tktran said:
I'm almost certain that I'm going to offend here but I'll go ahead anyway and appologise later.
With all due respect I'm the one with the drivers right here in front of me, I'm the one who's spent over 100 hours listening to open baffle, infinite baffle, ported, I'm the one who built test cabinets to find the best configuration for my preference and room, I'm the one who then spent hours with the god awful SW measuring TS parameters for individual drivers to get a more accurate picture whilst going back and modelling.
When you spend this much money on stuff its hardly worth taking things lightly. So to say I've been thorough in planning, testing and executing is an understatement. Its been over 6 months in the works and its going to likely be another 6 months before they are finish. There is only so much I can do before I actually get the drivers here and start testing, so the first step is a blind leap of faith based on preliminary modelling, reviews, opinions, statistical data, reputation, quality and anything else you can find on the net. I took a chance on the Seas just like most people do who don't build an established kit or design.
The Seas drivers aren't upto scratch in comparison to the mating drivers. No alignment, no loading, no fairy dust cable is going to change that. Take a listen to them on an open baffle, no box influences and ****** all bass but guess what, they still sound muted compared to the ATC.
I really don't know what more I could have done in the last few months that I haven't already tried.
I wouldn't take TS parameters as gospel as far as loading goes, you'd be surprised how far the units I have deviated from published specs and each other. The L22 is suited to both ported and sealed. Like I said above I tried sealed, ported and OB.
The L22 RNX4P isn't the 8535 though.
I guess the first impression count. If I'd listened to them hidden behind a grill and was suitably blown away and then saw them, it wouldn't matter.
With the internet everything is like a candy store, a shame but true.
Take a look at the sub scene for justification on that.
The Seas by all accounts have a good reputation for thier price range. If I've tried the one of the best at that price point then of course I'll have to fork out more money. The Seas, as far as I'm concerned doesn't get any better, why waste even more time, like 6 months wasn't enough, trying to get it to do things it can't do?
Finally sorry for tone but I almost pulled my hair out in despair when I read your post, I've tried lots but am still not happy.
Just some little remarks:
Do you measure in-room response at the listening position?
If yes, take a careful look at the say 80-400 Hz response. You might have a room resonance there. Such a resonance can significantly muddy the bass and "slam". 3 dB may matter and are easily overlooked on the usual messy FFT picture, plus speaker and microphone placement are very sensitive. Measurements should be done in many variations before forming an opinion.
You may have the best aligned speakers in the world, yet a +6 dB room resonance in the 100 Hz region will make them sound slow, muddy and imprecise. BTW I have dipoles and still manage to have 6 dB peaks and valleys in room. I mildly EQ for that and it makes a world of difference.
In general, some minor (+- 1 dB) differences in critical regions (80-400, 1500-3000 Hz) can change perceived sound quality dramatically. Bass goes from clean to muddy, mid from nasal to throaty, tweeters go from harsh to natural sounding etc.
Just my experience.
Oh BTW for years now I use the exact same drivers in my system. But I keep on finding ways to improve via cabinet type, X-O, low end extension placement, room EQ and placement in-room.
Do you measure in-room response at the listening position?
If yes, take a careful look at the say 80-400 Hz response. You might have a room resonance there. Such a resonance can significantly muddy the bass and "slam". 3 dB may matter and are easily overlooked on the usual messy FFT picture, plus speaker and microphone placement are very sensitive. Measurements should be done in many variations before forming an opinion.
You may have the best aligned speakers in the world, yet a +6 dB room resonance in the 100 Hz region will make them sound slow, muddy and imprecise. BTW I have dipoles and still manage to have 6 dB peaks and valleys in room. I mildly EQ for that and it makes a world of difference.
In general, some minor (+- 1 dB) differences in critical regions (80-400, 1500-3000 Hz) can change perceived sound quality dramatically. Bass goes from clean to muddy, mid from nasal to throaty, tweeters go from harsh to natural sounding etc.
Just my experience.
Oh BTW for years now I use the exact same drivers in my system. But I keep on finding ways to improve via cabinet type, X-O, low end extension placement, room EQ and placement in-room.
I beleave that people cross the W210 woofers over in the range 300-400 Hz depending on room an application.
The W210's are corner placed to get rid of early reflections, TacT beleaves that this is the best base for room correction.
A guy at the yahoo users group has done a W110 with one 10" woofer. The building plans for the W210, W410 and W110 can be downloaded from the yahoo TacT users group file area.
Woofers and parts for a kit can be purchased from TacT dk and possibly uk.
The 10" woofers are not typical subwoofers, as they have a lower driver mass and strong double magnets. They are woofers ment to be crossed high made especially for TacT by vifa.
I have heard the W210 woofers and they are very musical.
In order to do corner placement, you need to compensate using a digital time delay, but this is easy with your setup.
The W210's are corner placed to get rid of early reflections, TacT beleaves that this is the best base for room correction.
A guy at the yahoo users group has done a W110 with one 10" woofer. The building plans for the W210, W410 and W110 can be downloaded from the yahoo TacT users group file area.
Woofers and parts for a kit can be purchased from TacT dk and possibly uk.
The 10" woofers are not typical subwoofers, as they have a lower driver mass and strong double magnets. They are woofers ment to be crossed high made especially for TacT by vifa.
I have heard the W210 woofers and they are very musical.
In order to do corner placement, you need to compensate using a digital time delay, but this is easy with your setup.
ShinOBIWAN said:
Disabled Account
Joined 2003
ShinOBIWAN,
If I caused some offence, this was not my intention.
Unfortunately I have not followed this thread from post number #1, but of course there are 38 pages of stuff to go through. Perhaps I should have, before my first post.
I do not know what you've tried, what you haven't, and I was merely trying to offer some constructive criticism or other options to try, before you dump the drivers and spend more money on others.
Have you talked to others who have successfully implemented your drivers? What they tried, what worked or didn't, for them? Troels Gravesen is the only person I've seen who has published his work with the L22RNX4P- that is why I put it across as a suggestion.
(Troels uses the L22RNX4P as a replacement (same cabinet, similar tuning, to the 8535))
What was your vented alignment and how did it sound to you?
Another question: Have you experimented with stuffing? In subwoofers it changes Qtc, but in midwoofers I find that this can totally change the behaviour of the midrange too.
My first post was a picture comparing vented to sealed, using SEAS T/S data. Obviously Troels data is different, and will be your T/S data may be different. That's why I ask if you've measured near-field and in-room response. I think that's reasonable, and I'm not sure how many people could give useful feedback without knowing more information.
Actually I came across this thread because I saw your For Sale post.
Just my 2c, and just trying to help.
If I caused some offence, this was not my intention.
Unfortunately I have not followed this thread from post number #1, but of course there are 38 pages of stuff to go through. Perhaps I should have, before my first post.
I do not know what you've tried, what you haven't, and I was merely trying to offer some constructive criticism or other options to try, before you dump the drivers and spend more money on others.
Have you talked to others who have successfully implemented your drivers? What they tried, what worked or didn't, for them? Troels Gravesen is the only person I've seen who has published his work with the L22RNX4P- that is why I put it across as a suggestion.
Of course. My point was that in my experience, what can sound good vented can sound totally different when sealed. NOT JUST A LACK OF BASS, which is the common perception. This happened with the 8535- muted, closed-in, lacking dynamics and slam. Could it be possible that it happens with other drivers? Absolutely! Could it be possible that this happens with the L22RNX4P? Quite possibly.
