the line be for a 8" with a Fs of 40hz, Qts 0.4 and Vas 40 L (this is a hypothetical 8" but it is as realastic a hythetical driver I cna think of).
Regards
navin
"Unless winISD is playing me up, a box with my mid and a Q of 0.5 is about 90L
It will do 0.6 at around 10L so I'll go for this size"
Qts 0.5 = 90 L and Qts 0.6 = 10L
If that is what winISD is saying something is wrong.
I like the tapered TL too. How long should the line be for a 8" with a Fs of 40hz, Qts 0.4 and Vas 40 L (this is a hypothetical 8" but it is as realastic a hythetical driver I cna think of).
Regards
navin
navin said:
8" is a pretty big midrange. The idea with the closed line is that the waveform passes thru at least a wavelength before it gets back to the cone -- so ideally the length is 1/2 a wl of the lowest frequency of interest. The progressive damping and the taper allow you to get away with somewhat less. So i guess one could formulate a rule-of-thumb of 1/2 wl at the XO frequency. One could do some reverse engineering on the B&W Nautilus and get an idea of how long they figured they could get away with.
I speculate that if you open it up at the end and make it aperiodic you might get way with even less (or greater effectiveness).
Both can probably be modeled in Martin King's simulator.
dave
Hi again,
Navin with the win isd figs, the Qtc doesn't change much above 10L. I assume it's designed to go in a small box. At 2.6L Q=0.9, at 6L Q=0.7 at 11.8L Q=0.6. After this, increasing box size doesn't make a lot of difference.
Dave, I've been mocking up a 10L(+ a bit more) cab, and was wondering whether the taper can go down to nothing, and end about 40mm behind the baffle. I'll try and attach a drawing which will show the cab side on, with the front view next to it, and a side view showing the layers. The actual bit of ply that will divide will be stepped from layer to layer.
Cheers
Rob
Audiolab fell through - curses to e-bayers advertising what they don't have..😡
Navin with the win isd figs, the Qtc doesn't change much above 10L. I assume it's designed to go in a small box. At 2.6L Q=0.9, at 6L Q=0.7 at 11.8L Q=0.6. After this, increasing box size doesn't make a lot of difference.
Dave, I've been mocking up a 10L(+ a bit more) cab, and was wondering whether the taper can go down to nothing, and end about 40mm behind the baffle. I'll try and attach a drawing which will show the cab side on, with the front view next to it, and a side view showing the layers. The actual bit of ply that will divide will be stepped from layer to layer.
Cheers
Rob
Audiolab fell through - curses to e-bayers advertising what they don't have..😡
RobWells said:
I don't know about this ap, but in a voigt tapering to nothing is not a good idea. I'd feel better if you could give it at least a little area at the end -- even if you have to shorten the bottom bit a bit.
dave
- Status
- Not open for further replies.