thanks guys!
Honestly - I simply can't answer this question for the H-sub as I only built it last evening to show the "how to" of making a simple swing for these.
By the way, given the formula I missed to outline in
Audio and Loudspeaker Design Guide Lines
f = 1 / ((2 * PI) * SQRT (L / g))
the pendulum frequency of a 15" H-sub (L ~ 0.4 meter) results in roughly
1 / (( 2 * 3.14 ) * SQRT ( 0.4 / 9.81)) = 0.78Hz
So, this ~ 1 Hz resonance frequency of the swing is more than a decade below the deepest audio frequency we want to reproduce with that sub - should do fine IMO.
Remember the frequency depends *only* on the length of the swing.
Michael
Honestly - I simply can't answer this question for the H-sub as I only built it last evening to show the "how to" of making a simple swing for these.
By the way, given the formula I missed to outline in
Audio and Loudspeaker Design Guide Lines
f = 1 / ((2 * PI) * SQRT (L / g))
the pendulum frequency of a 15" H-sub (L ~ 0.4 meter) results in roughly
1 / (( 2 * 3.14 ) * SQRT ( 0.4 / 9.81)) = 0.78Hz
So, this ~ 1 Hz resonance frequency of the swing is more than a decade below the deepest audio frequency we want to reproduce with that sub - should do fine IMO.
Remember the frequency depends *only* on the length of the swing.
Michael
Last edited:
Hi Michael, Stig,
I am really interested in the work you are sharing with us in this thread, so please keep it up!
Michael, with respect to the H-baffle hanging doesn't the presence of the H-baffle negate the benefit of the swing? As I understood your explanation, it was the absence of the tightly coupled mass of the baffle or cabinet, which provided the benefits in performance. If you are swinging the H-baffle aren't you just swinging all those reaction forces as well?
I may have misunderstood what you said in your explanation several pages above, so I would appreciate any clarification you can provide.
Ciao!
I am really interested in the work you are sharing with us in this thread, so please keep it up!
Michael, with respect to the H-baffle hanging doesn't the presence of the H-baffle negate the benefit of the swing? As I understood your explanation, it was the absence of the tightly coupled mass of the baffle or cabinet, which provided the benefits in performance. If you are swinging the H-baffle aren't you just swinging all those reaction forces as well?
I may have misunderstood what you said in your explanation several pages above, so I would appreciate any clarification you can provide.
Ciao!
No you haven't misunderstood - you are absolutely correct.
On the other hand I try to support those who want to experiment as good as I can – not knowing yet where the benefits of swings start or end *in praxis*.
I'm happy Stig Erik having been so thoroughgoing to drop baffles completely with good results – as I possibly wouldn't have had the courage to do so soon.
This way I just trashed my (already small) open baffles immediately.
The trade in - in my case - is a severe loss of bandwidth (lower XO has to go up ~ 100 - 200 Hz to stay within the same IM limit I defined for design) but I found it well worth the price for the sonic improvement.
But back to what you wanted clarification about.
Yes the *ultimate best* would be if we could separate "local mass" (speaker motor mostly) from all "dislocated mass" (any baffles or boxes attached to the speaker).
*In praxis* this ultimate best can be done with nude speakers on a swing only – any other attempt is a compromise in one way or another.
To what extent those compromises are still beneficial in the context I described – I really don't know – hope time (meaning all brave "evolutionarios & revolutionarios"
out there) will tell ...Michael
Last edited:
They look to be one of those screw type cams that are used with "some assembly required" furniture such as a bookcase or desk. Usually available at the local hardware store.
This is great work you guys are doing - I'm loving this thread and all of the "out of the box"
thinking as well as the hard work (or should I say "playing") that is showing great results!! Verrrrrry Nice!! Thank You!! 
Ok, my turn. Here are the much more conservative feet (instead of fully suspended swing) :
I cut some cork pads in slightly different sizes and stacked them in 3 layers. They seemed quite stiff to me. So I put 2 of them more or less midway in the fore-aft direction -- to be a pivot (of this seesaw). But this alone was too unstable, so I inserted 4 more soft foam pads at corners. Then the whole thing became a big metronome, slightly "flipping" back and forth 3~4 times when being pushed. Its 'flipping' frequency might be about 1.x Hz by roughly estimate. The whole lot of mess (mass) on top made it move slowly.