(Troels uses the L22RNX4P as a replacement (same cabinet, similar tuning, to the 8535))
What was your vented alignment and how did it sound to you?
Another question: Have you experimented with stuffing? In subwoofers it changes Qtc, but in midwoofers I find that this can totally change the behaviour of the midrange too.
My first post was a picture comparing vented to sealed, using SEAS T/S data. Obviously Troels data is different, and will be your T/S data may be different. That's why I ask if you've measured near-field and in-room response. I think that's reasonable, and I'm not sure how many people could give useful feedback without knowing more information.
Actually I came across this thread because I saw your For Sale post.
Just my 2c, and just trying to help.
ShinOBIWAN said:
I have not read most of the comments on the AT drivers.. as to the Supravox 215 GMF..
I've looked over a LOT of drivers for potential pairing with the ATC mid and the that particular Supravox is always at the top of the list (..even when looking at different diameter drivers). As to the cheapness of the driver, yes it doesn't look as well crafted as even much less expensive drivers like the Seas you have (..and no where close to the AT drivers), but appearances can be decieving. Let go through a few things here (on their page(s)):
http://www.supravox.fr/haut_parleurs/215_GMF.htm
First and foremost we have a relativly high eff. design with a quoted 97 db. The higher the eff. - usually the better the dynamics (subjectivly). Of course more eff. designs have more restrictive bandwidth (usually), but not the Supravox (especially considering the sd). So it is something of a miracle that fs is just below 50 Hz for a driver that has an eff. (1w/1m) greater than 95 db. Of course the higher eff. also allows the opportunity to add mass (in this design with 2 drivers in parrallel) for reducing fs to an extent, but more importantly to add a degree or diaphram resonance control.
Next thing on the list (that is noteworthy) is the suspension. Progressive pleated suspensions are just plain better than other designs in a midbass driver (provided that suspension is "broken-in"). (..in particular they store less energy near the diaphram.) Its particularly rare to see a paper pleated suspension that offers more than a couple of millimeters of excursion (..usually they are cloth and as a result add a good deal more mass to the mms of the driver with an excursion over half a millimeter). Here the linear excursion is +/- 5 mm's (which is also extraordinary considering that most pro 10 inch drivers typically have at best 3.5 mm's of excursion). Considering the material and the linear excursion potential of this driver (and that once the suspension is "broken-in"), I'd say that the level of resistance will be VERY low for this driver under normal operating conditions. This essentially will allow the suspension to "get out of the way", allowing the signal to control the dampening of the woofer magnetically. With higher quality amplification you will typically achieve greater levels of "transparency" quite similarly to the behaviour of a true ribbon (with better amplification and no additional electrical resistance).
As to the frame - it looks like a good cast aluminum frame with better than average limted air flow restriction. It should be fairly non-resonant. Nothing spectacular (like PHY's frame), but nothing particularly poor either.
Now the gauss of the magnet IS special. 1.5 T is quite a bit higher than average (which is usually 1 to 1.1). This suggests a magnet material that is FAR more rigurously manufactured than normal. The benefit is typically greater "transparency" due to the additional force on the VC and greater linearity of the field (due to smaller/better grain structure of the magnet). I wouldn't be at all surprised if the magnet on the Supravox is a small neo magnet (and a high quality on at that). The magnet cap may look a little cheap, but its really just there for a bit of shielding (and a place to put their label).
The last thing that really catches my attention is the mms of the driver. Consider that this is an 8 inch driver - with a mms of only 10 grams. Over that much surface area you have a VERY lightweight diaphram. Again, just like the low resistance provided by the surround, the mass of the diaphram will store less energy in the short term and as a result react more quickly (time based) to changes in spl. Note that AT (in their T/S params) includes the psuedo param of "acceleration factor" which is force factor (Bl) divided by mms (times 1000). Their standard 8 inch flex unit has an acceleration factor of 423, their 4 inch flexuit is considerably better at 667, the Supravox 215 GMF is 850 (i.e better than most good midranges available).
Now you can (to a degree) actually see the benefit to the low mass in the spectral decay graphs they provide:
http://www.supravox.fr/mesures/mes215GMF2.htm
Notice that stored energy (THD) initially appears pretty crappy on this driver (at least in comparison to something like a good tweeter). But the bandwidth we are most concerned with is below 400 Hz. (Please note that unlike most manufacturers - Supravox actually has a large aneohic? chamber capable of measuring this low, so the graph is FAR more realistic than what you would normally get with a psuedo measurement via the extreme near field, and as a result appears far worse than what others provide.) In any event, you'll see the primary (aka fundamental) as well as harmonic distrotion with the graph. The three important things to note here are (within the desired bandwidth):
1. the speed of decay for the primary.
2. the sp difference in level between the primary and the harmonic.
3. the speed of decay (and its level) for the harmonics.
For #1 we have about a 10th of a millisecond - which is better than a lot of GOOD tweeters.
For #2 we have aproximatly 9-12 db of "clean" decay. Moreover unlike most drivers the decay extends (lower in freq.) beyond Supravox's ability to measure.
Combine parameters 1 & 2 and you have pretty much the recepie for driver "transparency" (or "muddiness" if the result is poor). Note that the 215 signiture is considerably better in this respect, but it will not have the dynamics of the 215 GMF.
For #3 The result is exceptional with almost 20 db of fast harmonic decay. The price for this is a nasty broad band resonance centered around 700 Hz - but with 24 db an octave slopes (or more) at 400 Hz this is a non-problem.
As to the enclosure - well that depends a lot on the size of the enclosure you are willing to put up with and with respect to how much sp "head room" you want (and overall freeedom from compression) vs. low freq. extension.
Still I think that these drivers in parrallel in an aperiodic "sealed" box with digital eq. (L. transform or otherwise) is probably the best solution for you (..unless you like ear bleeding sp levels). Furthermore, because of the delicate nature of the diaphrams, you'll need a good laminar airflow around the driver's themselves (in the box). (I'll get to the exact construction method for such an airflow device in another post if you want.. & note that it will be critical with these drivers otherwise you'll screw up much about the drivers that is good because of airflow resistance and reflections.) The drivers are already spec'ed at 97 db (though I doubt its quite that high in actuallity) + you'll have another 6 db due to parralllel VC's and voltage amplification with a doubling of surface area. I'd also limit extension to almost precisely 50 Hz. Combine these factors (add in BSC and limitations of your box's volume), and eq. will not be as extensive as you might think (i.e. you shouldn't have problems with achieving fairly high sp levels and avoiding all compression effects - unlike what you have now).
As for the 24 inch 2450, size can be a deceptive performance standard. Remember I mentioned the Beyma 12LX60 as my "pick"? If you look at their larger drivers with the same magnet structure you'll see that the THD is really no better with the larger drivers (at least at 1 watt/1 meter). Moreover you are better with multiple drivers Paralleled even when considering increased sp levels. Another HIGHLY advantageous effect of having more drivers is the ability to have more smaller boxes distributed throughout the room to minimize room effects and decrease eq. (Note this is the most effective acoustic method for dealing with room nodes, and is the most effective method period for dealing with room nodes over a large coverage area.) Horizontal distribution is "key" with this method (i.e. perhaps 4 subs behind your loudspeakers spanning the distance from speaker to speaker with equal spacing between each sub.)
