Of course my bass driver's SPL is not 95dB midrange nominal at 30Hz. But I calculate with it in the box, not free air. So the driver itself follows nature with declining SPLs for same amount of voltage applied (which also means needing more energy to maintain the flat SPL here), the rest is done by the box, according to simulation. And again: just a play with numbers. (Room, box placement, damping, furniture etc will all affect the outcome).
Couple of months have passed and I just arrived to be able to understand this statement
Yeah, you're right. Assuming all drivers are all the same efficiency and amp gains are matched, for a sinusoidal, constant big level signal sweep we have about the same power need for all drivers 'cause their sensitivity is the same, impedance is the same (ignore woofer's usual impedance peak now), voltage is the same so essentially all drivers will draw the same amount of current at the same voltage level so power is the same too.
But we're talking about musical signals where towards higher freqs energy is getting less (hitting the recording mic) and so the recorded signal is getting weaker too. That's reason #1 why we can choose less powered amps for midrange and even less powered amps for tweeters (assuming their gain remains identical when speaker sensitivities are identical too).
Reason #2 why we can opt for smaller amps for tweeters is their 3dB more sensitivity which requires 1/2 power (1/2 gain or -3dB gain) if I'm correct to produce the same SPL like midrange and woofer.
So at the end I arrive at ca.
- 1/1 woofer power need, 95dB sens.
- 1/2 midrange power need (smaller signal for midranges), 95dB sens.
- 1/4 tweeter power need (smaller signal for tweeters) multiplied by another 1/2 power need due to less gain need (98dB sens.) -> 1/8 power need.
I think this would work well. At least on paper now
Let's take tweeter maximum RMS capability: 100W
Be pessimistic, don't give it more than 50W RMS: 3dB headroom is left for peaks.
According to upper recipe we arrive then at 400W RMS and 800W RMS for woofer.
And if I apply some room EQ/DSP onto the whole signal, making it stronger/weaker here and there... I quickly recognize that the planned 1250W bass amps (into 8 Ohms) is pretty ok for the task.
Depending on amp gain(s) this power, especially at woofer side, could be driven by a Line-level XLR signal via a passive preamp (volume pot and input selector only).
But if I want to go mad with the woofer and use even more watts for whatever reason I'll need to amplify the signal going to the woofer on the line level already, before the woofer. At same gain, input sensitivity for an insanely strong amp is different than "normal" and it needs to be fed by correspondingly strong voltage to get that performance out of the amp.
This would mean an active preamp, introducing one more amp stage into the woofer's amplification. Or we can build that extra preamp stage into the strong class D monoblock itself so that it can be fed by usual XLR line out signal levels just like the midrange and tweeter's amps.
However, due to Class D's EMI (although it's stated to be very low) I'd put this extra preamp stage into the separate preamp. If needed at all. Haven't checked this amp module's input sensitivity yet (it will come soon).
That's where I'm now.
How about your project ? Some progress.. ?
Cheers.
Last edited:
Hello Vortex!
I like your project as I am thinking a 4-way active system, here are my opinions:
1) tower speakers are not the perfect solution with respect to their placement and fine tuning. if you have a dedicated room, I prefer subs + satellites.
2) one mono sub (2 or 3 even better place wisely) may sound better than 2 stereo subs with respect to standing waves, far better than EQ-ing. If you roll them down below 150 Hz there is no sound directivity, so stereo subs make no acoustical sense.
3) for imaging you may need a wider distance between the left and right tweeters and midranges, while for mid-base coupling, the distance L-R has to be smaller.
So I would go one sub next to the wall (start form corner and move around), two satellites on standing posts for mids and tweeters (bookshelf-like) the farthest apart but not too close from the side walls, and two standfloors towards center for midbase.
Estetically it will not look as good as your Avalon-like towers but much easier to fine tune their placement.
4) As drivers I would use one 15 or multiple 12s for subs, 6-8" for midbase, 4 or 5" midrange (although I prefer compression drivers with horns - IMHO more suited for vocals and instruments than cone speakers), 1" tweeter or supertweeter for +10 KHz.
5) if you go with an active crossover, you set the gain for each way so you don't care too much about drivers sensitivity matching.
Keep posting about your progress!
Regards!
I like your project as I am thinking a 4-way active system, here are my opinions:
1) tower speakers are not the perfect solution with respect to their placement and fine tuning. if you have a dedicated room, I prefer subs + satellites.
2) one mono sub (2 or 3 even better place wisely) may sound better than 2 stereo subs with respect to standing waves, far better than EQ-ing. If you roll them down below 150 Hz there is no sound directivity, so stereo subs make no acoustical sense.
3) for imaging you may need a wider distance between the left and right tweeters and midranges, while for mid-base coupling, the distance L-R has to be smaller.