Similar support method was also applied to the subwoofer baffle (at the center).
Although the sub baffle is very short (low), it's also very shallow in the fore-aft direction, so its profile of side-view is still relatively tall. The cork pivot plus soft foam feet made it move softly and slowly just like the main channels.
And it turned out.... well, seemed not as good as all-soft-feet support.
I confess, I was not very patient, my evaluation might be not thorough enough. I just felt the vibration on baffles was slightly larger than that of all-soft-feet support. Then I pulled out the cork pivot to compare and had an intereting observation.
When supported by the cork pivot, the lower half of the baffle vibrated less, while it moved more on the top (both in the fore-aft direction). Like a metronome - a vertial swinging lever - acts in a very very short excursion.
Without the cork pivot, the lower half of the baffle moved more, while much less on top. It seemed it became sort of a virtual pendulum, hung by a virtual pivot (maybe formed by the inertia mass) somewhere over the top of it. But of course it moved under lots of friction, far from the freedom of a real pendulum.
So the soft foam pads are bearing shear stress, thus the whole lot of 'payload' can twist its bottom. And the existence of cork pivot stops such movement. All in all, things are getting more and more complex.... (The situations of the vibrations in the 2 conditions mentioned above would change under different frequncies. Those very low end notes, say, under 30Hz, if strong enough, would excite some larger motions which I couldn't tell it's calmer or not on top or bottom... )
The positions, strengths and frequency of the source of excitation, the inertia of the whole system (and the distribution), the positions of real and virtual pivot(s), the dampings, even the air load by the baffle.... etc. all come into play. It's so hard to pinpoint their individual actions and thier interactions.
I'm lost
Oh well, before I build some real swings, I think I'll just leave them there with the soft pads....
An externally hosted image should be here but it was not working when we last tested it.
I cut some cork pads in slightly different sizes and stacked them in 3 layers. They seemed quite stiff to me. So I put 2 of them more or less midway in the fore-aft direction -- to be a pivot (of this seesaw). But this alone was too unstable, so I inserted 4 more soft foam pads at corners. Then the whole thing became a big metronome, slightly "flipping" back and forth 3~4 times when being pushed. Its 'flipping' frequency might be about 1.x Hz by roughly estimate. The whole lot of mess (mass) on top made it move slowly.
An externally hosted image should be here but it was not working when we last tested it.
Similar support method was also applied to the subwoofer baffle (at the center).
An externally hosted image should be here but it was not working when we last tested it.
Although the sub baffle is very short (low), it's also very shallow in the fore-aft direction, so its profile of side-view is still relatively tall. The cork pivot plus soft foam feet made it move softly and slowly just like the main channels.
And it turned out.... well, seemed not as good as all-soft-feet support.
I confess, I was not very patient, my evaluation might be not thorough enough. I just felt the vibration on baffles was slightly larger than that of all-soft-feet support. Then I pulled out the cork pivot to compare and had an intereting observation.
When supported by the cork pivot, the lower half of the baffle vibrated less, while it moved more on the top (both in the fore-aft direction). Like a metronome - a vertial swinging lever - acts in a very very short excursion.
Without the cork pivot, the lower half of the baffle moved more, while much less on top. It seemed it became sort of a virtual pendulum, hung by a virtual pivot (maybe formed by the inertia mass) somewhere over the top of it. But of course it moved under lots of friction, far from the freedom of a real pendulum.
So the soft foam pads are bearing shear stress, thus the whole lot of 'payload' can twist its bottom. And the existence of cork pivot stops such movement. All in all, things are getting more and more complex.... (The situations of the vibrations in the 2 conditions mentioned above would change under different frequncies. Those very low end notes, say, under 30Hz, if strong enough, would excite some larger motions which I couldn't tell it's calmer or not on top or bottom... )
The positions, strengths and frequency of the source of excitation, the inertia of the whole system (and the distribution), the positions of real and virtual pivot(s), the dampings, even the air load by the baffle.... etc. all come into play. It's so hard to pinpoint their individual actions and thier interactions.
I'm lost
Oh well, before I build some real swings, I think I'll just leave them there with the soft pads....