If however you were to utilize only one driver (and though I don't know the THD figures for it - but would find out b4 making a puchase), I'd look at the 24" Hartley:
http://www.diyaudio.com/forums/showthread.php?s=&threadid=50124&pagenumber=5
visual here:
http://www.diyaudio.com/forums/showthread.php?s=&threadid=50124&perpage=10&pagenumber=8
and here:
http://www.arduman.com/aa/Sayfalar/francisco/francisco.htm
ps. good to here that the A-Holes are incompetent as well.
Cheers Scott
As always Scoitt a huge thanks for the indepth and interesting discussion. Your one of the best folks on here.
Looking at the waterfall it does indeed look great in the range that I'm looking to run it at.
It seems I've actually been looking in the wrong places, so I'll almost certainly be moving over to the 215GMF.
Could you elaborate on box construction and airflow requirements further please? My target volume is around 50L max with a single 215, unfortunately there's no way I can justify 2 since I've already spent over £3000 and all I have to show for it is just have a bunch of drivers, amps, electronics, a dedicated PC XO, expensive tins of paint and a lot of scrap MDF 😉 Quite scary actually when its put like that!
So for now its definitely just the single 215 per side, more extravagance will have to wait until the next project.
I'm kinda losing motivation now since I put a lot of effort into what's been shown in this thread so far and its basically all been scraped and I'm virtually back to square one with the only certain thing being the XO details between the ATC and SS ring. I've got cabinets to completely redo and then start all over again with building test enclosures for the supravox - I really don't know if I could face that again so soon after.
So please any help you could offer to extract the best performance from the 215 would be a big help.
Thanks,
When the time comes I'll look into driver choices more deeply.
ScottG said:
As always Scoitt a huge thanks for the indepth and interesting discussion. Your one of the best folks on here.
Looking at the waterfall it does indeed look great in the range that I'm looking to run it at.
It seems I've actually been looking in the wrong places, so I'll almost certainly be moving over to the 215GMF.
Could you elaborate on box construction and airflow requirements further please? My target volume is around 50L max with a single 215, unfortunately there's no way I can justify 2 since I've already spent over £3000 and all I have to show for it is just have a bunch of drivers, amps, electronics, a dedicated PC XO, expensive tins of paint and a lot of scrap MDF 😉 Quite scary actually when its put like that!
So for now its definitely just the single 215 per side, more extravagance will have to wait until the next project.
I'm kinda losing motivation now since I put a lot of effort into what's been shown in this thread so far and its basically all been scraped and I'm virtually back to square one with the only certain thing being the XO details between the ATC and SS ring. I've got cabinets to completely redo and then start all over again with building test enclosures for the supravox - I really don't know if I could face that again so soon after.
So please any help you could offer to extract the best performance from the 215 would be a big help.
Thanks,
When the time comes I'll look into driver choices more deeply.
good post Scott . I have same dilema about ATC mid and low freq driver , but I wanna go with bigger one 12 " or 15" .
what do you think about 400-2000 EXC (maybe with silver coil )?
can it be good mach with ATC ? its pretty expencive , ~1150 euro a piece . never heard those supravox ones .
other candidates are Volt RV3143 an Lambda TD12 or TD15 .
what do you think about 400-2000 EXC (maybe with silver coil )?
can it be good mach with ATC ? its pretty expencive , ~1150 euro a piece . never heard those supravox ones .
other candidates are Volt RV3143 an Lambda TD12 or TD15 .
I'm looking into buying some 215 GMF but I can't find anywhere that does them. Sure I've found French and German sites but with no english version of the website.
I'd rather not buy from folks that I can't speak their language just in case of problems.
Anyone know of dealers in France/Germany/Europe that have english sites?
Thanks
Ant
I'd rather not buy from folks that I can't speak their language just in case of problems.
Anyone know of dealers in France/Germany/Europe that have english sites?
Thanks
Ant
ShinOBIWAN
I have come into this thread late, and was wondering whether you are still using the active XO you posted in the beginning. I also noticed an article that talked about Thiel speakers, and Mr. Theil indicated that they use 1st order XOs for time coherence. As you see the goals seem to be the same, but the methods are totally different.
Do your test correspond to what you hear quite well?
I have come into this thread late, and was wondering whether you are still using the active XO you posted in the beginning. I also noticed an article that talked about Thiel speakers, and Mr. Theil indicated that they use 1st order XOs for time coherence. As you see the goals seem to be the same, but the methods are totally different.
Do your test correspond to what you hear quite well?
soongsc said:
I'm not using the XO that I outlined in the beginning, like many things they were scrapped in favour of a digital XO.
Time/phase alignment can be achieved in many ways, so I wouldn't pay credence to anyone that says there is only one way.
ShinOBIWAN said:
Did you test the phase characteristics (minimum phase) along with the frequency responses for each driver after they were in the enclosure or before?
I've been messing around with the Seas L22's again over the weekend and decided to go ahead and port them by just routing out a hole just blelow the drivers for front firing and hten popping some 3" drain pipe in there which was 17cm long.
I also cut away the wall seperating the two chambers of the cabinets - remember how I had a seperate smaller chamber for the amps to mount in? I then sealed the whole thing back up to finally create an internal volume of 65ltrs for the amps to work in as opposed to the original 40ltrs.
Afterall this the simulation predicted an F3 of around 30hz with a port tuning frequency of around 30hz.
I figured what the hell and gave it a go, since I was planning on scrapping the cabinets anyway so wasn't bothered about messing them up for this little experiment.
The last time I tried ported I believe that the volume was around 35ltrs and tuning was from 23 upto 35hz, done by simply cutting the port down after each listening test. I remember that they were boomy and ill defined, you know the kind of bass that just seems blurred and slow, more like tones rather than anything like sharp edges, well that was the impression compared to the sealed 20ltr test box no matter where I tuned them too. I believe they sounded best at around 23hz tune which is natural as this driver has a tendancy to peak on the tuning frequency over around 30hz when I was modelling.
Please also bear in mind that they were NOT paired with the ATC or Scan at this stage! I had them crossed over to a pair of Mission 771e's using a behringer DCX2496 at 400hz. Since that is all I could test with at the time.
Finally the boxes also did not have any sort of bracing and I'd built them much less rigorously than the finals. They were tested with varying amounts of stuffing, such as behind the drivers, on the walls, very thin fill and no fill at all. I didn't try them with the dedshete absorbers at all since they would have been wasted and it isn't cheap stuff and I should also mention that the port was rear firing.
This time around the cabinet was very sturdy with doubled walls and extensive bracing, stuffing was limited to a small amount behind the drivers and then only on the cabinet walls, dedshete panels installed throughout and the port was front firing instead of rear. However the most important aspect is the inclusion of room correction EQ that wasn't in the test boxes! Sealed generally doesn't excite room nodes as much as ported and my room has a nasty hump at 55hz and 120hz. More short sightedness on my part and again I spoiled my sealed implimentation with it after I had already dismissed ported in the early stages of planning.
I must say that I have egg on my face. I've completely misjudged these drivers and been rude to tktran when his advice was actually the right stuff. Just because I'd spent time on ONE particular ROUGH ported solution and dismissed it, I'd actially failed to try the finer details and perhaps a more indepth look at its possibilities. I guess in fact what I did subconciously was to build rough test boxes then look for the most promising INITIAL results and follow that line of enquiry, of course that meant that sealed, which often yields better results initially was chosen as that route to follow instead of pursing a robust investigation of not the initial promising result but actually of the two.