So I would go one sub next to the wall (start form corner and move around), two satellites on standing posts for mids and tweeters (bookshelf-like) the farthest apart but not too close from the side walls, and two standfloors towards center for midbase.
Estetically it will not look as good as your Avalon-like towers but much easier to fine tune their placement.
4) As drivers I would use one 15 or multiple 12s for subs, 6-8" for midbase, 4 or 5" midrange (although I prefer compression drivers with horns - IMHO more suited for vocals and instruments than cone speakers), 1" tweeter or supertweeter for +10 KHz.
5) if you go with an active crossover, you set the gain for each way so you don't care too much about drivers sensitivity matching.
Keep posting about your progress!
Regards!
Hi morgot9999 !
Thank you for your friendly reply and advice. Let me reflect on your points one by one for simplicity.
1) I don't have a dedicated room yet, but I'm going to build my house in the next 2-3 years. There I'll have a dedicated music & movie room in the cellar, no parallel walls at all.. but for now my living room is rather small for my planned hi-fi. I'm going to build them anyway.
My experience with subs is they still have directivity even at 100Hz. At least as long as I can hear where bass is coming from, that's the case. I'd rather use subs with a XO freq. of less than 100Hz.
As long as I don't have the final room, in the existing room I have to grow upwards to wisely use available space. And I need around 400 liters for the woofers for the desired freq. curve & deep freqs.. it would be corrected by DSP then.
For me aesthetics and woman-compatibility is also important so to be able to place them virtually almost anywhere, I'll make them like tall pyramides, with a step on the front which won't be seen by the woofer. The tall boxes will contain 1 woofer each, nothing else. Midrange section will be a much smaller tall box, again with a step on the front. The midrange box will float put on the large box on neodymium magnets, the small tweeter box will again float on the midrage box, also on neodymium magnets. So no woofer resonance is getting passed to midrange and tweeter sections.
The whole system, also the woofers' tall boxes will float on strong neo magnets. There's a company here manufacturing such magnets, nice chrome polished finish, they look great but more importantly they have a "pulling power" (or turned around: "tossing" power) of about 80kg each so I only need couple of them to make one side float. And then the other side. And all 6 amps, DAC, everything.
2) I follow the blog (vlog) of PS Audio's CEO, Paul McGowan. According to their experience with subs and rooms, subs shouldn't be placed into corners or close to walls.. some healthy distance is always better for good musicality. Placement is important but I'm not sure playing around with placement itself is enough to say OK for the woofer section. Nowaday's DSPs are very well done and I think a small correction for the bass section won't harm.
3) Interesting opinion. Actually point like source is the best, people say. An array of WMMMMT on one axis above eachother can be seen as a point source, at least in one dimension. When I pull them apart from eachother, I loose this in this one plane too. Although not a big deal due to narrow crossover range (24dB/oct) I still don't see this as a good idea. It might sound even better but for me somewhat the classic way seems to be more consistent."Seems"...
4)
Smaller subs and more of them might be a wise decision if their fs is low enough so they can be tuned within a large box to go deep. More surface.. I understand. However it doesn't look that cool (sorry) and I always wanted a biiig badass monster driver which might be capable of audiophile grade musicality too when done well. That's the challenge now. I think the 21" RCF woofers with 2kW RMS capability (each) will be a great choice for experimenting first with an unconventional PA driver and then giving the final system a prestigious look, effortless bass and plenty of headroom while playing nowhere near to the bass section's maximum when the bass amp's gain is turned down to match the midrange-tweeter amps' gains.
I had a look at tons of midrange drivers, smaller ones too but until now I cannot replace the Faital 6FE100 drivers.. they just have an outstanding freq response (very flat), they can be driven hard if needed and combined they provide the sensitivity I needed here. Another thing is resonance frequency and impedance peak, these Faitals have their resonance freq. and impedance peak good below planned crossover frequency so I completely avoid their own resonance freqs. with them.. they'll only play above that which is good, especially when driving them with a low-damping-factor tube amp which will have an easy task to drive them close to nominal and flat impedance all the way up & down, during the whole midrange frequency range. On the other hand they still have an acceptable off-axis response near the 2nd XO freq. so imaging won't be harmed much I think.
Same applies for the AMT tweeter, I don't know what a supertweeter means but the AST25120 is pretty much of such a tweeter type I think. Alternative would be a big Mundorf AMT with some more max power capability but I don't think I'll ever max out the rated 100W in the tweeter section. Still thinking but no horns and no classic dome types, I simply love the pure direct ribbon and AMT sound character. XO freq to tweeter is as early as possible (safely) to enjoy the most out of its great transient response and sound 'character'.
5)
True but I'll make all 6 monoblocks for myself as well. This takes time. A temporary passive filter will do the work and it might come handy as well later on. For this backwards compatibility I stick to almost the same sensitivities despite having an active system.
