3-way active system: gain structure explained for proper planning

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
It's not the potentiometer track law that is the relevant factor. It's the way gain is.....it's a relative term

As an IT guy I imagine gain to be a multiplying relation between several (mostly dB based) parameters where at the end there's also a dB-like result of these (maybe SPL of each ways).

The point is in keeping these resulting numbers proportional to the master vol control changes (keep delta) (either on the computer or on the preamp at the very front line input) so that after volume increase my overall frequency response stays just as "flat" as it was at low master volume levels.

Hmmm. Interesting, the whole thing. And then we also better know the limits, how much we need to be able to compensate with the pots.

Btw I'll have a 97dB bass, 97dB mid and 98dB tweeter section in my new setup (bass rather 94-95dB in the box according to WinISD). Bass 18" 500W RMS, 8 Ohm. Mid 120Wx4 (2 in serial then these in parallel), tweeter 100W RMS. Trying to estimate how much amp power do I need for these. Ca 60 m2 living room. The calculator couple of posts before helps a lot I think. Until amps are ready I'll drive this system with a passive xo temporarily, despite some minimal speaker sensitivity differences. Project for early 2018. I like when highs lift a very little bit anyway :D

I'll go nuts in amp selection because so many people look a bit strange when hearing this that I want to make it just for spite: Mosfet Class AB with Exicon ECF20N20-s (I omit Class D, my first thought), tube KT150 AB parallel pushpull for mids (toroid PSU and toroid OT type just to make it even more unique) and tube OTL for the tweeters. Just didn't have time to consult with those who understand electronics better than me - including you and my father too (he has built couple of tube and BJT amps long ago).


Gentlemen, so far so good, thank you everyone for the valuable inputs, I thought first this topic will be empty for days.. amazing community. I go on reading linked stuff - you can go ahead further discussing, I'll keep an eye as well.. good night for Europe, have a nice afternoon America. Cheers. :hbeat::worship::worship::worship:
 
Last edited:

ICG

Disabled Account
Joined 2007
I chose the most "difficult" way by intention to learn finally how the heck we're going to bring levels together so that even in such a different setup the whole frequency range is getting louder or quieter equally on all 3 bands when turning the volume knob at preamp side line-level.

The most simple setup would be of course:
- take 3 identical amps (6 for stereo)
- speakers have about the same sensitivity
.. and voilá !

But this is just not always the case and I would like to understand the fundamentals behind this all. DIY room-stereo, no PA system here, no DSP.

.. and voilá? Well, it does not work that way. The most important thing is, none of your drivers will be super-linear. Even perfect drivers do not behave the same in the enclosure, the baffle step will give you a level increase at a certain frequency, depending on how wide your baffle is. The speakers need equalizing, so far the passive crossover did that job for you but that's not the case anymore if you go active. And the drivers will have different roll off slopes and rise/drop of the frequency response which will result in most cases in asymmetrical acoustical slopes if you use symmetrical electrical filters. You'll likely get phase problems too, the phase of speakers shift with the frequency naturally. A delay of a mid or high channel can sometimes reduce that but that's an additional feature you have to realize in hardware.

'Just' rebuilding the passive crossover into active filters doesn't work either because of the interaction of the drivers and crossover via the impedance and phase with the amplifier. Even the 'bad' things of a passive crossover (serial resistance, ESR, losses etc) are molded into the behaviour of the crossover and cannot be replicated easily.

All in all it will be much more expensive than a DSP and you'll invest months changing the filters and measuring single and combined channels over and over again. It's unbelievable much more work you have to do than just adding a subwoofer to some already working speakers. Keep in mind, you are developing the speaker and the electronics completely new from scratch and you have to be fit in both.
If you don't want to use a DSP, the filtering and equalizing will be a lot of work, the gain structure is definitely the least of your problems. If you have to ask about the gain structure, I'm afraid you don't have the knowledge or experience what to do. I've already seen it a lot of times ending in a shipwreck and a lot of money and effort down the drain.

I would strongly suggest you'd better use a DSP, get a better result, save money and a lot of time or it would probably better to drop that plan if you don't have someone local who got a lot of time and experience. DSPs are very cheap to get nowadays, the Sure/Wondom is only 28€/4ch, the needed programmer board and cables etc, you can re-use that for every new project, you don't need the programmer board again every time. Or get a MiniDSP.
 
.. and voilá? Well, it does not work that way. The most important thing is, none of your drivers will be super-linear.