However without a really thorough investigation of ported I'd quite clearly missed details - and probably still am.
What am I talking about?
After all the mods the Seas have actually never sounded better, gone is that muffled and ill definied quality I was talking about, they actually really seem to sing now and with big dynamics and just as transparency is at least 90% of the ATC, they really seem like a great match now. The crossover has had to be modified to a 4th order acoustic at 380hz which to me sounds quite exceptional in the overall tonal balance when crossing over to the ATC. You can't tell where the L22 ends and the ATC begins - its now that much of a difference, where as before it was clear that the sound changed in tone.
Its funny because I've removed the LT circuit and I've got just as much bass but it seems more lively, quicker and certainly more of dynamic. Listening to tracks that I've listened to over and over again when setting up the sealed solution had been a revelation. Massive Attack has some nimby yet big bass lines, these now sound cracking, really jumpy and effortless. They can really move around in pitch and keep it very clean. They also go loud now without that compressed sound that I got with the LT forcing the drivers to 20hz.
I'm still going to need a sub since they have an in room F3 of 27hz, I do not want to use any EQ to boost this at all since the drivers are ported and run quickly into problems.
I also cut away the wall seperating the two chambers of the cabinets - remember how I had a seperate smaller chamber for the amps to mount in? I then sealed the whole thing back up to finally create an internal volume of 65ltrs for the amps to work in as opposed to the original 40ltrs.
Afterall this the simulation predicted an F3 of around 30hz with a port tuning frequency of around 30hz.
I figured what the hell and gave it a go, since I was planning on scrapping the cabinets anyway so wasn't bothered about messing them up for this little experiment.
The last time I tried ported I believe that the volume was around 35ltrs and tuning was from 23 upto 35hz, done by simply cutting the port down after each listening test. I remember that they were boomy and ill defined, you know the kind of bass that just seems blurred and slow, more like tones rather than anything like sharp edges, well that was the impression compared to the sealed 20ltr test box no matter where I tuned them too. I believe they sounded best at around 23hz tune which is natural as this driver has a tendancy to peak on the tuning frequency over around 30hz when I was modelling.
Please also bear in mind that they were NOT paired with the ATC or Scan at this stage! I had them crossed over to a pair of Mission 771e's using a behringer DCX2496 at 400hz. Since that is all I could test with at the time.
Finally the boxes also did not have any sort of bracing and I'd built them much less rigorously than the finals. They were tested with varying amounts of stuffing, such as behind the drivers, on the walls, very thin fill and no fill at all. I didn't try them with the dedshete absorbers at all since they would have been wasted and it isn't cheap stuff and I should also mention that the port was rear firing.
This time around the cabinet was very sturdy with doubled walls and extensive bracing, stuffing was limited to a small amount behind the drivers and then only on the cabinet walls, dedshete panels installed throughout and the port was front firing instead of rear. However the most important aspect is the inclusion of room correction EQ that wasn't in the test boxes! Sealed generally doesn't excite room nodes as much as ported and my room has a nasty hump at 55hz and 120hz. More short sightedness on my part and again I spoiled my sealed implimentation with it after I had already dismissed ported in the early stages of planning.
I must say that I have egg on my face. I've completely misjudged these drivers and been rude to tktran when his advice was actually the right stuff. Just because I'd spent time on ONE particular ROUGH ported solution and dismissed it, I'd actially failed to try the finer details and perhaps a more indepth look at its possibilities. I guess in fact what I did subconciously was to build rough test boxes then look for the most promising INITIAL results and follow that line of enquiry, of course that meant that sealed, which often yields better results initially was chosen as that route to follow instead of pursing a robust investigation of not the initial promising result but actually of the two.
However without a really thorough investigation of ported I'd quite clearly missed details - and probably still am.
What am I talking about?
After all the mods the Seas have actually never sounded better, gone is that muffled and ill definied quality I was talking about, they actually really seem to sing now and with big dynamics and just as transparency is at least 90% of the ATC, they really seem like a great match now. The crossover has had to be modified to a 4th order acoustic at 380hz which to me sounds quite exceptional in the overall tonal balance when crossing over to the ATC. You can't tell where the L22 ends and the ATC begins - its now that much of a difference, where as before it was clear that the sound changed in tone.
Its funny because I've removed the LT circuit and I've got just as much bass but it seems more lively, quicker and certainly more of dynamic. Listening to tracks that I've listened to over and over again when setting up the sealed solution had been a revelation. Massive Attack has some nimby yet big bass lines, these now sound cracking, really jumpy and effortless. They can really move around in pitch and keep it very clean. They also go loud now without that compressed sound that I got with the LT forcing the drivers to 20hz.
I'm still going to need a sub since they have an in room F3 of 27hz, I do not want to use any EQ to boost this at all since the drivers are ported and run quickly into problems.
COOL! Its good to here that things are working-out without a great deal of additional expenditure.
I have just finished a fairly lengthy reply on the enclosure for the Supravox 215 GMF, but it appears you don't need it.🙄 (for me, 😀 for you.)
I think I'll still post it though to give you and others some ideas of what I consider to fairly "optimal" enclosure techniques.
I have just finished a fairly lengthy reply on the enclosure for the Supravox 215 GMF, but it appears you don't need it.🙄 (for me, 😀 for you.)
I think I'll still post it though to give you and others some ideas of what I consider to fairly "optimal" enclosure techniques.
Note: this was written without seeing Shin's recent succesful implementation of the Seas units.. BUT I'd thought it might hold at least some interest. Furthermore it appears that I will need to split it into two posts so here goes..
Sorry about not replying earlier, life got in the way..🙄
Typically Sealed enclosures give a better sense of dynamics than any other method, so I did some simplistic modeling for 1 Supravox 215 GMF (without eq. and no baffle step compensation) and came up with a little over 60 liters (64) of internal volume for the optimal "loading". Thats pretty close to your target of 50 liters.
So then 64 liters will net you aproximatly -3db at 100 Hz, -5 db at 75 Hz, and -10 db around 48-50 Hz. Thats really quite good for a sealed alignment of this volume, and it means that you won't need too eq. much for a 45- 50 Hz flat response - (in fact considering the driver's additional eff. over the mid and treble units while factoring in BSC and L.T.. I'd say that at reasonably loud listening levels you'll rarely be using more than 12 watts - not to shabby.) Now this doesn't mean that you must use 64 liters, a 50 liter volume will produce about a 1 db decrease at 50 Hz over the 64 liters (and less of a decrease than that at 75 Hz and 100 Hz) – so there isn’t a big difference here.
So then what do we need to do to get this right in one shot for an otherwise seamingly simplistic "box"?
In my opinion there are several problem areas that are overlooked by many people, (including long-time industry professionals), that contribute significantly to the sound of an enclosed loudspeaker (particularly in the midbass).
1. "Micro-Rocking". This is a motional behaviour problem between the driver and the enclosure. Use a fairly rigid baffle (like a good mdf) and you'll still have some back and forth movement of the driver that does cause problems relating to "transparency" (or #'s 1 & 2 that I provided on my previous post). Part of the problem is "lossy" in nature and part of it is resonant with something like good mdf. You can make it purely "lossy" (and remove most of the resonance character) by using something like 3M's Microsorb (which is pretty much the best material of its kind). That however is NOT ideal. Instead what is really desired is a mounting that is both rigid and non-resonant.
