Double 15-sketch (works with one driver only, too but I haven´t drawn one yet)
070806, intro
Quick explanatory detour before we start off: I have had the drivers used in Lynn O:s "Ariel" at home for a couple of years now, and built boxes for them with an MLTL line instead of the labyrinth described by Lynn, but the result when doing a quick comparison with my Spendor S100:s was a bit of a disappointment. The drivers sounded clearly inferior compared to the character of the Spendor's. Since I knew the bass was never going to be as good (the S100:s are a full range three way system, remember), I lost a bit of motivation in finishing them.
This is an open baffle speaker with inspiration mainly from three sources:
The name of the project is, therefore: Beyond endorphine minimalists. Not very good, I know, but they will receive another name after birth, as is so often the case with children, especially brain-children.
070807
Below is the current sketchwork for the baffle. Beware, it is NOT final and will change. Drivers as of today are Saba greencones, 20 cm wideband drivers and 10 cm tweeters. I recently ordered two Jensen P15R from two different ebay sellers, let´s hope they arrive unharmed and prove to be usable.
Oh, by the way, I was going to put another 15" driver in the baffle too, to help with the frequencies from 100Hz down to around 40 - all according to the discussion in "Beyond the Ariel", see link in section 1 in the ordered list above.An Eminence beta 15 perhaps. But at first, there will be only the one Jensen 15" driver.
Right now, most components are bought or I have them already. Currently waiting for delivery of Jensens, a couple of caps and inductors, and for the baffles to get cut. It might be another week or two before something of interest happens here.
070921
Looking at formulas and tables for open baffle falloff frequencies, it seems that not only is absolute baffle width important, but also placement of the driver on the baffle. According to Linkwitz, if the driver is placed assymetrically the rolloff starts higher than if you place it in the very center of the baffle. You need to use the shortest distance from driver center to edge. The formula I use for calculating is
v/4*width
where v equals the speed of sound, and the width is the shortest distance from the center of the driver to the edge of the baffle, in meters. Times two, since the sound wave has to travel along the back of the baffle too before it gets nulled by the corresponding back wave.
Thinking about this, the double-woofer sketch above seems a bit of a silly solution, right? With this placement, we get around 25 cm from baffle edge to each drivers center. Putting the values into the formula: speed of sound approx. 340 m/s, width 0.5 meters. This gives us 340/4*0,5=170Hz. So breakpoint for start of falloff for each driver is 170 Hz with this solution, while the falloff breakpoint of the smaller wideband driver ends up at 340/4*0,9≈94 Hz. Approximately, of course, and I haven't taken the contribution from the side supports into consideration here; they turn this speaker into some kind of U-baffle, but I wouldn't know how to calculate falloffs for a U-baffle anyway. At any rate, I would rather like the falloff breakpoints reversed... thus, yet another sketch. This effectively switches rolloff points for the drivers:
With breakpoints switched, the acoustical falloff of the wideband driver is now at around 170 Hz (those dipole acoustic falloff slopes are at 6dB/octave, if I am correctly informed). It can be combined with an electric first order crossover, at around 200Hz or perhaps even 300-400 Hz. Combined, this gives us 12dB/octave falloff, roughly harmonising the acoustical falloffs to work with Lukasz Fikus crossover filter design ideas, which I find very appealing.
Not having two 15" drivers means I lose output level, so yes, I will use two. Not in the main baffle though, the other one will go into a separate bass unit. Perhaps something like what mr. Fikus does with his bookshelf dipole subs.
I have cut out two baffles from birch board, and also side supports. Next up is cutting the holes for the drivers. Unless I change my mind again at the very last moment, I'll try out this last sketch for the hole patterns.
080102
Ooops, it seems we have a new year... Not much happening here, but I am happy to see that Lynn Olson is still thinking on and planning for his new speakers too... the same could not be said of the Lampizator, who starts building a new design for himself or someone else about every fortnight or so. I am very impressed!
Not that I haven't been busy - since the last blog entry our family have been occupied with finding, buying and moving into a new house. As I write this, the 12 week old little kitten who moved in at the same time as we did is prowling around examining an obviously particularly interesting spot on the living room carpet. By some strange coincidence this happens to be the very spot of interest for todays blog.
A new house means many new problems and possibilities, of course. The Big IB or horn subwoofer-in-the-attic dreams, which I guess most nerds must nourish in situations like mine, will be something to patiently wait with. Our house is small, and will continue to be so for a couple of years, so we had to compromise and make a combined bedroom/living room. From above, it will look something like this when all the furniture is moved into place. In all, it is about 18 sqm (5.3·3.5 m).
Sorry, this is a very crude sketch, I know.
Below the room drawing is a calculation of the 1st, 2nd, 3rd and 4th harmonics of resonance frequencies for the width, length and height of the room. The Hz scale is not logarithmic. The most problematic resonances look like the ones around 125-140 Hz where we have 3rd harm. for length and 2nd harm. for height almost on top of each other, and at around 65-70 where we have 1st height and 2nd length harmonics meeting. In theory, that is.
The spot where the cat was playing with the mat is were we´re at now. Look at the three room dividing rectangles in the middle of the above drawing. They are bookshelves seen from above. The two centre bookshelves - it is really a shame to call'em bookshelves, but they will look a lot like that so I'll just continue shall I? - are about 110 cm/43" high while the full height bookshelf is supposed to be around 200 cm/79". I plan to use the two centre bookshelves for hiding the BEM OB:s while they aren´t playing (or for playing, when someone is listening from some other room). When playing for active listeners, they can be pulled out of their slots and moved to the side walls. They will need wheels...
The centre bookshelves will probably need to be custom built. But the real stroke of genius (well - that is as close as it gets, coming from me, anyway) was when I realized that the "hole" above the BEM OB Hider Bookshelves could have... a large piece of glass or perspex matte screen above it! Sic!
Errrm well... The matte screen is (I hope) seen above obscuring the double doors in the background. It could be used for back projection for movies or for watching TV, and when the projector is off it would still act as a divider between living room and bedroom, at the same time letting the precious light (as you may know we Scandinavians love light) through from the double doors into the living room half. I imagine the BEM OB:s with a thin, screen printed or hand painted linen cloth in front (no, the left speaker cloth cover won't be cut through in reality, it is just a drawing thingy, mate). There are shelves inside the BEM OB Hider Bookshelves too, these could hold the power amps + PSU:s. The top shelves are used for electronics, I just drew two Aikidos in there to show it. Perhaps there will be doors in front of these topshelves.