Ideas and advice for Aluminium spheres project

[andyburlow]

I have recently found some aluminium spheres, which I believe were used as vacuum vessels and I wish to make them into a speaker enclosure using a full range driver.
I have measured the internal volume as holding 2.1 litres of water and the spheres are about 175mm in height.
I would like to get the spheres machined to take a driver in the side of the sphere like in the first photo, this will allow me to bolt a base to the neck.
I think I could get a 4.5 inch driver in there and if not definitely a 3 inch driver. I may have to weld some lugs in there if there is not enough meat left after machining (wall is about 5 mm).
Can anyone advise if the lack of volume would be a waste of time fitting a driver like the CSS FR125S or should I stick with a 3" like the HiVi B3N?
I have considered getting the spheres machined to take the smaller and cheaper drivers but if the FR's will fit and work well enough would prefer to use them.
I would like to use the speakers for normal level music or maybe as satellites for a surround system (I have 5 spheres) and expect to use a sub.
This is my first foray into the world of DIY Audio so please forgive the vague questions.
Many thanks, Andy

[LineArray]

Originally posted by andyburlow
Can anyone advise if the lack of volume would be a waste of time fitting a driver like the CSS FR125S

Hi Andy,

you may interpret the result by yourself. For a closed
box 9 litres would be OK for FR 125S.

So normally i would say, use a driver with smaller Vas.

But since i love these drivers i tend to give it a chance:

If you can live with XO frequency to the sub between
100 and 120 Hz you could build a box with a resistive
vent, sometimes called "aperiodic box" or "resistance box".

If you fill the box entirely with an appropriate wadding
you may increase acoustical volume to 2.4 litres, but
keep in mind, that you will loose some volume by
truncating the spheres and inserting the drivers.
So in the end you may end up with 2 liters again ...

The controlled "leakage" has to be designed to adjust Qts
between 0.6 and 0.8 . The value of >1.0 you get for a box
without leak is too high and will sound fairly boomy...

Design of the crossover to the sub will be a bit tricky ...
I would use the natural rolloff of the satellites.
If you don't need very much sound level, you can run them
fullrange for first approach. Maybe add a highpass with
-3 dB rolloff < 80 Hz later, to reduce cone excursion.

I strongly recommend to build two test enclosures from
conventional material first (maybe cubes with some internal
diffuser since cubes are not very beautiful due to standing
waves inside ), to design the leakage properly and see whether
the satelites meet your needs.

If everthing is adjusted well, then start modifying your vessels,
since that looks like big effort ...

How do the vessels sound, if you knock on them with a hammer?
Maybe you should add some damping pads to get them quiet.

To me the project looks tricky but still interesting ...
Interesting look, interesting material and acoustical quality
can be interesting too ...


Let's hear what other specialists think about it in this forum


Cheers

[andyburlow]

Thanks for the feedback, you are correct in your assumption that I will loose volume due to the machining, maybe as much as a third may dissapear.
Maybe these are destined to have smaller drivers which will suit the lower Vas.
I think your idea of internal damping is good as they have a low ring when knocked with a hammer although it does not hang around for long.
As the machining will not be easy (try holding a sphere in a vice) I might not get many attempts to experiment with driver size although starting small and working upwards is always a good idea.
Can anyone recommend a 'good' 3 inch driver?

[LineArray]

Hi Andy,

what about Bandor 50 ?

http://www.bandor.com

Low XO to the mono sub will be a problem,
maybe a vented design will be necessary ...

Will not be a party box, depends on your needs.

Cheers

[sreten]

Hi,

the Dayton RS100 drivers have a cutout of 80mm, for 175mm dia,
this leads to an included angle of 56 degrees with no flat for the
driver mounting. The thickness of your walls is a design issue.
It will appear to be wider the smaller the driver to provide
a flange to the hole, assuming removing a flat section.
To a degree you can choose how wide to make this flange.
The cutout is small for a 4", they are really nearer a 3".
The B3S the cutout is 70mm.

A proper 4" I'd say would need mounting on an additional ring.

http://www.zaphaudio.com/smalltest/

Zaphaudio.com is the place to go but you need to ask yourself
some serious questions regarding limited sensitivity and volume
levels regarding 2 channel music.
Stereo subs and highish c/o point seem your best bet.

Zaph is fairly adamant 3" is a far as you can go for fullrange, and
that 3" are better with an additional small neo-type tweeter.
So one could say for a 4" you should assume + tweeter.
Decent dispersion matters for AV use, not so much stereo.

You should also assume AV filtering, optimum stuffing and some
tricks for the volume might be possible, e.g. a stuffed hollow
cylinderical stand could add a litre or so of effective volume.
Optimum stuffing can add 30% to effective volume.

Some of the compromises you have to work out for yourself.
There are other sphere projects on this forum.

If you have a 40hm capable AV then the RS100-4 looks promising.
Note after adding a BSC filter it will be 8 ohm in the upper range.

The B3N looks the easiest to fit (some sort ring to support the
folded circular flange might be needed) and information on the
electrical filters needed is available in other threads.

The CSS drivers I'd say are non-starters. Volume is too low to use
any of their advantages effectively. Fostex's are efficient but no
excursion, so pointless with a juicy AV amplifier.

Based on simple mounting and a round frame I think it comes
down to the B3N on its own, the RS100's or RS100's + neo's.

/sreten.

[es44]

I'm thinking that if more volume is needed, how about letting the base be part of the whole speaker ?

Just a thought, i have no experience with speakerbuilding yet, but read a great many posts here while drooling

Best regards
Ebbe

[Apoclypse Boy]

Originally posted by andyburlow
Thanks for the feedback, you are correct in your assumption that I will loose volume due to the machining, maybe as much as a third may dissapear.
Maybe these are destined to have smaller drivers which will suit the lower Vas.
I think your idea of internal damping is good as they have a low ring when knocked with a hammer although it does not hang around for long.
As the machining will not be easy (try holding a sphere in a vice) I might not get many attempts to experiment with driver size although starting small and working upwards is always a good idea.
Can anyone recommend a 'good' 3 inch driver?

I'm busy with soe 300mm plastic ones , you're small Al ones will work a treat in a Lathe, machine the face flat and tap the front, 5 mm will be ample. Damping will make all the diff. Ran mine empty and sounded crap , aded one layer of cold tar, 1" HD foam and 1 pound of Dacron, i'm really amazed how they sound

[andyburlow]

Good advice about the damping and machining, thanks. What drivers have you used? You have a bit more volume than me.

[jkeny]

Originally posted by Apoclypse Boy


I'm busy with soe 300mm plastic ones , you're small Al ones will work a treat in a Lathe, machine the face flat and tap the front, 5 mm will be ample. Damping will make all the diff. Ran mine empty and sounded crap , aded one layer of cold tar, 1" HD foam and 1 pound of Dacron, i'm really amazed how they sound

Where did you get the plastic spheres? I was thinking of buoys but all the modern ones are inflatible, I believe. I wanted to experiment, cheaply, with spheres for Jordan JX92S drivers.

[OzMikeH]

Get a block of hardwood.
machine out a socket for the sphere. it doesn't have to be perfect.
Glue the sphere into the socket with lots of white PVA wood glue.

Allow lots of time for glue to cure, a week should do it.

Do your machining, rest frequently to precent the sphere from heating up and softening the glue.

When finished soak in water. pull sphere out.
do not use a socket more than once. the soaking wrecks it.