2. "Air-Flow Resistance". This problem is at its worst the closer the resistance is to a driver. In effect, at close range (less than 12 inches), it will mechanically dampen the driver's motional behaviour and can even increase the driver's mms and lower the fs and eff.. This is pretty much the opposite of what we want (..magnetic signal controlled dampening). "AF" resistance greater than a foot in length can also be detrimental and lead to a delayed release in box pressure (i.e. a build-up and release in the box) that will increase emf problems with the driver.
3. "Back-Diaphram Reflections". This is a problem with higher freq.s and "thin" profile non-ridgid diaphrams on drivers. (at lower freq.s pressurization is the problem.) In this instance then the worry is that higher freq. reflections will reflect of of the driver's frame or motor assembly and from the box itself - causing higher freq.s to actually "bleed-through" the diaphram and become audible.
4. "Box Resonance". This can include Micro Rocking, but in this instance I'm concerned with both external vibrations (in-room sound generated by the box) and more importantly internal pressure building and release that is also related to longer AF resistance.
So then what's the most cost effective way to deal with these issues for a sealed enclosure?
Well #'s 1, 2 (the "close range variety), and 3 CAN (if constructed properly) be effectivly eliminated with a Laminar Air-flow and some un-conventional baffle "mounting".
OK then, the air flow device is really nothing more than a duct with a low drag-coefficent for the driver's rear-ward "stroke". In this case then the "duct" is within the box and encapsulates the driver within the box and "dumps" the driver's rear radiation out of the "duct" and into the box. Now we don't want to buid up an in-band pipe resonance within the duct.. (but I'll get to that solution in a bit.)
The usual method for such an air-flow device is as follows from a cross-section view of the side of the speaker with the driver mounted on the front):
Baffle (as a part of the box); driver mounted to baffle (Sealed); Duct/Pipe (with an interior diameter just larger than the driver's diameter) mounted to the baffle internally (Sealed) surrounding the driver; Duct/Pipe extending beyond driver's magnet and typically as much as 2/3rd's the internal depth of the speaker; Duct/Pipe with exit within 1/3rd of the internal box's rear panel; conical reflector (like a funnel) with a diameter the size of the interior diameter of the Duct/Pipe attached to the internal box's rear panel in-line with the exit of the Duct/Pipe.
Now to get rid of a potential pipe resonance, to decrease the drag-coefficient, and to decrease reflections we will spend a few dollars on a lot of simple (but extremely effective) plastic drinking straws without the flexible "bend". (you could also use medical grade tubing like Pro-Flo, but it isn't neccesary). We will stuff the Duct/Pipe with these straws until they physically "stay-put" simply from the contact pressure. (btw, you don't want a "crimp" in the straw, but some "collapsing" is inevitable and not detrimental.) The straws will "span" the length of the Duct/Pipe starting from around the driver's magnet (the frame's end cap on the magnet closer to the diaphram) and end at the exit. Unless you can find some long straws it will likely take 2 straw lengths worth (..and no, the staws do not need to perfectly "adjoin" to effectivly create on long straw). BTW, I believe at least one manufacturer utilizes exactly this method for their reference speaker (the Beauhorn Virtuoso).
You might have recognized that many of the straws (those that are directly behind the driver's magnet assembly) will not have any beneficial effect with regard to air-flow. In fact we don't need to stuff straws here but we do need something here for the straws so that they will stay in place. Consider then a rigid dowel that is the diameter of the driver's end "end"-cap. Now such a dowel could act as a driver "Micro-Rocking" reduction device that attaches to reflector and then on to the rear panel of the box acting as a "brace" for the driver. However you'll loose some volume this way and I'll explain a much superior method for "Micro-Rocking" elimination momentarily. In otherwords I'd just fill the Duct/Port full of straws alone with no such device.
OK so at this point we have close range air-flow resistance and reflections pretty well taken care of (though I'll point out a method latter for reducing reflections further). But we don't have a reduction in Micro-Rocking.
What I propose now is to make the Duct/Port out of a rigid and massive material with an edge thickness that will allow the driver to be mounted on the Duct/Port (instead of on the baffle). The Duct/Port then will be sealed (but NOT "attached") to the Baffle via silicone caulk. The support for the pipe will be rigid legs in the box attached to the Duct/Port and the box's bottom plate.
As far as Micro-Rocking is concerned this structural "build" will not only remove any lossy or resonance characteristic that would occur with a normal baffle, but it also eliminates any box resonance that would "transmit" to the driver via its connection to the baffle.
So then whats the "material" for the Duct/Port? Why Cement, what else? Specifically Quikrete (and in particular Quickwall) - the kind found at hardware/home improvement stores that does NOT have, or need, any aggregate (aka pebbles), and is usually white in color (though it can be dyed black). This typically only requires the cement (and dye coloring), a 5 gallon mixing bucket, a drill based mixing paddel, a drill (should be strong), form release, and a form.
Everything but the form release and the form is pretty self explanatory. The form release can easily be obtained via PURE silicone lubricant spray. The form itself is nothing more than two tubes of the right diameter for the form, a flat bottom plate, and a top plate with appropriate screw-in's" jutting into the form that will stay there for screwing/bolting the driver to the Duct/Port. Screw-in's can also typically be found in the same hardware store. (If you wanted to recess the driver some then you would need a positve impression of the driver's frame surround "extending" to the depth you desired the driver to be "inset".) The form's tubes are usually made from cardboard and are often called "footing" forms and can typically be found right where the cement is in the same store.. tubes can also be found at packaging/mailing centers (and can even be special ordered through manufactureres). The plates can be made from MDF scrap. Note that you only need to make one form (but you'll need enough screw-in's for 2 Duct/Ports). The screw-in's are best glued with a fairly weak glue to the top form plate so that it is easy to take the plate off with the screw-in's staying in-place in the concrete Duct/Pipe. Silicone caulk usually works quite well to attach and seal the cardboard pipes to the plates (but do NOT use caulk in the form gap, ONLY on the outer rim of the outer pipe and inner rim of the inner pipe).
Sorry about not replying earlier, life got in the way..🙄
Typically Sealed enclosures give a better sense of dynamics than any other method, so I did some simplistic modeling for 1 Supravox 215 GMF (without eq. and no baffle step compensation) and came up with a little over 60 liters (64) of internal volume for the optimal "loading". Thats pretty close to your target of 50 liters.
So then 64 liters will net you aproximatly -3db at 100 Hz, -5 db at 75 Hz, and -10 db around 48-50 Hz. Thats really quite good for a sealed alignment of this volume, and it means that you won't need too eq. much for a 45- 50 Hz flat response - (in fact considering the driver's additional eff. over the mid and treble units while factoring in BSC and L.T.. I'd say that at reasonably loud listening levels you'll rarely be using more than 12 watts - not to shabby.) Now this doesn't mean that you must use 64 liters, a 50 liter volume will produce about a 1 db decrease at 50 Hz over the 64 liters (and less of a decrease than that at 75 Hz and 100 Hz) – so there isn’t a big difference here.
So then what do we need to do to get this right in one shot for an otherwise seamingly simplistic "box"?
In my opinion there are several problem areas that are overlooked by many people, (including long-time industry professionals), that contribute significantly to the sound of an enclosed loudspeaker (particularly in the midbass).