Well.. that's it. I think I give more weight to appearance than you but I still think this system will be a pleasure to listen to. Given the active setup, there're a lot of fine tune possibilities.
Thank you for your friendly reply and advice. Let me reflect on your points one by one for simplicity.
1) I don't have a dedicated room yet, but I'm going to build my house in the next 2-3 years. There I'll have a dedicated music & movie room in the cellar, no parallel walls at all.. but for now my living room is rather small for my planned hi-fi. I'm going to build them anyway.
My experience with subs is they still have directivity even at 100Hz. At least as long as I can hear where bass is coming from, that's the case. I'd rather use subs with a XO freq. of less than 100Hz.
As long as I don't have the final room, in the existing room I have to grow upwards to wisely use available space. And I need around 400 liters for the woofers for the desired freq. curve & deep freqs.. it would be corrected by DSP then.
For me aesthetics and woman-compatibility
is also important so to be able to place them virtually almost anywhere, I'll make them like tall pyramides, with a step on the front which won't be seen by the woofer. The tall boxes will contain 1 woofer each, nothing else. Midrange section will be a much smaller tall box, again with a step on the front. The midrange box will float put on the large box on neodymium magnets, the small tweeter box will again float on the midrage box, also on neodymium magnets. So no woofer resonance is getting passed to midrange and tweeter sections. The whole system, also the woofers' tall boxes will float on strong neo magnets. There's a company here manufacturing such magnets, nice chrome polished finish, they look great but more importantly they have a "pulling power" (or turned around: "tossing" power) of about 80kg each so I only need couple of them to make one side float. And then the other side. And all 6 amps, DAC, everything.
2) I follow the blog (vlog) of PS Audio's CEO, Paul McGowan. According to their experience with subs and rooms, subs shouldn't be placed into corners or close to walls.. some healthy distance is always better for good musicality. Placement is important but I'm not sure playing around with placement itself is enough to say OK for the woofer section. Nowaday's DSPs are very well done and I think a small correction for the bass section won't harm.
3) Interesting opinion. Actually point like source is the best, people say. An array of WMMMMT on one axis above eachother can be seen as a point source, at least in one dimension. When I pull them apart from eachother, I loose this in this one plane too. Although not a big deal due to narrow crossover range (24dB/oct) I still don't see this as a good idea. It might sound even better but for me somewhat the classic way seems to be more consistent."Seems"...
4)
Smaller subs and more of them might be a wise decision if their fs is low enough so they can be tuned within a large box to go deep. More surface.. I understand. However it doesn't look that cool (sorry)
and I always wanted a biiig badass monster driver which might be capable of audiophile grade musicality too when done well. That's the challenge now. I think the 21" RCF woofers with 2kW RMS capability (each) will be a great choice for experimenting first with an unconventional PA driver and then giving the final system a prestigious look, effortless bass and plenty of headroom while playing nowhere near to the bass section's maximum when the bass amp's gain is turned down to match the midrange-tweeter amps' gains. I had a look at tons of midrange drivers, smaller ones too but until now I cannot replace the Faital 6FE100 drivers.. they just have an outstanding freq response (very flat), they can be driven hard if needed and combined they provide the sensitivity I needed here. Another thing is resonance frequency and impedance peak, these Faitals have their resonance freq. and impedance peak good below planned crossover frequency so I completely avoid their own resonance freqs. with them.. they'll only play above that which is good, especially when driving them with a low-damping-factor tube amp which will have an easy task to drive them close to nominal and flat impedance all the way up & down, during the whole midrange frequency range. On the other hand they still have an acceptable off-axis response near the 2nd XO freq. so imaging won't be harmed much I think.
Same applies for the AMT tweeter, I don't know what a supertweeter means but the AST25120 is pretty much of such a tweeter type I think. Alternative would be a big Mundorf AMT with some more max power capability but I don't think I'll ever max out the rated 100W in the tweeter section. Still thinking but no horns and no classic dome types, I simply love the pure direct ribbon and AMT sound character. XO freq to tweeter is as early as possible (safely) to enjoy the most out of its great transient response and sound 'character'.
5)
True but I'll make all 6 monoblocks for myself as well. This takes time. A temporary passive filter will do the work and it might come handy as well later on. For this backwards compatibility I stick to almost the same sensitivities despite having an active system.
Well.. that's it. I think I give more weight to appearance than you but I still think this system will be a pleasure to listen to. Given the active setup, there're a lot of fine tune possibilities.
Last edited:
Hi Vortex
I started the building process but it's a long way to go. My room is planned 7x4.4.x2.3m in the basement, with no wife factor. Not interested in home theater, only stereo. Plan it wisely, from the begging, including speaker placement, rack, power outlets and so on.
