Thanks guys. I bought a PVC pipe today, 37mm on the inside was the best I could get. Unfortunately, and I've just realised this now, my design is a lot shorter than the original. If I want an inch spare at the end of the pipe the most it could be is around 76mm. I'm using this calculator and it recommends a ~100mm pipe, unless I go for a smaller diameter.
I am a little confused as to what I should be setting the 'Tuning Frequency' in that calculator to. The Resonant Frequency of the driver is 97Hz, should I be setting the tuning frequency to the same?
I am a little confused as to what I should be setting the 'Tuning Frequency' in that calculator to. The Resonant Frequency of the driver is 97Hz, should I be setting the tuning frequency to the same?
I've been doing a little research on ports. I'm still not sure what frequency I should be aiming for when tuning the box. Any quick help on this would be much appreciated. According to the calculator I'm using the original box is tuned to 90Hz, does that sound right?
I was also reading here about external ports. Would this be a reasonable idea, if I'm not able to get enough length without going to a really small-diameter pipe.
I can only put off my decisions about ports another couple of days. I have less than a week to finish the whole thing.
Final question is about the polyfill - this stuff is the same as what's put inside pillows, if my research is correct, right? I can just take it out of an old pillow? Or is it better to buy somewhere?
I was also reading here about external ports. Would this be a reasonable idea, if I'm not able to get enough length without going to a really small-diameter pipe.
I can only put off my decisions about ports another couple of days. I have less than a week to finish the whole thing.
Final question is about the polyfill - this stuff is the same as what's put inside pillows, if my research is correct, right? I can just take it out of an old pillow? Or is it better to buy somewhere?
Last edited:
Thanks. I don't think I can just simply convert it because - at least according to the calculator I'm using (posted a couple of posts above), slotted ports are calculated differently to non-slotted. The diagram has a slotted port, mine won't be.
According to the calculator the box is tuned to 86Hz I think, not 90Hz as I said above. Since the resonant frequency of my driver is 97Hz compared to the 95Hz of the original, I'll tune it to 88Hz (sound right?).
This means I'll be using a tube of 1.1in with a vent length of 2.7in. Hopefully I can get this size PVC pipe, otherwise I'll have to change it slightly. I'll buy another pipe tomorrow.
According to the calculator the box is tuned to 86Hz I think, not 90Hz as I said above. Since the resonant frequency of my driver is 97Hz compared to the 95Hz of the original, I'll tune it to 88Hz (sound right?).
This means I'll be using a tube of 1.1in with a vent length of 2.7in. Hopefully I can get this size PVC pipe, otherwise I'll have to change it slightly. I'll buy another pipe tomorrow.
What frequency it's supposed to be tuned to gets quite involved. There's such a thing as an "alignment" for a bass reflex cabinet. Thiele identified a couple dozen "maximally flat" alignments, and others continued the work. There are a lot of possibilities. The basic parameters are cabinet volume, taking box losses into account, the "Q" of the speaker, and the ratio of the tuning frequency to the resonant frequency of the speaker for that alignment.
This is all very interesting and worthy of study- I've put quite a few weeks into this subject recently- but my understanding is you are in that tunnel-vision deadline world of school where relevant avenues of study remain untraveled unless they affect your grades. Hence, my recommendation for trial and error based on a reasonable guess of the right starting point.
Possibly, this project may have whetted your interest to where you come back and make a deeper study of some of this fascinating field when you're not under school pressures. Good luck!
This is all very interesting and worthy of study- I've put quite a few weeks into this subject recently- but my understanding is you are in that tunnel-vision deadline world of school where relevant avenues of study remain untraveled unless they affect your grades. Hence, my recommendation for trial and error based on a reasonable guess of the right starting point.
Possibly, this project may have whetted your interest to where you come back and make a deeper study of some of this fascinating field when you're not under school pressures. Good luck!
Yeah I would love to be able to put a great amount of research in this and get something that's going to be exactly what I want, but I have only 5 more days to finish this up, so I just don't have the time to spend on it. I'd love to come back to this later on when I don't have the pressure of a deadline.
I'd rather not put a sub into this, that seems like much more effort. What tuning would you recommend?
Sorry I've realised this is a crazy question. What are the risks of tuning to around 85Hz. Why would this cause overloading? Is it better to tune a bit higher? The frequency graph
An externally hosted image should be here but it was not working when we last tested it.
is pretty flat up until about 130Hz, it seems pointless to tune higher than 100Hz.
I believe Sreten is referring to the fact that -at- the tuning frequency, the cone's motion will be at its minima, that is, barely moving at all. Virtually all the energy is coming out of the port at that point. So the cone is protected from excessive travel (excursion) and thus distortion is reduced.