Voilá was meant for the quick and dirty way. It's already better than the same speakers with a passive crossover setup.

I'm not sure yet if I'll begin with a DSP or jump right into my first fixed config with a 4th order Linkwitz Riley crossed around 170/2800 Hz as my speakers would "like" it. bass mids highs.

So far regarding baffle simulations with exact driver parameters I'm pretty flat within 1.5dB compared to the end result which will be insanely affected by room itself and placement etc. just like for all of us normally. Somebody told me it's already a more than 6dB deviation if I turn my head in front of my speakers or stand up.. so I'm really not striving for crazy accurate setup.. just.. you know .. be decent so far I can be with the speakers themselves and then the rest is well .. luck ? Yepp. Room is room. Moving the speaker stands a bit to the wall .. or back .. left.. right .. whatever.. if I find a sweet spot - at the end my ear will decide - and regarding it's sensitivity we shouldn't even plan a flat frequency response but make a dip instead at around 3.5kHz because here we are more sensitive and for a flat listening experience we should dampen the sound around this - still it's not a common practice .. ah, enough. I just wanted to know the basics of gain-matching throughout the chain because diying stupid is stupid. :D


DSP.. :scratch:I have some precius material from nativedsd.com and some other hi-res sites, DSD, DXD, even in DSD256 resolution. Now the question is what the miniDSP does with my digital signal and in what quality compared to a Matrix Audio X-SABRE Pro and no-DSP solution. Somebody also recommended me the dbx DriveRack PA2. I think I'll give some kind of DSP a try and in worst case I can sell the stuff at eBay with some loss if I don't like the sound or if it doesn't provide me that "wow" effect. I already heard systems with "wow" sound - all passive ones.. so this is very-very depending on how I will feel first with the setup if I listen to it (gain corrected).

A lower-resolution DSP'd solution is probably still better overall than an ultimate high-res DAC without any room correction. Or not. As long as I cannot judge I'll try both ways, with and without.

Baffles will be cylinders so vertically time-aligned, I can move mids and highs to the front or to the back as much as I will and I also planned to do so, aligning each of them to the bass driver's voice coil.

I have a local who will help me but cannot answer my questions right now. This local has a new family with 2 children (just like me and my sister) so around Xmas time this local is completely unreachable via emails. We'll for sure discuss the whole topic throughout from A .. to Z .. mid 2018.
 
Last edited:

ICG

Disabled Account
Joined 2007
Voilá was meant for the quick and dirty way. It's already better than the same speakers with a passive crossover setup.

Then the passive crossover wasn't done right. Let me guess - just calculated?

I'm not sure yet if I'll begin with a DSP or jump right into my first fixed config with a 4th order Linkwitz Riley crossed around 170/2800 Hz as my speakers would "like" it. bass mids highs.

Each of these drivers got the potential for a very good sound. Active that is. Otherwise the mid driver got a way too low spl, so that's no wonder it didn't sound good passive. The 18" subwoofer and the 6" mid aren't a good fit. Your mid driver is 7dB quieter than the bass which is a lot. And that means you'll need literally 5x the power on the mid driver of the bass to get it to the same level/linear. Maybe you prefer the loudness effect but the sensational, firstly impressive effect becomes tiresome pretty quickly for normal music and non-party use. In any case, that makes your midrange the weakest of the drivers. For home use or even party still plenty of max spl (~110dB) though.

So far regarding baffle simulations with exact driver parameters I'm pretty flat within 1.5dB compared to the end result which will be insanely affected by room itself and placement etc. just like for all of us normally.

Well, you won't stay within 1,5dB once you're crossing them over, the phase WILL have some surprises for you. But I guess you think you are convinced you're already a very good speaker developer. I guess I'm out soon.

Somebody told me it's already a more than 6dB deviation if I turn my head in front of my speakers or stand up.. so I'm really not striving for crazy accurate setup.. just.. you know .. be decent so far I can be with the speakers themselves and then the rest is well .. luck ? Yepp. Room is room.

Uhm, well, you know, that's most likely not even the half he said. You might also need some training for your ears. Listening and judging speakers, even more, especally developing them, isn't that easy.

[...] if I don't like the sound or if it doesn't provide me that "wow" effect. I already heard systems with "wow" sound - all passive ones.. so this is very-very depending on how I will feel first with the setup if I listen to it (gain corrected).