If you hear the location of the sub it means that it excites a room mode or it's volume is too high.
Separate the subs boxes to the WMT or MT to have the placement flexibility. Monstrous towers are made to impress and bring large profits.
Even if you stack the mids to woofers, the cabinets should be stiff enough that will not resonate, so you don't really need the vibration decoupling between the boxes. Is much worse to hear the box resonance than care about the transmitted vibration from one box to the other. (I have a pair of diy 30L, ringing at 200 Hz at high volume, cause: no bracings).
Keep in mind, the air pressure energy must be kept inside the box, and not transmitted outside the cabinet.
Floating boxes on neo magnets. I think it's overkill, put your money in something else. Soft vibration dampers tuned low enough (15 Hz?) should do the job (check Sorbothane, Wacker, Vibracousic for inspiration)
Two 21's indoor? Hopefully you want them horn loaded
Smaller 12-15's could be much faster in transients than 18-21's, because of lower cone mass. It depends on the music genre, think to reagge vs. dubstep.
I like Faitals too
AMTs are too expensive for my pocket.
What topology you think for the monoblocks? Maybe we build them together ^_^.
I follow Paul McGowan too, check also Steve Guttemberg, the Audiophiliac.
My opinion, any commercial speaker has to sell, so it's compromised from the beginning: looks, cost, size, market taste. How many 400L subs have you seen? The same for active systems, they are acknowledged in performance, but not a commercial success (check Lenard Audio).
Regards
I started the building process but it's a long way to go. My room is planned 7x4.4.x2.3m in the basement, with no wife factor. Not interested in home theater, only stereo. Plan it wisely, from the begging, including speaker placement, rack, power outlets and so on.
If you hear the location of the sub it means that it excites a room mode or it's volume is too high.
Separate the subs boxes to the WMT or MT to have the placement flexibility. Monstrous towers are made to impress and bring large profits.
Even if you stack the mids to woofers, the cabinets should be stiff enough that will not resonate, so you don't really need the vibration decoupling between the boxes. Is much worse to hear the box resonance than care about the transmitted vibration from one box to the other. (I have a pair of diy 30L, ringing at 200 Hz at high volume, cause: no bracings).
Keep in mind, the air pressure energy must be kept inside the box, and not transmitted outside the cabinet.
Floating boxes on neo magnets. I think it's overkill, put your money in something else. Soft vibration dampers tuned low enough (15 Hz?) should do the job (check Sorbothane, Wacker, Vibracousic for inspiration)
Two 21's indoor? Hopefully you want them horn loaded
Smaller 12-15's could be much faster in transients than 18-21's, because of lower cone mass. It depends on the music genre, think to reagge vs. dubstep.
I like Faitals too
AMTs are too expensive for my pocket.
What topology you think for the monoblocks? Maybe we build them together ^_^.
I follow Paul McGowan too, check also Steve Guttemberg, the Audiophiliac.
My opinion, any commercial speaker has to sell, so it's compromised from the beginning: looks, cost, size, market taste. How many 400L subs have you seen? The same for active systems, they are acknowledged in performance, but not a commercial success (check Lenard Audio).
Regards
Hi, interesting all what you write here.. a pleasure to read
What I can quickly relate to.. - sitting at work
1. Neo magnets. Cheap. And the effect is fancy. Why not ?
2. Box for bass: plywood, no mdf. Stiffened from inside. Too big to make them out of concrete. However, the significantly smaller midrange and tweeter boxes might be made of concrete while weight is still manageable.
3. Two 21" woofers, in front facing reflex towers. Nothing special.
Smaller ones and transients: not necessarily. Thiele-Small tells you the magic, bigger cone is compensated with stronger magnet and here we go.
4.Class D for bass drivers, custom parallel pushpull class AB tube for mids (KT150, fully balanced, differential amp), Exicon Mosfet blocks for tweeters, based on the The End Millenium's quite basic design.
And some crazyness, like:
- driving class D modules with linear PSU
- driving tube amp with smps
What I can quickly relate to.. - sitting at work
1. Neo magnets. Cheap. And the effect is fancy. Why not ?
2. Box for bass: plywood, no mdf. Stiffened from inside. Too big to make them out of concrete.
However, the significantly smaller midrange and tweeter boxes might be made of concrete while weight is still manageable. 3. Two 21" woofers, in front facing reflex towers. Nothing special.
Smaller ones and transients: not necessarily. Thiele-Small tells you the magic, bigger cone is compensated with stronger magnet and here we go.
4.Class D for bass drivers, custom parallel pushpull class AB tube for mids (KT150, fully balanced, differential amp), Exicon Mosfet blocks for tweeters, based on the The End Millenium's quite basic design.