But below that frequency, this type of cabinet design offers no protection to the driver. So the driver will (unless high-passed) exceed its mechanical limits as the cone excursion increases drastically, thus increasing distortion. Fortunately, at some point, the driver self-destructs.
But below that frequency, this type of cabinet design offers no protection to the driver. So the driver will (unless high-passed) exceed its mechanical limits as the cone excursion increases drastically, thus increasing distortion. Fortunately, at some point, the driver self-destructs.
Well that's good news.
I'm a little confused though. Are you saying that I should tune the box above or below the speaker's resonant frequency? And I've heard that within a box, the speaker's resonant frequency increases - the 97Hz in the specs is in free air, correct?
The resonant frequency increases relative to Fs in a sealed box. But this is vented of course. The usual starting point is the driver's Fs. Then you can adjust it up or down, but in this case, tuning to Fs is what I'd do.
And then I'd only play it softly at first, with some smoky old jazz records or chamber music (since it's not high-passed). Dub-step would not be advisable.
And then I'd only play it softly at first, with some smoky old jazz records or chamber music (since it's not high-passed). Dub-step would not be advisable.
Yes, the spec is measured in free air.
Tuning the cabinet to (pulling a "reasonable" number out of the air) 80 Hz may give you a reasonably flat response down to 75 Hz or so, BUT ...
The problem is the frequencies BELOW this 75 Hz. Most recorded (pop) music material has a lot of energy well below 75 Hz - the standard electric bass guitar goes down to 40Hz, and the most energetic part of a kick drum is perhaps in the 30 Hz to 70 Hz range. If you put such music into the speaker, there's no "back pressure" at these frequencies, and the cone will have rather large excursion, but not much sound will be emitted at these frequencies (unless you put your ear right up to the cone, but that's cheating). If you play it too loud, even though it may not SOUND "too loud," these lower frequencies could damage the driver. SLOWLY turn up the volume and watch how much the cone moves with the bass.
A sealed box won't extend the low frequency response as much as a (bass reflex) vented port, but it will have the "back pressure" at these lower frequencies, and help protect the driver.
It's possible to make a passive "crossover" or high-pass filter that will take out the frequencies below 70-80 Hz and protect the driver in a bass-reflex cabinet from these frequencies, but for such a low frequency the components are both expensive and physically large, and so it's not practical.
I think it's good advice to have this as a sealed cabinet, at least for when you're showing it off and someone else might turn up the volume. You can make the port later. This will also give you a good idea of the difference between the response of a sealed cabinet versus a bass reflex.
Ok, thanks for that. I'll take that into account when I decide what to do. I'm never going to have these very loud - they're in my living room with loads of other people. I'd like to have something I can reference for my audio engineering work, but the accuracy of the lowend is not a huge issue as my room isn't treated. Something pleasant to listen to for music, video games, movies, etc. at medium volumes is my aim.
Just out of curiosity - will boosting the lowend with a software EQ (or a baffle step compensation circuit) have the same risks?
I'm going to glue the front + rings together tonight and tomorrow, then I can glue the front on, attach the speakers, pack it with polyfill (again - what's the best source for this?), then I can decide about the port and glue the bottom on. Oiling and sanding wherever fits best.
Just out of curiosity - will boosting the lowend with a software EQ (or a baffle step compensation circuit) have the same risks?
I'm going to glue the front + rings together tonight and tomorrow, then I can glue the front on, attach the speakers, pack it with polyfill (again - what's the best source for this?), then I can decide about the port and glue the bottom on. Oiling and sanding wherever fits best.
Googling shows polyfill at Parts Express:
Acousta-Stuf Polyfill 1 lb. Bag 260-317
but the stuff at Walmart should do fine:
Mountain Mist Fiberloft Polyfill Pillow Stuffing - Walmart.com
I recall that baffle step compensation is only 6 dB (or is it 3, I forget), and shouldn't be a problem. Using a bass tone control or software EQ can boost things by 12dB or more, and at medium to high volume can be a worry for any system. The problem isn't always overexcursion of the cone, it is often the power handling (heat generated) of the voice coil.
Acousta-Stuf Polyfill 1 lb. Bag 260-317
but the stuff at Walmart should do fine:
Mountain Mist Fiberloft Polyfill Pillow Stuffing - Walmart.com
I recall that baffle step compensation is only 6 dB (or is it 3, I forget), and shouldn't be a problem. Using a bass tone control or software EQ can boost things by 12dB or more, and at medium to high volume can be a worry for any system. The problem isn't always overexcursion of the cone, it is often the power handling (heat generated) of the voice coil.
Sewing stores/craft stores. Big chains that cater to these markets (Wally World, Target (was Zellers), The Bay, Sears in this neck of the woods)
dave
Hi,
FWIW there is nothing remotely unconventional about this speaker design
and this thread is full of the conventional mistakes and misunderstandings
of those that know very little to nothing about speaker design.