I think you don't know what a DSP does. It's not an effect unit and does not create a 'wow' sound.

A lower-resolution DSP'd solution is probably still better overall than an ultimate high-res DAC without any room correction. Or not. As long as I cannot judge I'll try both ways, with and without.

A DSP is always only as good as you set it up. The setup changes so much more than the brand/model or specs of a DSP. And no, pressing 'auto EQ' does not provide good sound.

Baffles will be cylinders so vertically time-aligned, I can move mids and highs to the front or to the back as much as I will and I also planned to do so, aligning each of them to the bass driver's voice coil.

Attention, phase and delay aren't the same! It can be handled the same in a certain (close) range but that's it. With a DSP you don't have to do that either.

I have a local who will help me but cannot answer my questions right now. This local has a new family with 2 children (just like me and my sister) so around Xmas time this local is completely unreachable via emails. We'll for sure discuss the whole topic throughout from A .. to Z .. mid 2018.

Hm, if you got him, why ask him AND here too if you have to wait anyway?
 

ICG

Disabled Account
Joined 2007
For the mids you shall check #23 - the gray part and re-calculate overall sensitivity (knowing 1 midrange of this type is doing 91dB).

I want to apologize for not reading your post properly, I'm really sorry. The use of 4 of these 6" mid drivers introduce another possible problem though, depending on the cross over frequency. For high power use, the air motion transformer has to be xo'd at at least 1,5kHz, maybe a bit higher. 4 6" one directly above the next results in a massive height of 66,8cm/24". That leads to a comb filter effect and very strong vertical beaming (not to mention a possible height problem of the speaker, and, in turn, placing the tweeter accordingly). To arrange them in a 2x2 grid lessens that problem vertically but introduces it horizontally as well and also an uneven dispersion of the mids (narrow) compared to the heights (wide). So, using four of these mid drivers isn't a good solution. I'd suggest you go for a single 10" PA mid or, in this case probably much better, mid-bass driver with good spl as that solves the most problems. That will also fit much better to the 18" sub. Or use a loud 6" PA-mid driver and swap the 18" for 2 12" bass drivers, it's the same cone surface but they can play much higher.

The rest I don't want to comment. I did but oh.. wasted an hour ..

It took you an hour to write 3 lines? :eek: No, really, I'd like to help because I'm afraid you lack in understanding the tools if you're expecting a 'wow-effect' from the use of a DSP. A short term wow-effect happens if you press the loudness-button, bass exciter etc. I understand you don't want to hear that but there are a lot of conceptual errors in your plans but my critics are constructive. If you don't want me to reply further, that's fine, just say so please.
 
Well thanks. You don't know much about me so if you're willing to help I appreciate and big Danke, allerdings vorsichtig Bitte: I have a built-in quite sensitive fuse against "I bet.." or "it was badly designed" or several assumptions like this. If you're going into argumentations with assumptions you walk the same stupid path like I walk if I design my system based on assumptions too (instead of facts). ;)

It's not easy to meet my father, a math genius electric engineer since he has a new family and we live 220km away but we already planned a big "hifi & electronics" session somewhere to Feb and he reviews then the setup. (For the speakers based on sensitivity and freq curves his AMEN is here via mail, however he asked me if I want to crush down the house or what. I answered yes. He LOL'd. :) )

You know, I heard lots of audiophile systems, participated in ABX tests, countless listening tests, omg, quit that, it brings only frustration and less joy.. there's a point one has to stop and that's it. 2 of my not-that-close-anymore-friends are crazily into it and luckily the one is the typical triode-guy with 10-ish watts, the other owns a PA sound amplification company and has a 100 m2 dampened, measured, "calibrated" cellar at home with a 3-way active zillion-costing setup, uhhhh... thaaat's sound for Mahler. (Also quality vise). That big 3-way beautifully designed system was way more living for me than the Audio Note silver cabled stuff driven by 2x 2A3-s. I balance myself somewhere in-between, sadly I cannot resist rotating the volume control clockwise when I'm really into the mood so "partying" with SE Class A is no option for me.

My fuse is getting blown if you consider me the next "wannabe ultimate speaker maker" - do you own such a company or do you fear huge money loss if I accidentally find again the holy grail of speakers or what ? :D Everybody strives for the perfect speaker here but this consists of loooooots of parameters - and everybody places the weights somewhere 'cause there isn't a one-for-all ultimate design. Some are putting weights here, some there, an absolute perfect system doesn't exist. At least not for everybody. One might be quite satisfied with a certain sound and that's it. I also have a private life and don't want to spend my next years with searching for the ultimate else life really passes by and one day I wake up having still no ultimate system but lived long without joy of music itself.

