The golden ratio is chosen for visual appeal. In practice fabricating a baffle with a true irrational number as its dimensions is impossible. For sonic benefits what you are really after is that for the two harmonic series that results in the two dimensions, none of the upper harmonics coincide until it is very high up, in other words the two tones from the two dimensions are 'dissonant'. 5:8 is actually a minor 6th. I surmise (haven't tried) that root 2 (1.414) might sound good. Even if we are imprecise and we round to 1.4 it still does not sound like any regular musical interval. By this logic 1.3 or 1.7 would sound good too.
[EDIT: I was thinking out loud and as I think more root 2 is a diminished fifth which would suggests 2 keys to gravitate to and a good speaker dimension ratio must be totally out of tune.]
[EDIT: I was thinking out loud and as I think more root 2 is a diminished fifth which would suggests 2 keys to gravitate to and a good speaker dimension ratio must be totally out of tune.]
Last edited:
As far as bafflestep goes, the width of the cabinet is usually the main event.
Bear in mind most drivers are designed to work in a cabinet a bit wider than the driver.
All this standing waves and modes and Golden Ratio business is a bit overrated IMO. At low frequencies the main event is the volume of the box.
At high frequencies, the speaker is so inefficient it hardly matters. Think 2% efficiency for most drivers due to the very poor cone to air coupling.
The surprise? The classic 2 foot by 1 foot by 1 foot cabinet is hard to beat. It just works well in the Super HL5+ according to Alan Shaw of Harbeth. He uses 12mm HDF with panel damping in the BBC tradition. But 18mm on the front baffle. Very light. A brace across the middle makes sense with a tall cabinet.
Harbeth - Loudspeakers - Super HL5plus - Hifi speaker
The M30.1 is nearer Golden Ratio on front baffle. Looks nice enough. But not actually the 0.618:1:1.618 ratio overall.
Harbeth - Loudspeakers - Monitor 30.1 domestic - Hifi speaker
Good drivers, good crossover and that is 90% of the battle. Good cabinet is the rest.
Bear in mind most drivers are designed to work in a cabinet a bit wider than the driver.
All this standing waves and modes and Golden Ratio business is a bit overrated IMO. At low frequencies the main event is the volume of the box.
At high frequencies, the speaker is so inefficient it hardly matters. Think 2% efficiency for most drivers due to the very poor cone to air coupling.
The surprise? The classic 2 foot by 1 foot by 1 foot cabinet is hard to beat. It just works well in the Super HL5+ according to Alan Shaw of Harbeth. He uses 12mm HDF with panel damping in the BBC tradition. But 18mm on the front baffle. Very light. A brace across the middle makes sense with a tall cabinet.
Harbeth - Loudspeakers - Super HL5plus - Hifi speaker
The M30.1 is nearer Golden Ratio on front baffle. Looks nice enough. But not actually the 0.618:1:1.618 ratio overall.
Harbeth - Loudspeakers - Monitor 30.1 domestic - Hifi speaker
Good drivers, good crossover and that is 90% of the battle. Good cabinet is the rest.
Diffraction can help with the staggered offsets (tweet to baffle edge).
There are measurements backing it up.
Me, I use foam so I don't worry about it (may get my felt cut some day).
I think golden ratio helps spread out standing waves inside the box.
I had a pioneer b20 in a box with no insulation, not 1 : 1.618 ratios, but 1' deep x 2' wide x 3' tall. Sounded pretty good (minus the honky sound probably created by the big whizzer).
There are measurements backing it up.
Me, I use foam so I don't worry about it (may get my felt cut some day).
I think golden ratio helps spread out standing waves inside the box.
I had a pioneer b20 in a box with no insulation, not 1 : 1.618 ratios, but 1' deep x 2' wide x 3' tall. Sounded pretty good (minus the honky sound probably created by the big whizzer).
You will find that for offset tweeter placement at the top of a tall baffle, the optimal placement is close to golden ratios between the sides and top, with the top having the shortest distance. I.e. Top:side1:side2 = 1:1.618:1.618^2 = 1:1.618:2.618. So if the distance from the tweeter to the top is 1 unit, the distance to the left or right should be 1.618 units and the distance to the right or left should be 2.618 units.
For woofers, generally the bigger the baffle the better to minimise losses due to omni-directional radiation at low frequency. As the size of the cone becomes significant compared to the baffle, the ripple becomes less severe so there is less need to place a woofer offset.