And some crazyness, like:
- driving class D modules with linear PSU
- driving tube amp with smps
Dear all, my big project is delaying not only due to corona situation now but first I need to build two smaller systems (1 is being built already, 2x KT88 SE amps)..
Nevertheless, I'm still keeping this big project of mine on radar.
I got some interesting feedbacks from some audiophile listener folks like I should't compare PA midranges to Scan-Speak, Seas, Vifa etc.. (suggesting me I might choose another midranges than the chosen Faital 6FE100 ones). Of course they couldn't argue from technical point of view but you know, it still made me think a little bit about the whole thing.
What're your thoughts about it ? We as DIY-ers measure a lot and go rather (not only!) by specs than being pure empirical (meaning: by-ear mostly, even if SPL graphs aren't that beautiful, etc).
Tbh I couldn't find real measurements for loudspeaker distortion, which is apparently a topic on most forums when talking about using PA stuff in room kits. I don't think PA shall be any worse just by itself because it's PA. By construction it might be worse or even better, who knows, but I don't like to make blind assumptions without real measurements.
So what do you think, shall I replace the Faital 6FE100 drivers with Aurum Cantus AC-130F1 ones ?
Faital Pro Aurum Cantus
Both have great specs, however
- I don't see off-axis graphs for the Aurum Cantus
- on the other hand they're a little bit smaller than the Faitals and cone breakup arrives a tiny bit later
- Faitals sustain 100W RMS while ACs 60W only
- ACs have a tiny little bit less sensitivity of 89dB average (still perfect for my application to arrive to woofer level 95-ish SPL with a parallel connection)
Just for recap, intended use stays the same: 4pcs for 1 side, 2-2 in series then these lines in parallel. Active 4th order @ 125Hz/2kHz, neighbours are 21" RCF on the bottom and Aurum Cantus AMT on the top, all active (gain issues don't matter).
Can it be that I'm better off with these ACs despite somewhat less overall max power in midrange section ? (Bass 2kW max, tweeter 100W max). I intend to use normal-loud listening levels from around 3-4 meters, not more, no disco. (huge bass is just for an absolutely satisfying reserve if needed, especially when watching movies on the stereo).
To put it simple: would you trade some power for some less distortion ? (Where I'm actually not sure if these ACs have less distortion at all, but let's assume).
Nevertheless, I'm still keeping this big project of mine on radar.
I got some interesting feedbacks from some audiophile listener folks like I should't compare PA midranges to Scan-Speak, Seas, Vifa etc.. (suggesting me I might choose another midranges than the chosen Faital 6FE100 ones). Of course they couldn't argue from technical point of view but you know, it still made me think a little bit about the whole thing.
What're your thoughts about it ? We as DIY-ers measure a lot and go rather (not only!) by specs than being pure empirical (meaning: by-ear mostly, even if SPL graphs aren't that beautiful, etc).
Tbh I couldn't find real measurements for loudspeaker distortion, which is apparently a topic on most forums when talking about using PA stuff in room kits. I don't think PA shall be any worse just by itself because it's PA. By construction it might be worse or even better, who knows, but I don't like to make blind assumptions without real measurements.
So what do you think, shall I replace the Faital 6FE100 drivers with Aurum Cantus AC-130F1 ones ?
Faital Pro Aurum Cantus
Both have great specs, however
- I don't see off-axis graphs for the Aurum Cantus
- on the other hand they're a little bit smaller than the Faitals and cone breakup arrives a tiny bit later
- Faitals sustain 100W RMS while ACs 60W only
- ACs have a tiny little bit less sensitivity of 89dB average (still perfect for my application to arrive to woofer level 95-ish SPL with a parallel connection)
Just for recap, intended use stays the same: 4pcs for 1 side, 2-2 in series then these lines in parallel. Active 4th order @ 125Hz/2kHz, neighbours are 21" RCF on the bottom and Aurum Cantus AMT on the top, all active (gain issues don't matter).
Can it be that I'm better off with these ACs despite somewhat less overall max power in midrange section ? (Bass 2kW max, tweeter 100W max). I intend to use normal-loud listening levels from around 3-4 meters, not more, no disco. (huge bass is just for an absolutely satisfying reserve if needed, especially when watching movies on the stereo).
To put it simple: would you trade some power for some less distortion ? (Where I'm actually not sure if these ACs have less distortion at all, but let's assume).
Last edited:
(Alternatively I could go with 4x AC-130/50CKs but then I'd need to handle that hump right at 1.5k. Power looks good, 150W).
Hi Vortex
PA drivers are made to play Loud. They perform best at 75-100% of their power rating delivering concert level SPL's (+100 dB), while hifi drivers can be used for midnight listening (50-60 db).
Not sure about this 6", but a 12-15" bass driver at 10% of it power rating will do no bass notes. The explanation comes from the suspension stiffness (spider+surround) and the cone travel.
