|
Post by karatestu on Dec 11, 2019 9:44:15 GMT
Hello.
Some prefer speakers with only one drive unit (full range) and the benefits that they bring. Others have 2, 3, 4 way and the crossovers that approach brings. Then there are the slimline towers with multiple mid bass drivers. BMTMB, MTM and every combination thereof. Some have bass drivers side firing to keep the front baffle thin. Then there are the quasi omni directional and semi omni speakers. That is where my interest lies.
My speaker project has morphed in to a ridiculous arrangement of multiple bass drivers and tweeters with some mid bass drivers thrown in for good measure. My 12" bass drivers have a fairly high Vas even for a sealed enclosure so i went isobaric to try and keep the box volume down (there are benefits other than that and an efficiency penalty). I have tried these bass cabs down firing, up firing, side firing, back firing and the traditional front firing. Down firing and rear firing are truly awful ways to point the bass drver with the set up i have. Side firing just sounds weird to me - can't put my finger on it. Up firing and forward firing are both good for me with the front firing probably edging it due to the close proximity of the driver to the floor (less baffle step).
Currently i have the 12 inchers in clamshell isobaric configuration with just the front baffle separating them and the rear driver out of phase with the front. That way i get to keep the box volume to it's minimum (without changing drivers of course). It sounds good but looks ugly with the back of the cheap Chinese drivers and magnet not being the best aesthetically.
I only have one 5.25" mid bass per channel firing upwards as per the NVA cubes. I have always liked this direction of fire for the mid bass. However i have plans to try additional 5" drivers side firing and maybe front firing all at the same time as the up firing one. Might end up like a Linn Brik's god forbid.
Recently i have been going through a tweeter breeding project. I have always had the original single tweeter firing forwards. It is essential imo. But being the inquisitive sort i wondered about adding others firing in different directions. Tweeters were pulled from other speakers and even a pair of small full rangers out of some PC powered jobbies. First was up firing, then inward firing, then outward firing. I considered putting one on the back but the rats nest of wires would not permit it.
I really like the result even with some low quality tweeters and most of them just blu tacked and gaffa taped to the baffles. It makes me want to replace the added ones with decent Visaton models the same as the front firing tweeter and make proper driver cut outs to help with diffraction. I have all four tweeters per channel wired in series so i have been able to get rid of the 12R padding resistor that was required when only using one tweeter.
It works imo and plays to my tastes perfectly (ie pinpoint imaging is not that important to me). So what would you call this frankenspeaker (apart from a mess) ? Front firing 12" bass, up firing (and possibly forwards and to both sides) 5" mid and 1" soft dome tweeters firing forwards, upwards and to both sides. Alldrivers are doped NVA style and no xover apart from a single high pass filter capacitor for the tweeters per channel. I must be bonkers.
|
|
|
Post by John on Dec 11, 2019 9:54:19 GMT
Love to see some pictures of your project
|
|
|
Post by MikeMusic on Dec 11, 2019 10:06:03 GMT
Wot John says !
|
|
|
Post by MartinT on Dec 11, 2019 10:17:01 GMT
I have been an advocate of multiple drivers each handling an optimum area of the frequency spectrum - provided they actually match each other in characteristics and the crossover is well designed.
The worst I ever heard was a Lumley speaker with about 8 different drivers which were a mismatch cooked up in hell. Terrible, incoherent, sound with very poor bass.
More recently my JM Lab (Focal) Mezzo Utopias almost got it right but the Tioxid tweeter was still prominent with what sounded like a shelf lift in the crossover.
The best match are my current Ushers with beryllium midrange and tweeters, and fast Eton woofers that can keep up with them. Very coherent, very musical. I feel that Usher have since spoiled the Be-20s with the diamond tweeter that is too prominent, just like in the Focals.
|
|
|
Post by jandl100 on Dec 11, 2019 10:56:01 GMT
I've tried all manner of speakers from single driver planars (I can't get on at all with single driver cones) to my current end-game speaker which is a 4-way 6-driver with omni-directional carbon fibre petals for upper mids and treble, and symmetrically side-firing lower mid and bass metal cone drivers. Sounds like a real mess of driver types and orientations, but it meets all of my needs. I've never heard a speaker that beats it all-round for my tastes.