1. "Micro-Rocking". This is a motional behaviour problem between the driver and the enclosure. Use a fairly rigid baffle (like a good mdf) and you'll still have some back and forth movement of the driver that does cause problems relating to "transparency" (or #'s 1 & 2 that I provided on my previous post). Part of the problem is "lossy" in nature and part of it is resonant with something like good mdf. You can make it purely "lossy" (and remove most of the resonance character) by using something like 3M's Microsorb (which is pretty much the best material of its kind). That however is NOT ideal. Instead what is really desired is a mounting that is both rigid and non-resonant.
2. "Air-Flow Resistance". This problem is at its worst the closer the resistance is to a driver. In effect, at close range (less than 12 inches), it will mechanically dampen the driver's motional behaviour and can even increase the driver's mms and lower the fs and eff.. This is pretty much the opposite of what we want (..magnetic signal controlled dampening). "AF" resistance greater than a foot in length can also be detrimental and lead to a delayed release in box pressure (i.e. a build-up and release in the box) that will increase emf problems with the driver.
3. "Back-Diaphram Reflections". This is a problem with higher freq.s and "thin" profile non-ridgid diaphrams on drivers. (at lower freq.s pressurization is the problem.) In this instance then the worry is that higher freq. reflections will reflect of of the driver's frame or motor assembly and from the box itself - causing higher freq.s to actually "bleed-through" the diaphram and become audible.
4. "Box Resonance". This can include Micro Rocking, but in this instance I'm concerned with both external vibrations (in-room sound generated by the box) and more importantly internal pressure building and release that is also related to longer AF resistance.
So then what's the most cost effective way to deal with these issues for a sealed enclosure?
Well #'s 1, 2 (the "close range variety), and 3 CAN (if constructed properly) be effectivly eliminated with a Laminar Air-flow and some un-conventional baffle "mounting".
OK then, the air flow device is really nothing more than a duct with a low drag-coefficent for the driver's rear-ward "stroke". In this case then the "duct" is within the box and encapsulates the driver within the box and "dumps" the driver's rear radiation out of the "duct" and into the box. Now we don't want to buid up an in-band pipe resonance within the duct.. (but I'll get to that solution in a bit.)
The usual method for such an air-flow device is as follows from a cross-section view of the side of the speaker with the driver mounted on the front):
Baffle (as a part of the box); driver mounted to baffle (Sealed); Duct/Pipe (with an interior diameter just larger than the driver's diameter) mounted to the baffle internally (Sealed) surrounding the driver; Duct/Pipe extending beyond driver's magnet and typically as much as 2/3rd's the internal depth of the speaker; Duct/Pipe with exit within 1/3rd of the internal box's rear panel; conical reflector (like a funnel) with a diameter the size of the interior diameter of the Duct/Pipe attached to the internal box's rear panel in-line with the exit of the Duct/Pipe.
Now to get rid of a potential pipe resonance, to decrease the drag-coefficient, and to decrease reflections we will spend a few dollars on a lot of simple (but extremely effective) plastic drinking straws without the flexible "bend". (you could also use medical grade tubing like Pro-Flo, but it isn't neccesary). We will stuff the Duct/Pipe with these straws until they physically "stay-put" simply from the contact pressure. (btw, you don't want a "crimp" in the straw, but some "collapsing" is inevitable and not detrimental.) The straws will "span" the length of the Duct/Pipe starting from around the driver's magnet (the frame's end cap on the magnet closer to the diaphram) and end at the exit. Unless you can find some long straws it will likely take 2 straw lengths worth (..and no, the staws do not need to perfectly "adjoin" to effectivly create on long straw). BTW, I believe at least one manufacturer utilizes exactly this method for their reference speaker (the Beauhorn Virtuoso).
You might have recognized that many of the straws (those that are directly behind the driver's magnet assembly) will not have any beneficial effect with regard to air-flow. In fact we don't need to stuff straws here but we do need something here for the straws so that they will stay in place. Consider then a rigid dowel that is the diameter of the driver's end "end"-cap. Now such a dowel could act as a driver "Micro-Rocking" reduction device that attaches to reflector and then on to the rear panel of the box acting as a "brace" for the driver. However you'll loose some volume this way and I'll explain a much superior method for "Micro-Rocking" elimination momentarily. In otherwords I'd just fill the Duct/Port full of straws alone with no such device.
OK so at this point we have close range air-flow resistance and reflections pretty well taken care of (though I'll point out a method latter for reducing reflections further). But we don't have a reduction in Micro-Rocking.
What I propose now is to make the Duct/Port out of a rigid and massive material with an edge thickness that will allow the driver to be mounted on the Duct/Port (instead of on the baffle). The Duct/Port then will be sealed (but NOT "attached") to the Baffle via silicone caulk. The support for the pipe will be rigid legs in the box attached to the Duct/Port and the box's bottom plate.
As far as Micro-Rocking is concerned this structural "build" will not only remove any lossy or resonance characteristic that would occur with a normal baffle, but it also eliminates any box resonance that would "transmit" to the driver via its connection to the baffle.
So then whats the "material" for the Duct/Port? Why Cement, what else? Specifically Quikrete (and in particular Quickwall) - the kind found at hardware/home improvement stores that does NOT have, or need, any aggregate (aka pebbles), and is usually white in color (though it can be dyed black). This typically only requires the cement (and dye coloring), a 5 gallon mixing bucket, a drill based mixing paddel, a drill (should be strong), form release, and a form.
Everything but the form release and the form is pretty self explanatory. The form release can easily be obtained via PURE silicone lubricant spray. The form itself is nothing more than two tubes of the right diameter for the form, a flat bottom plate, and a top plate with appropriate screw-in's" jutting into the form that will stay there for screwing/bolting the driver to the Duct/Port. Screw-in's can also typically be found in the same hardware store. (If you wanted to recess the driver some then you would need a positve impression of the driver's frame surround "extending" to the depth you desired the driver to be "inset".) The form's tubes are usually made from cardboard and are often called "footing" forms and can typically be found right where the cement is in the same store.. tubes can also be found at packaging/mailing centers (and can even be special ordered through manufactureres). The plates can be made from MDF scrap. Note that you only need to make one form (but you'll need enough screw-in's for 2 Duct/Ports). The screw-in's are best glued with a fairly weak glue to the top form plate so that it is easy to take the plate off with the screw-in's staying in-place in the concrete Duct/Pipe. Silicone caulk usually works quite well to attach and seal the cardboard pipes to the plates (but do NOT use caulk in the form gap, ONLY on the outer rim of the outer pipe and inner rim of the inner pipe).
continuation:
Both the internal support legs and the bottom panel of the box must also be made from the same concrete (which again requires a form). (Picture a box form with 4 tubes, for the legs, extending upward from it.) You won't need a form release for the tube legs - they can just be left "on" the legs (but you'll need 8 tubes then for the legs). The Pipe/Legs should have end radiuses that will allow full end surface contact to the Duct/Port - (i.e. they will need a concave "dip" that matches the convex nature of the outside of the Duct/Port). A strong (super) glue should be used to conect the legs to the Duct/Port (..and obviously the legs are part of the box's bottom plate so they don't need attachment to it). Additionally, the bottom plate portion is a good place to make the plate a bit thicker and add some lead "shot".
It would also be HIGHLY advantageous if the top box panel was also made from the same concrete (though of course it will NOT be attached to any part of the Duct/Port). I'd suggest a bottom plate at least twice as thick as the "overall" side panel/baffle, and a top plate at least 1 1/2 as thick as the "overall" side panel/baffle. You should also STRONGLY consider an edge "impression" on the top and bottom plate/panels for the side panels to "fit" into (like fitting a lid on some tupperware).