Okay, back to business.

- 1st XO is planned around 150-170Hz. With these Faital mids I'm going to avoid the impedance peaks of the drivers (which is at Fs 61Hz and I'm skipping it) so overall impedance curve might look better then. Looking hours long at my spectrum analyzer with different kind of music to figure out what I really eat I think it's also a good XO point. Lower - no go due to mids, I don't want to cross them near their natural starting slope, some reserve is always good to stay on the safe side. And power.. yeaah.. XO upper - no go either, I don't want a beautiful female voice split apart so that the bass is still playing it significantly. The bass will be lifted a bit above floor but not much.. A bit it will play depending on voice freq vs. 24dB slope but I rather listen to soprans than alts. 150-170Hz is also good for sound "coming from the mids" well above floor: for a bass sound we don't know very exactly where it comes from while for mids we for sure know, the switch is somewhere at around 100-150Hz so again if I try to keep the bass out of "higher" midranges it's generally good for my stereo soundstage overall. A bit overlap will be but not that much with this XO freq so I don't expect Anne Buckely stretching into my bass drivers too and again, for Mark Knopfler sometimes ok.

Vertical dispersion will be anyway sub-optimal considering the 4 mids in a line-array and the tweeter on top of them. I plan no MTM design here by intention, read couple of articles about it, advantages, disadvantages. Not sure I can repeat all of these however, just gathering the "goodies" .. I prefer a more distributed non-MTM rather than a more perfect MTM with a smaller sweet spot regarding listening position, I won't listen always alone. Adding all of the baffle heights together this system will be just at head level somewhere with the tweeter when I'm sitting on the sofa as usual and I had no issue with my existing setup (2x12" drivers on top of eachother and then 1x mid, 1x high - same height). It's also good considering the tweeter's vertical dispersion which isn't that "balanced" like with a dome type so if I lift the speakers even higher to be with my head at middle of midrange speaker array I'll already hear somewhat less from the tweeter. Nevertheless, the very first boxes will be 1 side only and without piano finish so I can change or re-plan if needed, mdf can be chopped and heat the house if the first design turns out to be awful. I'm going to use mic, laptop, a DACT measurement system and calm weather - I don't loose much here except time - but I win experience.

- second XO point at 2800 Hz considering a balance between how much mids the tweeter has to carry to protect it (despite rated 100W RMS and factory recommended XO at 1000Hz considering a probably steep slope) and how mids are breaking apart in 0° vs. 45°degree based on freq. response + the lift at around ca. 3800Hz which I would like to avoid, this applies to lots of speakers strangely. The human ear is mostly sensible in this rather wide region too so I might change 2nd XO to 2500Hz but really no lower to protect the tweeter... That's why I tend to use a DSP, for experimenting/calibration for sure and then I'll see if it stays or not. This way I pretty much balanced human voice to the mids, except some Mark Knopflers (his voice goes into bass somewhat too and is fine for me). Violins - another favourite - also play well in the mids and highs, the Aurums are sooo sweeet, ahhh... heard them (not sure if exactly these), fell in love. Crazily detailing, tickling my ears - can't wait to check them with some fine Jean-Michel-Jarre. Yeah, I'm a ribbon (or quasi-ribbon) lover and this one comes without a transformer - no coupling capacitor needed to avoid "short" at amp side in an active design.

So in general I wouldn't let my bass drivers go into mid regions if not needed, despite pretty good capability. (I could almost design a technically somewhat stretched 2-way system with this bass drivers and tweeters but the Aurums aren't the cheapest ones to have more of them to spread load and I also don't want to leave 8 Ohms, neither further increase their sensitivity vs. bass).

Grouping the mids into 2x2 array ? Horizontal disp.. no-go.

I got your point with 4 mids vs. 1 bigger mid, also considered. It's not that easy to find midranges that have such a flat freq curve like these Faitals, on the other hand maybe this vertical line-array-setup will ruin my FR later on -
who knows. Maybe this would fit too within gain correction tresholds.. or .. dunno. Will have to check parts-express again, I spent 2-3 days looking for good-looking drivers (by specs) and that alone is a crazy thing.. I finally found the Faitals.. and .. to be honest.. there are soooo-sooo many complete line arrays and "normal" designs omiting these ideal circumstances and still sounding amazing that I'm not really afraid of this issue.