Thanks –
Its been an anodized profile that's used to make stylish curtains – the profile is hung from the ceiling and the gliders move inside the profile.
Luckily we have an excellent metal store here that also sells all sort of profiles. I cut it to small pieces the length of the MDF thickness and that's it...
Maybe IKEA or the like sells something similar - dunno
You possibly could adapt a simple tube profile too - or one of the special profiles for floors...
yes - I used a square knot :
http://www.wikihow.com/Tie-a-Square-Knot
– but feel free to choose any other you like:
http://de.wikipedia.org/wiki/Liste_von_Knoten
Usually, I hang my speakers with stainless stranded steel wires of ~1.5 mm.
Michael
Great system CLS
Now I understand better that you desperately look for any "non-swing" alternatives!
Thanks for your detailed report about cork pivots.
To get the most from impulse compensation you should treat each driver separately anyway – remember the each driver *has* to move to allow for impulse compensation – its just that we do not want
1) leakage paths
2) huge radiating areas attached or building an energy transmitting structure
In case of horns, I think it may depend on the rigidity of the horn itself if we possibly could consider it to be sort of local mass – or if its better to decouple the horn from the driver.
I'm lost
Oh well, before I build some real swings, I think I'll just leave them there with the soft pads....
It would be a lot of effort that's for certain!
Michael
..
This is great work you guys are doing - I'm loving this thread and all of the "out of the box"
*playing* (at its best) - definitely ! Michael
Last edited:
Ahoy Mate! Looks as if that six years I spent in the US Navy will finally pay off - that's the reason for the c2c in my screen name!!
The square knot was the first rope tying trick we learned. My 1st wife preferred a half-hitch - but that is an entirely different story.
Attachments
Hi Michael,
Thanks for the kind words. I've already spent quite a lot of time and effort on the system. It's certainly my joy, but also a burden, especially when some thoughts of major modifications are coming up....
I've been thinking about the mounting method of this:
http://www.diyaudio.com/forums/multi-way/123512-ultimate-ob-gallery-24.html#post1989991
Please look at the mounting of the sub - the woofer is separately mounted by a wood frame, instead of on the baffle. And there's a tiny gap between baffle and driver.
I think I'll steal this idea but with the drivers suspended. This way I can keep most of my existed stuff, "just" enlarge the baffle holes and hang those woofers.
Thanks for the kind words. I've already spent quite a lot of time and effort on the system. It's certainly my joy, but also a burden, especially when some thoughts of major modifications are coming up....
I've been thinking about the mounting method of this:
http://www.diyaudio.com/forums/multi-way/123512-ultimate-ob-gallery-24.html#post1989991
Please look at the mounting of the sub - the woofer is separately mounted by a wood frame, instead of on the baffle. And there's a tiny gap between baffle and driver.
I think I'll steal this idea but with the drivers suspended. This way I can keep most of my existed stuff, "just" enlarge the baffle holes and hang those woofers.
....
I think I'll steal this idea but with the drivers suspended. This way I can keep most of my existed stuff, "just" enlarge the baffle holes and hang those woofers.
For a first try you possibly could just loosen the woofers and move them a few mm towards you so that there is some play when hanging. You can enlarge the baffle to the outer diameter of the woofers later on...
If you hang them on thread rods like Stig Erik has shown earlier – I suggest to go around the magnet with the rope / steel wire as well.
This makes the thread rod the (stable) pivot - but takes a lot of the burden to hold the weight of the motor from them.
Make sure you have enough free length for the swing (shouldn't be any problem with your structure)
There are certainly more elegant ways to do - but it might be better to make it work first and improve on the design later on..
Michael
Last edited:
I have to say guys - I find this thread fascinating. I could even say inspiring. You've actually motivated me to get off my duff and tinker. I've hung a D28-2 and an Eton 7-380 and I find it remarkable.
It's hard to put into words but there's a sense of presence of expanded detail that is pretty surprising when you see bare drivers hanging from string - you just don't expect it to sound the way it does.
One thing I noticed while hanging the drivers, the D28 in particular, was that there was a strong pull from the nearby magnet to the mounting bolts. I don't want to stay into tin hat territory, but is the steel hardware (bolts, nuts, cable, etc) going to interfere with the symmetry of the driver's magnetic field? Should we be trying to using non-magnetic hardware? Sorry if this is a silly question.