So then, we only have #4 "Box Resonance" to deal with. To a fairly significant extent we have made our job that much easier with concrete bottom AND top plates/panels. The concrete is both massive and rigid, and will allow for a fairly easy (to construct) side panel structure. Remember we are concerned here with both external vibrations and internal box pressurization issues.
For the external vibration problem the best solution is a two-box solution (i.e. a box within a box with a small air-gap between them). In this instance though we don't need to create an external box and an internal box because we only have one top and bottom plate. However we will need to make 2 sets of side panels for the box (one a bit larger than the other). My recomendation here is 1/2 inch mdf that has been "fluted", (half-way into its depth), with a router (or a planer or tablesaw dado) vertically to give a "box-joint" edge profile. The inner box side panel will have its "fluting" on the outside, and the outer box side panel will have its "fluting" on the inside. The "fluting" of each will be stagered so that it "locks-in" with the other similar to a "box joint" BUT with about a few milimeters of air between the two. (note that for the additional space the “routed” channels will need to be about a 3rd again as wide as the “flutes”.) This structure should have superior rigidity while having a thinkness of only a bit more than an inch. (obviously the two boxes will be super-glued to the cement top and bottom plates - and hopefully they will "fit" into the plates themselves as suggested previously.)
For the internal pressurization problem sidewall rigidity is VERY helpful. Internal bracing is important here and rather easily done. You'll only need about 6 wooden dowels (per speaker) that will span the interior distance desired (usually 4 for the side-to-side bracing and 2 for the back-to-front bracing). The dowels should only have a diameter ONLY wide enough to accomidate the screw used to attach it to the inner "box" panel (obviously drill the screw's holes before screwing and use glue). The dowel placement should NOT be symetrical (i.e. the dowels should be staggered/offset from each other in relation to vertical and horizontal placement within the box).
In addition do NOT add box stuffing. A fiber mass will turn a good deal of the energy into heat, BUT it will also store and release energy latter AND cause air flow resistance. Instead will be going the NASA route and use silicone micro-spheres in a paint substrate.
http://www.hytechsales.com/insulating_paint_additives.html
Typically latex primer works best for the substrate..
You should be painting all interior surfaces of the box with multiple coats of this stuff EXCEPT for the honey-comb exit point of the Duct/Port, the reflector, and the aperiodic fiberglass portion of the vent. You should also paint the inner air-space portion between the two “box” side panels while still maintaing at least a millimeter of air-space. Finally the Driver’s Frame should also be painted with this stuff as should the interior of the Duct/Port where the straws are not located (i.e. right next to the driver in “front” of the magnet) – and this is the final method for reducing reflections. BTW, at least one manufacturer uses this in there reference speakers: Totem Acoustic.
If you can, it would be advantageous to paint the honey-comb portion of the straw-end that “sees” the driver, BUT you’ll need to go back after painting and essentially “bore-out” any paint in the straws themselves. (and this is simply to further reduce reflections from the “edges” of the straws near the driver.)
OK then, last but not least we have both a reflector and an aperiodic vent to make. Both will be made from the same cement (for the same reasons we used it elsewhere). The forms for the reflector can be from two large funnels (i.e. caulk the drip ends and stack the two where there is enough air-space to pour about an inch of concrete between the two). Essentially you’ll want a conical reflector that covers the surface area of the exit of the Duct/Port (so yes, they will be big funnels). If you manage to get the back end/edge nice and smooth then its fairly easy to glue the reflector to the back panel, otherwise you’ll need be a bit more creative with your adhesive. The open “front end” isn’t important because it shouldn’t be larger than the diameter of the driver’s magnet structure.
The aperiodic vent “opening” should be about 5 inches in diameter (for 50 liters) on the exterior-side, and about 7+ inches in diameter on the interior side (..depending on the thickness of the vent’s disc). A 1 inch thick piece of high density structural fiberglass foam (like the kind used in some HVAC ducting boxes – preferably dyed black) will cover, (be inset), this vent and create a resistive venting that will allow a decrease in pressure build-up in the box and lower the impeadance of the driver at fs without creating a pipe resonance or significant detrimental air flow resistance. The funnel-like chamfered edge created by the dissimilar diameters gets rid of pipe-resonance problems that sub-optimal aperiodic vents have (i.e. the scan speak and dynaudio versions) – obviously the diameter of the fiberglass will need to be 7 inches (and bevel cut on the side of the “disc” with an exacto knife for the 5 inch chamfer).
Now the concrete disc itself could have an outside diameter to match that of the Driver. Consider that you originally had 2 drivers vertically oriented on the front baffle. Now you would place the one driver where the top driver was and you COULD match the basic esthetic by placing the “vent” where the bottom driver was. Furthermore you could have both diameters esthetically the same (just about 9 ½ inches) to give you the look of two equal (9 ½ inch) diameter black circles vertically oriented on your front baffle. This disc would be glued to the two front “legs” in the box (which should be almost, but not, touching the interior of the front baffle), and again the disc will be sealed (but not attached) to the front baffle via silicone caulk. And again the surface of the contact point of the disc where it will attach to the legs should be concave to match the convex legs.
Well there is a LOT of info here to “digest”, and quite a bit of it is no doubt difficult to interpret. But I’ll try and “field” questions when they arise if you are still interested. Additionally, construction order and assembly order is also important – and again if your still interested I’ll do a follow-up on that as well.
Both the internal support legs and the bottom panel of the box must also be made from the same concrete (which again requires a form). (Picture a box form with 4 tubes, for the legs, extending upward from it.) You won't need a form release for the tube legs - they can just be left "on" the legs (but you'll need 8 tubes then for the legs). The Pipe/Legs should have end radiuses that will allow full end surface contact to the Duct/Port - (i.e. they will need a concave "dip" that matches the convex nature of the outside of the Duct/Port). A strong (super) glue should be used to conect the legs to the Duct/Port (..and obviously the legs are part of the box's bottom plate so they don't need attachment to it). Additionally, the bottom plate portion is a good place to make the plate a bit thicker and add some lead "shot".
It would also be HIGHLY advantageous if the top box panel was also made from the same concrete (though of course it will NOT be attached to any part of the Duct/Port). I'd suggest a bottom plate at least twice as thick as the "overall" side panel/baffle, and a top plate at least 1 1/2 as thick as the "overall" side panel/baffle. You should also STRONGLY consider an edge "impression" on the top and bottom plate/panels for the side panels to "fit" into (like fitting a lid on some tupperware).
So then, we only have #4 "Box Resonance" to deal with. To a fairly significant extent we have made our job that much easier with concrete bottom AND top plates/panels. The concrete is both massive and rigid, and will allow for a fairly easy (to construct) side panel structure. Remember we are concerned here with both external vibrations and internal box pressurization issues.