For woofers, generally the bigger the baffle the better to minimise losses due to omni-directional radiation at low frequency. As the size of the cone becomes significant compared to the baffle, the ripple becomes less severe so there is less need to place a woofer offset.
Hi Steve, talk about a fluke. My cabs are very near to the HLF5+. Mainly because these volts are meaty bastatrds and a 35L sturdy cab has driven the size. I'm using a 16mm solid aluminium baffle. The rest of the cabinet is 18mm with bracing.I'll be using self adhesive bitumen sheets to line the insides.
Just quoting because we are turning the page. FWIW, a mm on a tweeter just shifts the sweet spot slightly. Not worth worrying about.
I think this Volt 8" polycone plus SEAS 25mm tweeter project of yours really shouldn't be too hard. I have no idea why you think an 18mm aluminium baffle is a good idea though. No-one does this stuff. How you going to machine it?
And bracing a hugely heavy box is just not my idea of easy to build. Is it even necessary? The BBC didn't think so. Damping panels won't do anything on heavy rigid panels except add mass and stored energy which you will hear.
The only listening approach that makes any sense is to compare to live sound. That is the real yardstick. If that leads you to use classical music which you have heard live, so be it. But TBH, most people don't listen to anything other than an unreal studio sound, so are clueless.
This is how you do your Volt/SEAS speaker:
Rogers Loudspeakers › LS5/9
Not hard to flesh that one out for your particular drivers once you understand what it does well.
I can almost see where you are coming from, but the question isn't lending itself to an answer. What specifically do you think is at stake here?
18mm would be severe overkill. It might ring if you tap it with a hammer, but music is not going to excite it.
Golden Ratio is not better than other non-coincident ratio boxes that are based on irrational numbers.
It is pretty thou. I’ve done quite a few boxes that are close to Golden Ratio,
Root 2 works, but don’t use it twice.
dave
Time to get back on topic. This aluminium baffle may or may not work, time will tell.
The Golden Ratio, 1.618... or 0.618... , is one of the more interesting algebraic numbers:
Golden ratio - Wikipedia
Ramanujan found it particularly interesting as the simplest irrational constant to express as continued fractions. It is the limit of the ratio of successive Fibonacci numbers: 1,1,2,3,5,8,13,21,34,55,89,144, ...
FWIW, it is not a transcendental like Pi and Exp e.
It has a deep relationship with 5-fold symmetry, which in turn has a deep relationship with the 6 equal angle lines that can be constructed in our 3D space:
A New Path to Equal-Angle Lines | Quanta Magazine
And you thought that there were only 3 axes at right angles in 3D space that have equal angle symmetry!
Should we incorporate it in our rectangular monkey coffins? Probably does no harm. The idea is the vibrational nodes are in an irrational ratio, so overlap and reinforce as little as possible. No particular frequency is then overly emphasized.
However, does it really matter? Plenty of loudspeakers break the rules while sounding spectacularly good. Like this Spendor S100 example owned by Vrystaat:
Lot to like there. Vibration and damping is generally spectacularly misunderstood. It's why bridges fall down and buildings topple. Time really does tell.
The Golden Ratio, 1.618... or 0.618... , is one of the more interesting algebraic numbers:
Golden ratio - Wikipedia
Ramanujan found it particularly interesting as the simplest irrational constant to express as continued fractions. It is the limit of the ratio of successive Fibonacci numbers: 1,1,2,3,5,8,13,21,34,55,89,144, ...
FWIW, it is not a transcendental like Pi and Exp e.
It has a deep relationship with 5-fold symmetry, which in turn has a deep relationship with the 6 equal angle lines that can be constructed in our 3D space:
A New Path to Equal-Angle Lines | Quanta Magazine
And you thought that there were only 3 axes at right angles in 3D space that have equal angle symmetry!
Should we incorporate it in our rectangular monkey coffins? Probably does no harm. The idea is the vibrational nodes are in an irrational ratio, so overlap and reinforce as little as possible. No particular frequency is then overly emphasized.
However, does it really matter? Plenty of loudspeakers break the rules while sounding spectacularly good. Like this Spendor S100 example owned by Vrystaat:
Lot to like there. Vibration and damping is generally spectacularly misunderstood. It's why bridges fall down and buildings topple. Time really does tell.
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.