It's hard to put into words but there's a sense of presence of expanded detail that is pretty surprising when you see bare drivers hanging from string - you just don't expect it to sound the way it does.
One thing I noticed while hanging the drivers, the D28 in particular, was that there was a strong pull from the nearby magnet to the mounting bolts. I don't want to stay into tin hat territory, but is the steel hardware (bolts, nuts, cable, etc) going to interfere with the symmetry of the driver's magnetic field? Should we be trying to using non-magnetic hardware? Sorry if this is a silly question.
Has anyone tried this OB di-pole type with a large MF planar (or 2)? Some large MF planers are designed for open-baf. di-pole use below 700Hz.
I do some consulting for the US importer of such a large MF planar (the 2030), by FPS; see them at the factory website (HQ office in Tokyo) FPS INC. Flat Panel Speakers - Products. The 1030 is narrower but much less output.
The next best large MF planar maybe the BG Neo10 (1/3 less $ but 1/3 the size of 2030 and much higher Fs) or HPV MAD1 (high $); availability of both has been limited.
Contact me for a free copy of my Article/Overview of ribbon & planar-magnetic drivers.
Let me know if more planar driver info. is helpful.
Regards,
David Kennedy
Home Page
LIVE-FIDELITY
I do some consulting for the US importer of such a large MF planar (the 2030), by FPS; see them at the factory website (HQ office in Tokyo) FPS INC. Flat Panel Speakers - Products. The 1030 is narrower but much less output.
The next best large MF planar maybe the BG Neo10 (1/3 less $ but 1/3 the size of 2030 and much higher Fs) or HPV MAD1 (high $); availability of both has been limited.
Contact me for a free copy of my Article/Overview of ribbon & planar-magnetic drivers.
Let me know if more planar driver info. is helpful.
Regards,
David Kennedy
Home Page
LIVE-FIDELITY
Hello David,
4x FPS 2030 can work pretty well i think. All other models have too low SPL for my tastes. But the 2030 is VERY interesting...
They have several advantages vs the BG Neos:
-Lower extension
-Square size (easier to do an array)
-Good power handling
-high impedance (min seems around 6,5)
But the THD is high (for my standards) and they dont show the curve.
Anyway, i didnt know FPS, thanks for mentioning them
How much are the 2030 in the US (i'm from europe, but i'd like to have an idea)?
Sent a pm to receive me your article, thanks.
Changes again!
The 12" Beyma's I had were reserved for an other dipole project... so I had to find replacements. Luckily I have a lot of drivers to pick from.... This time I wanted to try three 8" Excels for lower mid/bass. It works very well, and makes the main dipole look a bit slimmer. XO is 1700, 350 and 65 Hz. The three 8" can go flat to 40 Hz with just 6 dB EQ, but they very quickly run out of excursion. Better leave the low bass to the 21" H-baffle...
The 12" Beyma's I had were reserved for an other dipole project... so I had to find replacements. Luckily I have a lot of drivers to pick from.... This time I wanted to try three 8" Excels for lower mid/bass. It works very well, and makes the main dipole look a bit slimmer. XO is 1700, 350 and 65 Hz. The three 8" can go flat to 40 Hz with just 6 dB EQ, but they very quickly run out of excursion. Better leave the low bass to the 21" H-baffle...
An externally hosted image should be here but it was not working when we last tested it.
This is an awesome thread. Thanks, Erik!
Simple experiment to do though. I won't be in a position to try it out for several months (currently in a design phase and not in a building phase) but could look into it in the rather unlikely event nobody beats me to it.
Based on my experience developing electromagnetic field solvers I'd expect a major perturbation in the fringing field to have only small effects on the field distribution within the gap. A few bolts and some chain is a minor perturbation and should have no significant on the gab; my bet would be measurements wouldn't show a difference between steel hardware, non-ferrous metal hardware, and non-metallic hardware. Large areas of sheet steel plate or the like would probably show a difference.
Simple experiment to do though. I won't be in a position to try it out for several months (currently in a design phase and not in a building phase) but could look into it in the rather unlikely event nobody beats me to it.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.