|
|
|
Post by karatestu on Dec 11, 2019 11:08:15 GMT
Thanks for the replies. The speaker world is divided that's for sure probably because along with the room they have the biggest effect on sound. We all like different things, place some parameters higher in importance than others. I think there is a huge trade off and it is all swings and roundabouts. With diy you can try to design and build a speaker as near as possible to what you think is perfection. I love to read about people's speaker projects, different ways of doing things and flying in the face of convention. What i don't like to read is internet "experts" dismissing a design or build having not heard it and with all their experience in a simulation programme. You have to make the things and listen in your own room with your own music and ears. Many of the things i have tried would be laughed at and dismissed as wrong because a book or forum post they have read said so. Some unorthadox things i tried worked and some did not. I don't want this to turn into a rant...............but it has I don't have electrical xovers to worry about because there aren't any (apart from a single high pass filter cap on the tweeters). But i often forget that there are mechanical xovers from the drivers themselves, the enclosure they are in and the way they are loaded. A sealed enclosure is a 2nd order high pass filter a ported speaker a fourth order high pass. Then you have the drivers rolling off all by themselves at either end of their frequency range. Then there are resonant frequency peaks to deal with. I have multiple drivers with frequency ranges that overlap heavily, relying on natural doped driver smooth roll off and the high pass filter that is the enclosure. It is a bit of a juggling act but i can honestly say i have never enjoyed music as much as i do now. I may have the odd small dip and peak in the frequency response here and there but i don't care, the things it does well more than makes up for that. With heavily doped 12" bass and 5.25" mid bass i am never going to win an efficiency contest but that is not needed for me. I have a small room and don't listen at ear splitting levels. I find it hard to explain what the doped driver xoverless approach brings to what i hear but it is so natural and sounds most like what i hear when playing with other musicians in the bands i am in. Open, lively, coherent, natural. It has been said it is a phase thing. I have no way of measuring things so i have to go by my ears and subjective speculation. I doubt i could design a decent passive crossover to save my life without a lot of practice. I love the freedom of putting a driver in a box of a certain volume, pointing in a certain direction and mixing up as many of them in as many directions as i think sounds good / best to me. Bollocks to all the text books and internet warriors. I avoid the highly technical forums for obvious reasons. But there are still phase issues to deal with even though i don't have xovers per se. Group delay still happens. I have drivers on different baffles pointing in different directions. You might call it quasi omni. So far i have tried not to put more of one of the same type and size of driver on the same baffle / panel. Multiple tweeters on the same baffle don't work imo. What was my point ? I have forgotten. Oh that was it. try everything and don't be surprised if you like it
|
|
|
Post by John on Dec 11, 2019 11:40:03 GMT
Speaking for someone who has a very radical design that most people who have not heard it would say cannot work. I love reading your openness to finding your own path
|
|
|
Post by jandl100 on Dec 11, 2019 11:43:23 GMT
Love to see some pictures of your project
|
|
|
Post by karatestu on Dec 11, 2019 11:54:51 GMT
I will post some up. However these are still very rough prototypes made from the crapest recycled 19mm chipboard with silicon sealant and decorator's caulk plastered everywhere to ensure air tightness.There are old holes plugged with bits of metal and wood, tweeters stuck on the side of the smaller cabs with blu tack and gaffa tape.
|
|
|
Post by MartinT on Dec 11, 2019 11:55:44 GMT
Two single-driver speakers have impressed me over the years, but both were cheats. The Horning Agathon Ultimate used a Lowther full range unit and sounded extraordinarily vivid, but it had twin 12" Beymer bass drivers at the back for the bottom octave. The Martin Logan CLX large panel speakers were probably the only panel speaker ever to have impressed, and were a lot more dynamic than Quads, but at the demo I attended they were still supplemented by a subwoofer.
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Dec 11, 2019 11:59:49 GMT
With the advert of the large 3D printing machines real world modelling now can be achieved in a matter of hours (smaller designs) Another vote against single driver, yes they do a few things well, but unlike the old farts that mostly listen to them I like to have a complete picture of the musical soundscape.
|
|
|
Post by karatestu on Dec 11, 2019 13:18:58 GMT
The tweeter breeding project started like this with a cardboard top baffle extension with up firing 1" metal dome tweeter from a pair of of B&W P4 transmission lines. Little cuboid sitting on the big bass cube with forward firing 12". Back then i had the isobaric configuration with both drivers having the same volume behind them (as RD instructed me it was the proper way of doing it - like his cubix speakers). I still thought the bass was too loud about it's resonance frequency ie not enough enclosure volume. The next tweeter i added to the inside faces of the little cuboids. They are tweeters from a pair of MS902S stand mounts.Nothing special just bog standard 1" metal domes blu tacked on (later i added gaffa tape to smooth diffraction a little). After adding each tweeter i noticed how much better it sounded due to lower distortion from sharing the work out. It really was astonishing especially with the first additional tweeter. Not ready to stop there i realised i had run out of tweeters that i could scavenge. That led me to ripping a part a broken pair of powered pc speakers. crappy little things with small oval full range speakers (well full range if you weren't bothered about bass). As these were going after the high pass filter capacitor i thought they would be ok. Again i thought distortion lowered a little even with these cheap crap drivers although diminishing returns were definitely setting in after the second tweeter was added. The other thing is the speakers are much more omni directional than before and this was evident in the listening. Much more spacious and airy with a widened sound stage and height. Pin point imaging was never the strong point of semi omni speakers but i never found them wanting in that respect. With the addition of all these tweeters i don't think it has changed a great deal. I also managed to remove the 12R padding resistor that was in series with the single front firing visaton soft dome tweeter (without doubt the best tweeter out of this thrown together line up). I put this down to the series wiring i employed for a chain of four tweeters and the fact they are firing in different directions. The high frequency that has to bounce off surfaces to reach my ears has further to go and so loses dB along with dB loss from the actual reflection. I had to adjust the high pass filter cap down in value due to the increased impedance of all those tweeters in series. Finally i gave the FR tweeter a little enclosure with gaffa tape There have been further developments regarding the bass enclosure but need to take a couple of pics before i post that.