For the project build them sealed and use EQ (MP3 or PC) to balance them.
They wont go very loud so tout them as very good nearfield PC speakers.
(The good drivers will cream most PC speakers for volume and bass.)
In the longer run for a usuable pair of speakers your going to need a stand
to bring them to ear level and this may as well contain some half decent
bass units, as they simply cannot do bass farfield well, at all, too small.
The stand being an entirely conventional braced box with a base plinth.
Two of these per side sealed and stuffed in parallel is a good starting point :
The Madisound Speaker Store (nothing close available anywhere near the price)
One each side face and glued/linked together for force cancelling
with resilient driver mountings is heading towards unconventional.
An unconventional first order series x/o may or may not work well, if it
does the result will be a fairly unconventional FAST, very good vs. cost.
rgds, sreten.
Sealed stuffing is not critical for a 1.1L driver in 2.5L.
FWIW there is nothing remotely unconventional about this speaker design
and this thread is full of the conventional mistakes and misunderstandings
of those that know very little to nothing about speaker design.
For the project build them sealed and use EQ (MP3 or PC) to balance them.
They wont go very loud so tout them as very good nearfield PC speakers.
(The good drivers will cream most PC speakers for volume and bass.)
In the longer run for a usuable pair of speakers your going to need a stand
to bring them to ear level and this may as well contain some half decent
bass units, as they simply cannot do bass farfield well, at all, too small.
The stand being an entirely conventional braced box with a base plinth.
Two of these per side sealed and stuffed in parallel is a good starting point :
The Madisound Speaker Store (nothing close available anywhere near the price)
One each side face and glued/linked together for force cancelling
with resilient driver mountings is heading towards unconventional.
An unconventional first order series x/o may or may not work well, if it
does the result will be a fairly unconventional FAST, very good vs. cost.
rgds, sreten.
Sealed stuffing is not critical for a 1.1L driver in 2.5L.
Last edited:
Hi Sreten,
I really want to learn all I can. If you are willing to point out the flaws, I for one would be really grateful. (I am reading and re-reading all the books I can get my hands on, but admit to being weak on the physics).
Thanks all. Ill leave the port til later and just do a sealed box for now.
I glued my front, top and back panels together. Unfortunately my front panels were pretty warped. Goddamnit. I sanded as best I could but there were still some significant gaps that had to be bridged by the glue. I just hope it'll hold.
I hooked up the speakers and put them in the box, they sounded pretty good. No vibrations in the box, which is good news. There was no bottom panel on , nor any polyfill so it's not what it'll finally sound like. I sine sweeped and it was pretty flat apart from a noticeable spike at 2-4khz. Not outrageous, but noticeable. Just using my ears, so I can't give a dB rating. Any way to alleviate this?
I glued my front, top and back panels together. Unfortunately my front panels were pretty warped. Goddamnit. I sanded as best I could but there were still some significant gaps that had to be bridged by the glue. I just hope it'll hold.
I hooked up the speakers and put them in the box, they sounded pretty good. No vibrations in the box, which is good news. There was no bottom panel on , nor any polyfill so it's not what it'll finally sound like. I sine sweeped and it was pretty flat apart from a noticeable spike at 2-4khz. Not outrageous, but noticeable. Just using my ears, so I can't give a dB rating. Any way to alleviate this?
Ok. Holy **** I'm annoyed. Went in to uni today and someone had misplaced the router bit I needed to do the lip of the speaker. Jesus ******* Christ. Because I'm an idiot I accidentally glued one of the baffles on the wrong way (DON'T say anything!), so now I have one amazingly good speaker and one speaker that rattles and vibrates and sounds like ****.
I'm gonna take them both in tomorrow and rout them with a 10mm bit. It means I'll have a huge gap around the speakers, but it's the only way I can fix this problem without spending $60 on a 24 pack of router bits when I only need the one.
Anyway, apart from that, I packed the good speaker with a bit of polyfill and the bottom is being glued on as I type. It sounds incredible. Sealed, no port, and it goes down to 70hz fairly flat. Still a spike at 2-4khz, but other than that its flat.
I'm gonna take them both in tomorrow and rout them with a 10mm bit. It means I'll have a huge gap around the speakers, but it's the only way I can fix this problem without spending $60 on a 24 pack of router bits when I only need the one.
Anyway, apart from that, I packed the good speaker with a bit of polyfill and the bottom is being glued on as I type. It sounds incredible. Sealed, no port, and it goes down to 70hz fairly flat. Still a spike at 2-4khz, but other than that its flat.
Finished it up today. Sounds pretty damn good. Thanks for everyone's help, MUCH appreciated. If anyone has any questions I'll try my best to answer.
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.