2x12" instead of 1x18" ? No-go. My girlfriend :female: told me size matters :mischiev: and I believe her. :tongue: I like strange or insane projects and here I just put my weight on "extraordinary", "unusual", "rare", "huge", "cool", "wanted it since my childhood", etc.. - it's pretty much uncommon in my country to see 18" drivers in a living room, especially not in a cylinder. Now those who visit me, will see them. Call it psychological warfare :devily: Apart from it, it challenges me too during overall speaker design. I don't like easy going stuff, that leads nowhere. Cheers. :judge:
 
Oh, for the wow effect: why do you think I "wow" because of pressing the loudness button ? You completely misunderstood it and present me as if I would be some kind of half-deaf boombox-techno-guy.

I learned piano 6 years long, alongside with music theory (solfeggio) where at the end of the last year there was a competition: the teacher played a Bach on piano 3 times, with 2 minute breaks in between. I didn't want to go 'cause I wasn't interested in solfeggio at all but I had to (parents..). But they convinced me, nothing to loose.

After the 1st play I finished my notes on the paper and reached it in. The teacher asked me if I'm sure. Yeah, sure. Who cares. Next week I went to the lesson and everybody was like "aaah, here he comes". I won the competition without the smallest failure after 1 listening only, ALL of the others had a minor issue here and there with 3 listenings. This is not what I wanted to tell about myself 'cause I'm not a king-kong type of guy shouting such into the world every day but I think I'm not deaf and when I say "wow" to music it usually means dynamics and finest details, ability to transmit feelings, etc. instead of simply wowing to a transducer at the bottom of my ***.

Let's drive this topic back to gains and enough of advice. :skull:
 
Let's start at the ear and your listening distance.

If I assume you want 110dB at your ear, with a listening distance of 2.5m from one pair of speakers:
You need 107dB from one speaker @ 2.5m (20*log (2.5m/1m)=+7.95dB
You an extra 8dB to convert that to 1m, i.e. 115dB @ 1m from the speaker.

If I assume all your drivers are rated at 95dB/W @ 1m, then you are 115dB-95db short.
You need 20dBW to get 95dB speakers up to 115dB

20dBW = 100W

Each of your 95dB/W@1m drivers needs 100W to get to 110dB at your ear when listening @ 2.5m

Since all the speakers are the same sensitivity, the gain in each channel can be the same.

If you know your maximum source voltage is 1Vac, then you need a combined gain through the crossover and power amplifier of 28.3Vac(100W into 8ohms)/1Vac = 28.3times (+29dB)
Job done for drivers that are ALL 8ohms, 95dB/W @ 1m and a 2.5m listening distance.

If you want to use a different listening distance, then the correction becomes:
20*log (listening distance /1m), instead of using +8dB

If you want to use a different driver sensitivity, then just subtract your chosen sensitivity from 95dB, to arrive at the new gain and for the new power.
i.e. you want to use a 90dB driver on one channel, then 95dB -90dB = +5dB
You need +5dB of extra gain and you need 5dB of extra power (extra power = inv log (5dB/10)= 3.16times more power)

There is a further correction you may want to implement.
Some believe that not as much power/SPL is required at very high frequencies.
You can correct for that by reducing the "at ear" SPL. Maybe choose 106db at your ear to save 4dB of power. But your gain on that channel remains, to maintain the same frequency response.
Similarly some believe that extra bass SPL is required to make that chest thumping sound.
You adjust the "at ear", or at chest SPL, by adding 3dB or 6dB to your target.
The power in the bass channel amplifiers goes up by the extra SPL you have chosen to use.
The gain in the bass channels remains the same to maintain the frequency response.
 
Last edited:

ICG

Disabled Account
Joined 2007
Well thanks. You don't know much about me so if you're willing to help I appreciate and big Danke, allerdings vorsichtig Bitte: I have a built-in quite sensitive fuse against "I bet.." or "it was badly designed" or several assumptions like this. If you're going into argumentations with assumptions you walk the same stupid path like I walk if I design my system based on assumptions too (instead of facts). ;)

It's not easy to meet my father, a math genius electric engineer since he has a new family and we live 220km away but we already planned a big "hifi & electronics" [...] his AMEN is here via mail, [...]