Just take a look at the Vas or Cms of your selection:
6FE100 deforms 0.48 mm for 1N force applied.
AC-130F1 for the same 1N will do 2.48 mm.
For a low cone travels, the air will not get pressurized enough to produce the SPL you are looking.
Another difference is the mass of the moving mass, Mms:
14g for Faital vs 9.7 g for Aurum.
As Force = Mass * Acceleration, a lighter cone will accelerate faster than a heavier cone => better transient response!
Take care about the -6dB loss when doubling the distance! This applies to free space (no walls). In a relatively small room (large rooms = theater hall) unless you have 100% absorption, the sound will bounce back. Pop a baloon 1m vs your-listening-distance and measure with a spl meter (phone may work).
PA drivers are made to play Loud. They perform best at 75-100% of their power rating delivering concert level SPL's (+100 dB), while hifi drivers can be used for midnight listening (50-60 db).
Not sure about this 6", but a 12-15" bass driver at 10% of it power rating will do no bass notes. The explanation comes from the suspension stiffness (spider+surround) and the cone travel.
Just take a look at the Vas or Cms of your selection:
6FE100 deforms 0.48 mm for 1N force applied.
AC-130F1 for the same 1N will do 2.48 mm.
For a low cone travels, the air will not get pressurized enough to produce the SPL you are looking.
Another difference is the mass of the moving mass, Mms:
14g for Faital vs 9.7 g for Aurum.
As Force = Mass * Acceleration, a lighter cone will accelerate faster than a heavier cone => better transient response!
Take care about the -6dB loss when doubling the distance! This applies to free space (no walls). In a relatively small room (large rooms = theater hall) unless you have 100% absorption, the sound will bounce back. Pop a baloon 1m vs your-listening-distance and measure with a spl meter (phone may work).
Last edited:
Hi morgot999. Thank you for your response.
I have meanwhile still some unclarities in the project, regarding the midrange section. But this is mainly caused by the mismatches between the Faital drivers' measured and manufacturer-presented data.
response curve by manufacturer vs a measurement in real (PDF, scroll down)
In this real life measurement the SPL curve of the Faital is rising, instead of being flat as stated by manufacturer. (I mean overall tendency by 'flat', not local peaks or dips). I think to maintain a good SPL across the wanted operational region it's crucial for every system that the drivers don't have a rising (or falling) SPL curve. This is btw nothing compared to the B&G Neo10. Sometimes I don't understand people who build with B&G planar drivers, like this one here. Having a look at the SPL curve, I could use such a driver (or more of them for power splitting, for hell of a money) under following statements/circumstances only:
- 1. to get a flat SPL curve, a single series cap is needed which acts as a 1st order -6dB/oct. filter, compensating for the nonstop-rising SPL of the driver itself, this way effectively reducing the final produced SPL to near flat. Value of the cap depends on the frequency where it would begin to work, I'd set it at around 350-400Hz
- 2. via this series cap, SPL is made flat-ish and overall average sensitivity settles at around 84-85dB
A head scratching task for those who want no single passive element in the signal path while trying to avoid DSP as well. (For DSP+electronic crossover people, no cap would be needed and that's it, everything would be modified by the DSP itself probably).
Anyway, back to the Faitals, the more stiff suspension isn't that radically reflected by SPLs what the numbers suggest. What I like in the Faitals is: I can place them very close to eachother, due to their romboid shape. This way the center-to-center distance is minimized and I can place them as a 2x2 matrix rather than on top of eachother. This way they'll sound more like 1 bigger midrange driver, at least for the most of the intended range (125-150Hz -> 2kHz). Just above these frequencies would they split and sound like 4 individual drivers (due to lambda decrease).
I understand your theory with stiffness but if they wouldn't produce that much SPL at low power, then 1W/2.83V/1m measurements would reflect this as well, to a bigger extent than what has been measured on the 2nd graph I linked.
The ACs are round and I'm still thinking about how to place them close to eachother in one group. General rules is placing them above eachother because then we have a point-like source in horizontal domain at least but when their frequencies add together and create a homogenous "virtual-midrange" anyways, I think it doesn't matter if I place them next to eachother horizontally too.. so in 2x2. But I might be wrong as well, just thinking loud.
Anyway, I think my system will employ a DPS and crossover by miniDSP, correcting the bass region only and trying to leave mid-high range intact. So good planning of mid-high section is important for me. (Although room acoustic issues would produce big peaks and/or dips anyway so a Dirac Live may come handy).