|
|
|
Post by Slinger on Dec 11, 2019 13:23:12 GMT
Be honest, you're actually building the head for one of Dr Who's enemies in the new series, aren't you, Stu?
|
|
|
Post by karatestu on Dec 11, 2019 13:29:57 GMT
Be honest, you're actually building the head for one of Dr Who's enemies in the new series, aren't you, Stu? Ha ha , yes Paul. Just working out who is gonna get exterminated first
|
|
|
Post by karatestu on Dec 11, 2019 13:56:58 GMT
This is breaking news i haven't even posted it on hfs yet. Just to show you the lengths i go to for highly scientific experimentation using state of the art materials and techniques I wanted to retry up firing bass drivers now that i have the drivers in clamshell configuration with a much larger volume behind the rear driver. Obviously the air volume between the two driver's cones is very little with only 19mm chipboard between them. It is hard to arrange the little cuboid next to the big cab when doing up firing bass drivers. So i came up with this little elegant solution which has me thinking now I really have surpassed myself this time. Using an ikea lack platform i spanned the space between the window ledge and the top of the driver's magnet but the heights were slightly out. This was solved with a little length of wood between the driver and the ikea lack panel. A big weight on the back (56 lbs) and little cuboid speaker plonked on top and we were ready to rock. There is quite some weight pressing down on the driver magnet which i guess can only be a good thing. Although there might be more vibration passed to the upper cab. I tried to bring the tweeter's dome to the centre of the 12 incher's cone for some crude time aligning. This works way better than the last time i tried up firing bass drivers. I am listening to it now and it is one of the best configurations i have come up with yet. I am still going to try putting all the drivers in the big bass cube and not using the smaller cab. But i will live with this for a while and think about how it can be refined for the final build.............if that ever happens
|
|
|
Post by speedysteve on Dec 11, 2019 16:56:53 GMT
Multiple drivers for me. And you can't beat volume for bass. "Ye cannae change the laws of physics Captain". If done properly, crossovers are not a problem at all.
I'm at 5 way, as that is the generally accepted optimum for front loaded horns regarding the division of the frequency range.
|
|
|
Post by MartinT on Dec 11, 2019 18:10:51 GMT
Stu, do you think the improvement came about from having so many 'treble' drivers balancing out the bass driver's sensitivity, or was it more from the omni characteristics of them firing in all directions? Or both?
|
|
|
Post by karatestu on Dec 11, 2019 19:40:35 GMT
Stu, do you think the improvement came about from having so many 'treble' drivers balancing out the bass driver's sensitivity, or was it more from the omni characteristics of them firing in all directions? Or both? Hi Martin. The main improvement when adding tweeters was the lowering of distortion. It was quite astonishing how much clearer things became and the amount of hidden detail exposed. With one tweeter and a first order high pass filter at corner frequency of just under 6 khz the tweeter is having to produce quite high spl below the crossover and at lower frequencies than is good for it. That is the classic grumble about first order xovers. When the first tweeter was added the drop in distortion was very large probably down to the tweeters having less work to do. At first i was not really listening for the added omni effect because the lowered distortion was such a jaw dropping moment. It became much more apparent when the side tweeters were added. I could then hear its effect much more clearly and the law of diminishing returns set in as regards lower distortion. I must say that the omni effect is very pleasing to me and the added tweeters would have been well worth it just for that. The driver sensitivity balancing was already done before i addded tweeters because the original tweeter had a 12 ohm padding resistor in series (no Lpad). That 12R balanced it with the doped 5" mid - i call it a mid but really it is a mid bass as it is run full range with no filter. The doped 12" is alsk run full range with no filters and the balance has usually been good, depending on how big an enclosure i had at the time or direction of firing. I expect all three things have had an effect. The lowering of distortion was by far the biggest benefit followed by the omni effect. Getting rid of one component (a single padding resistor) between the amp and the tweeter might have had an effect but it is too hard to tell with all the other stuff going on. It does make me happy to think that a component has been removed but then i have added three new components (tweeters). I expect (hope) it can only get better when i replace the up and side firing tweeters with three more Visaton G25ffl the same as the front firing one. They are £70 a pop and i need six of the buggers. Also when they are all nice and neatly installed into proper cut outs the dffraction should be considerably better.
|
|
|
Post by karatestu on Dec 13, 2019 6:42:54 GMT
I saw an ebay listing for 128 Visaton full range speakers for £250. I put it onmy watch list and then got a 20% off message from the seller so down to £200. Do i really need 128 drivers ? Could build a couple of very big line arrays with them
|
|
|
Post by MartinT on Dec 13, 2019 6:56:39 GMT
You certainly could!
|
|