For the external vibration problem the best solution is a two-box solution (i.e. a box within a box with a small air-gap between them). In this instance though we don't need to create an external box and an internal box because we only have one top and bottom plate. However we will need to make 2 sets of side panels for the box (one a bit larger than the other). My recomendation here is 1/2 inch mdf that has been "fluted", (half-way into its depth), with a router (or a planer or tablesaw dado) vertically to give a "box-joint" edge profile. The inner box side panel will have its "fluting" on the outside, and the outer box side panel will have its "fluting" on the inside. The "fluting" of each will be stagered so that it "locks-in" with the other similar to a "box joint" BUT with about a few milimeters of air between the two. (note that for the additional space the “routed” channels will need to be about a 3rd again as wide as the “flutes”.) This structure should have superior rigidity while having a thinkness of only a bit more than an inch. (obviously the two boxes will be super-glued to the cement top and bottom plates - and hopefully they will "fit" into the plates themselves as suggested previously.)
For the internal pressurization problem sidewall rigidity is VERY helpful. Internal bracing is important here and rather easily done. You'll only need about 6 wooden dowels (per speaker) that will span the interior distance desired (usually 4 for the side-to-side bracing and 2 for the back-to-front bracing). The dowels should only have a diameter ONLY wide enough to accomidate the screw used to attach it to the inner "box" panel (obviously drill the screw's holes before screwing and use glue). The dowel placement should NOT be symetrical (i.e. the dowels should be staggered/offset from each other in relation to vertical and horizontal placement within the box).
In addition do NOT add box stuffing. A fiber mass will turn a good deal of the energy into heat, BUT it will also store and release energy latter AND cause air flow resistance. Instead will be going the NASA route and use silicone micro-spheres in a paint substrate.
http://www.hytechsales.com/insulating_paint_additives.html
Typically latex primer works best for the substrate..
You should be painting all interior surfaces of the box with multiple coats of this stuff EXCEPT for the honey-comb exit point of the Duct/Port, the reflector, and the aperiodic fiberglass portion of the vent. You should also paint the inner air-space portion between the two “box” side panels while still maintaing at least a millimeter of air-space. Finally the Driver’s Frame should also be painted with this stuff as should the interior of the Duct/Port where the straws are not located (i.e. right next to the driver in “front” of the magnet) – and this is the final method for reducing reflections. BTW, at least one manufacturer uses this in there reference speakers: Totem Acoustic.
If you can, it would be advantageous to paint the honey-comb portion of the straw-end that “sees” the driver, BUT you’ll need to go back after painting and essentially “bore-out” any paint in the straws themselves. (and this is simply to further reduce reflections from the “edges” of the straws near the driver.)
OK then, last but not least we have both a reflector and an aperiodic vent to make. Both will be made from the same cement (for the same reasons we used it elsewhere). The forms for the reflector can be from two large funnels (i.e. caulk the drip ends and stack the two where there is enough air-space to pour about an inch of concrete between the two). Essentially you’ll want a conical reflector that covers the surface area of the exit of the Duct/Port (so yes, they will be big funnels). If you manage to get the back end/edge nice and smooth then its fairly easy to glue the reflector to the back panel, otherwise you’ll need be a bit more creative with your adhesive. The open “front end” isn’t important because it shouldn’t be larger than the diameter of the driver’s magnet structure.
The aperiodic vent “opening” should be about 5 inches in diameter (for 50 liters) on the exterior-side, and about 7+ inches in diameter on the interior side (..depending on the thickness of the vent’s disc). A 1 inch thick piece of high density structural fiberglass foam (like the kind used in some HVAC ducting boxes – preferably dyed black) will cover, (be inset), this vent and create a resistive venting that will allow a decrease in pressure build-up in the box and lower the impeadance of the driver at fs without creating a pipe resonance or significant detrimental air flow resistance. The funnel-like chamfered edge created by the dissimilar diameters gets rid of pipe-resonance problems that sub-optimal aperiodic vents have (i.e. the scan speak and dynaudio versions) – obviously the diameter of the fiberglass will need to be 7 inches (and bevel cut on the side of the “disc” with an exacto knife for the 5 inch chamfer).
Now the concrete disc itself could have an outside diameter to match that of the Driver. Consider that you originally had 2 drivers vertically oriented on the front baffle. Now you would place the one driver where the top driver was and you COULD match the basic esthetic by placing the “vent” where the bottom driver was. Furthermore you could have both diameters esthetically the same (just about 9 ½ inches) to give you the look of two equal (9 ½ inch) diameter black circles vertically oriented on your front baffle. This disc would be glued to the two front “legs” in the box (which should be almost, but not, touching the interior of the front baffle), and again the disc will be sealed (but not attached) to the front baffle via silicone caulk. And again the surface of the contact point of the disc where it will attach to the legs should be concave to match the convex legs.
Well there is a LOT of info here to “digest”, and quite a bit of it is no doubt difficult to interpret. But I’ll try and “field” questions when they arise if you are still interested. Additionally, construction order and assembly order is also important – and again if your still interested I’ll do a follow-up on that as well.
Vil said:
For a driver with such a low mms, large sd, good x-max, and a field coil - I think the NON silver VC version at around 700 Euros is a STEAL if you need a driver that will work between 300-400 Hz and 30 Hz superbly. (the silver VC seems ludicrous to me for the extra cost..)
Provided you have the extra room for the much larger resulting loudspeaker - yes, I think it would pair up rather well with the ATC mid.
>>>>For a driver with such a low mms, large sd, good x-max, and a field coil - I think the NON silver VC version at around 700 Euros is a STEAL if you need a driver that will work between 300-400 Hz and 30 Hz superbly. (the silver VC seems ludicrous to me for the extra cost..) Provided you have the extra room for the much larger resulting loudspeaker - yes, I think it would pair up rather well with the ATC mid.
thanks for reply Scott ,
do you think I should go vented or sealed ? there will be active setup with possibility to add some EQ (linear phase )
another question - I heard those field coils become very hot after one or two hours and some driver parameters changes a lot .is it true?
thanks for reply Scott ,
do you think I should go vented or sealed ? there will be active setup with possibility to add some EQ (linear phase )
another question - I heard those field coils become very hot after one or two hours and some driver parameters changes a lot .is it true?
Vil said:
Most of the designs I've seen are bass-reflex, sometimes Onken-like. I think it really depends on HOW you make your box and what sort of overall "target" system you want. You could adhere to the basic formulation I just provided for the 215 GMF and you should get excellent results (though obviously the box will be MUCH larger). Basic eq. is almost always better added with a sealed enclosure - but IF you have a room correction eq. then "timing" can be corrected even with a vented box (though dampening really can't be). I don't think a T-Line is practical for this driver, but dipole operation may be acceptable if the bandwidth is limited and you have suitable eq. (and this is largely due to the fact that you can change T/S params via the voltage change from the field coil).
as to the heat - yes I've heard that as well (from Terry Cain on one of the smaller drivers). I'd talk to the supplier about this and get something in writing before purchasing - basically an extended warrenty (and return without a restocking charge) on this issue. I'd also try to contact owners of this driver and get their opinion:
Some of the AC customers use this driver:
http://www.audio-consulting.ch/OurCustomers.htm
perhaps AC will provide email contacts..
(LOL just looked at the price differential between the silver and non-silver version here in the US, boy we are getting "hosed" here on the non-silver.. OR we are getting a "break" on the silver version.. depending on your perspective.)
I just did some modeling with WinISD and found Lambda Acoustic TD15S and 400-2000 EXC at 9V has almost identical Frequency responses with sealed box at about 200L . I think I need to make a 250L sealed test box and try Lambda TD15S first ...
with some EQ I can easily reach 25-30hz at -3db .
what do you think ?
with some EQ I can easily reach 25-30hz at -3db .
what do you think ?
Attachments
- Status
- Not open for further replies.