Feel free to write your opinion if different of all of the above.. and .. thanks again
I have meanwhile still some unclarities in the project, regarding the midrange section. But this is mainly caused by the mismatches between the Faital drivers' measured and manufacturer-presented data.
response curve by manufacturer vs a measurement in real (PDF, scroll down)
In this real life measurement the SPL curve of the Faital is rising, instead of being flat as stated by manufacturer. (I mean overall tendency by 'flat', not local peaks or dips). I think to maintain a good SPL across the wanted operational region it's crucial for every system that the drivers don't have a rising (or falling) SPL curve. This is btw nothing compared to the B&G Neo10. Sometimes I don't understand people who build with B&G planar drivers, like this one here. Having a look at the SPL curve, I could use such a driver (or more of them for power splitting, for hell of a money) under following statements/circumstances only:
- 1. to get a flat SPL curve, a single series cap is needed which acts as a 1st order -6dB/oct. filter, compensating for the nonstop-rising SPL of the driver itself, this way effectively reducing the final produced SPL to near flat. Value of the cap depends on the frequency where it would begin to work, I'd set it at around 350-400Hz
- 2. via this series cap, SPL is made flat-ish and overall average sensitivity settles at around 84-85dB
A head scratching task for those who want no single passive element in the signal path while trying to avoid DSP as well. (For DSP+electronic crossover people, no cap would be needed and that's it, everything would be modified by the DSP itself probably).
Anyway, back to the Faitals, the more stiff suspension isn't that radically reflected by SPLs what the numbers suggest. What I like in the Faitals is: I can place them very close to eachother, due to their romboid shape. This way the center-to-center distance is minimized and I can place them as a 2x2 matrix rather than on top of eachother. This way they'll sound more like 1 bigger midrange driver, at least for the most of the intended range (125-150Hz -> 2kHz). Just above these frequencies would they split and sound like 4 individual drivers (due to lambda decrease).
I understand your theory with stiffness but if they wouldn't produce that much SPL at low power, then 1W/2.83V/1m measurements would reflect this as well, to a bigger extent than what has been measured on the 2nd graph I linked.
The ACs are round and I'm still thinking about how to place them close to eachother in one group. General rules is placing them above eachother because then we have a point-like source in horizontal domain at least but when their frequencies add together and create a homogenous "virtual-midrange" anyways, I think it doesn't matter if I place them next to eachother horizontally too.. so in 2x2. But I might be wrong as well, just thinking loud.
Anyway, I think my system will employ a DPS and crossover by miniDSP, correcting the bass region only and trying to leave mid-high range intact. So good planning of mid-high section is important for me. (Although room acoustic issues would produce big peaks and/or dips anyway so a Dirac Live may come handy).
Feel free to write your opinion if different of all of the above.. and .. thanks again
Oh, to moving mass: I think it's directly reflected in frequency response.
As long as cone 'A' is a bit heavier then cone 'B', if magnetic forces, suspension etc. are all the same, they will both have the same transient response speed within the region where both produce the same SPL. The only effect of a heavier cone would be earlier mid- or high-frequency rolloff, but below the rolloff where both produce equal SPLs they're the same. Transients have more to do with cone mass, suspension stiffness and magnet force combined I think. (But just assuming).
As long as cone 'A' is a bit heavier then cone 'B', if magnetic forces, suspension etc. are all the same, they will both have the same transient response speed within the region where both produce the same SPL. The only effect of a heavier cone would be earlier mid- or high-frequency rolloff, but below the rolloff where both produce equal SPLs they're the same. Transients have more to do with cone mass, suspension stiffness and magnet force combined I think. (But just assuming).
Last edited:
Interresting thread, subscribed !
Would like as you to try cheapiest way to active filtering + EQ/DRC, tri-amp class D or single output transistor class A amps a la Hiraga Nemesis (less interrested by the choice of the speaker but if on a budget)
My main concern is about to get a good digital device, not sure the cheapest MiniDSP stuffs sounds good (the most expensive with AK chips perhaps, but ...expensive for a try)
Or my next build will be segmented ESL...
Just a personal inputt about the first post : some of the music is asking a low fequency range : 30 hz could be a -F3 targett.
Would like as you to try cheapiest way to active filtering + EQ/DRC, tri-amp class D or single output transistor class A amps a la Hiraga Nemesis (less interrested by the choice of the speaker but if on a budget)
My main concern is about to get a good digital device, not sure the cheapest MiniDSP stuffs sounds good (the most expensive with AK chips perhaps, but ...expensive for a try)
Or my next build will be segmented ESL...
Just a personal inputt about the first post : some of the music is asking a low fequency range : 30 hz could be a -F3 targett.
Well, welcome !
Can't believe I opened the first post more than 3 years ago.. since then I've learned sooooo much from here in this community and a lot of members omg.. this is the best DIY forum on Earth right now - at least for audio related topics
To frequencies: I intend to tune the box to fs=28Hz if I'm not mistaken, I don't have WinISD in front of me right now..
The whole big project will be slightly modified.. I accepted the need for some modifications since the very original plans and I also want to fasten up things a bit.
- I abadoned my plans of driving the midrange section with tube class AB monoblocks. Just doesn't make sense and I'm not sure if I can achieve a reasonable signal-to-noise ratio at around 200-300W-ish power without the constant "hiss" (despite being midrange drivers). I would rather employ a Hypex Class D, most probably the NCore 400.
- Same considerations for the tweeters. A Class AB mosfet amp would still be up to the task but when today's top of the line Class D-s are sooo great, why not use them ? Again, the NCore 400 (at somewhat less power, to match gains)
So I'm going an all Class D approach and live the tube amps in my workroom for the PC.
To miniDSP: the SHD measures actually pretty good. Not perfect, but it plays in the blue league here.
Here is the full review if interested. click
Can't believe I opened the first post more than 3 years ago.. since then I've learned sooooo much from here in this community and a lot of members omg.. this is the best DIY forum on Earth right now - at least for audio related topics
To frequencies: I intend to tune the box to fs=28Hz if I'm not mistaken, I don't have WinISD in front of me right now..
The whole big project will be slightly modified.. I accepted the need for some modifications since the very original plans and I also want to fasten up things a bit.
- I abadoned my plans of driving the midrange section with tube class AB monoblocks. Just doesn't make sense and I'm not sure if I can achieve a reasonable signal-to-noise ratio at around 200-300W-ish power without the constant "hiss" (despite being midrange drivers). I would rather employ a Hypex Class D, most probably the NCore 400.
- Same considerations for the tweeters. A Class AB mosfet amp would still be up to the task but when today's top of the line Class D-s are sooo great, why not use them ? Again, the NCore 400 (at somewhat less power, to match gains)
So I'm going an all Class D approach and live the tube amps in my workroom for the PC.
To miniDSP: the SHD measures actually pretty good. Not perfect, but it plays in the blue league here.
Here is the full review if interested. click
Thanks for the links...
You know projects... sometimes last few years. Just finished to refurbish an old cabinet and filter for a friend that asked 8 months ago and also just refurbished a ittle Proton amp thanks the good advices here from Mooly fellow and an ooooold thread about the same amp. I must say I think about going active filter and tubes since 10 years... but you know I'm always more involved by listen to music than anything
So I have to agree about what you're saying and let's be cool with the time spent, it's hobby for most of us
You know projects... sometimes last few years. Just finished to refurbish an old cabinet and filter for a friend that asked 8 months ago and also just refurbished a ittle Proton amp thanks the good advices here from Mooly fellow and an ooooold thread about the same amp
. I must say I think about going active filter and tubes since 10 years... but you know I'm always more involved by listen to music than anything So I have to agree about what you're saying and let's be cool with the time spent, it's hobby for most of us
Ahh, you're right.. but when I think about I was 37 at the time of the 1st post and now I'm approaching 41 in May, omg. Time is just passing by so cruelly and I should already listen to this big system instead of dreaming about it. Many years ago I would have said I don't have the money for the build, parts, etc. Sadly it's not about money now, although it isn't a cheapo system but still just moderate on costs... but I miss the time and place for that. Due to corona situation we couldn't move out of our flat (into a family house) and of course in a normal flat I'm not going to bother our neighbours with bass of several 100 Watts. But you know... DIY is DIY .. so I just managed to get some tubes and building my small & simple headphone amplifier based on others' suggestions on this forum. That's the only DIY happiness nowadays. I hope strict regulations are over soon and the real estate market is getting moving as well, so we can sell, we can buy ... Well.. yeah. Life
Last edited:
lol, otl tube headphone amp is on the waiting list too !
That's the way it is with hobby and experience, it comes with years and you become eventually too much old to profit at the max... deaf in ours.
One never has to forgett to enjoy what he has first, cofins are just a tool not an end... uh, they are but it's already too late then ! Today the knowledge seems more important than the tools to enjoy whatever tools I had at the end of the day to listen to music ... this day you realise indeed years are spending fast
That's the way it is with hobby and experience, it comes with years and you become eventually too much old to profit at the max... deaf in ours.
One never has to forgett to enjoy what he has first, cofins are just a tool not an end... uh, they are but it's already too late then
! Today the knowledge seems more important than the tools to enjoy whatever tools I had at the end of the day to listen to music ... this day you realise indeed years are spending fast
Then you might want to check this thread click
I'm building the version of forum member kodabmx, post #2. Tubes are already here, power transformers as well, I'm in the process of buying all the small parts. Maybe today I can gather all little pieces, capacitors, resistors, diode bridges, internal cabling, switch, volume control, etc.... Only the small box is missing, the chassis. It's a quite easy little project, still causing a lot of enjoyment while building. And then hopefully while listening to it too.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.