Yes, I've made assumptions. Yes, it was possibly wrong, ethically and factually as well to assume that much. On the other hand, the claim a 'math genius electric engineer' is automatically great at speaker development confirms exactly these prejudgements. Both got little to none benefits for such a development unless there's actual experience not in speaker building but speaker development. Both does not say anything about the ability, knowledge and experience about acoustics, speakers or the requirements. No offense, maybe he's the greatest speaker builder on earth, I can't know but these two facts about him do not say anything in that matter. At all.

You know, I heard lots of audiophile systems, participated in ABX tests, countless listening tests, omg, quit that, it brings only frustration and less joy.. there's a point one has to stop and that's it.

Yes, ofcourse. That's completely nonsense and waste of time if you can't draw anything from that. Maybe you should've participated in other listening sessions though, with ppl who are experienced in speaker development, from whom you can learn more relevant about speaker development.

My fuse is getting blown if you consider me the next "wannabe ultimate speaker maker" - do you own such a company or do you fear huge money loss if I accidentally find again the holy grail of speakers or what ? :D

Uhm, no. It would be rather the opposite because you'd be the best customer, buying things and only afterwards realizing that's not the way to go, which means, you'll buy something else. And IF you'd find the holy grail, well, then I'd sell even more because everyone will want these speakers too which means I'd sell tons of it. Unfortunately, I don't live in your area. Or your country. Or got a shop anyway. I tried not to sell you something, I tried to save you money, to prevent wrong purchases and waste of material and effort. Despite me being constructive, you don't seem to understand or appreciate that. I should probably open a shop in Hungary instead.

- 1st XO is planned around 150-170Hz. With these Faital mids I'm going to avoid the impedance peaks of the drivers (which is at Fs 61Hz and I'm skipping it) so overall impedance curve might look better then.

Uhm, no, it isn't. If you put it in any case of case :D the resonance goes up. You are a good speaker developer, shouldn't you know that? If you put it in a sealed enclosure with a Qts of (assumed ideal) 0,7 the fs/fb goes up to 77Hz. The needed volume for that would be 23,7l, for each of the Faitals that is. Together with the own volume of the drivers and the wood it's over 100l of enclosure volume already taken up. You could ofcourse chose a smaller volume, 15l each (=ca.70l with drivers and wood) would result in a still good Qts of 0,78 but a rise of the resonance up to 85Hz. Smaller isn't pretty much preferable because it starts to hunch up a tiny bit at 150-200Hz, so well within the planned reproduction range.

Looking hours long at my spectrum analyzer with different kind of music to figure out what I really eat I think it's also a good XO point. Lower - no go due to mids, I don't want to cross them near their natural starting slope, some reserve is always good to stay on the safe side. And power.. yeaah.. XO upper - no go either, I don't want a beautiful female voice split apart so that the bass is still playing it significantly.

Well, look what I've wrote above. Sorry, these drivers aren't suitable for your planned box.

- second XO point at 2800 Hz considering a balance between how much mids the tweeter has to carry to protect it (despite rated 100W RMS and factory recommended XO at 1000Hz considering a probably steep slope) and how mids are breaking apart in 0° vs. 45°degree based on freq. response + the lift at around ca. 3800Hz which I would like to avoid, this applies to lots of speakers strangely.

I already wrote it, you'll get comb filter effects and if you're xo'ing them that high, it will be absolutely heavy, practically unbearable once you start moving around, sit down/standing up. Yes, 1kHz is too low but at 3kHz are strong resonances and the impulse decay takes very long. They aren't really suitable for this kind of use. The highest you should part them is 2,5kHz and you'd already need a notch filter then.

So in general I wouldn't let my bass drivers go into mid regions if not needed, despite pretty good capability. (I could almost design a technically somewhat stretched 2-way system with this bass drivers

Uhm, no, you can't read that from the datasheet just because the lines in the graph go higher. The bass already got lively resonances at 350 and 700Hz which can be observed in the impedance and in frequency response measurements. It got there also quite the mountains in the waterfall decay measurements and distortion peaks. Don't let the well smoothed graph fool you, that's a pure subwoofer, it's not usable for anything midrange-ish.

While we're at it, it needs a rather large enclosure for a just medium-deep bass.

attachment.php


You've got a pretty big room, so 250l+100lmids=350l (+wood, which results in closer to 400l) isn't probably a big issue but is that deep enough for you?

I got your point with 4 mids vs. 1 bigger mid, also considered. It's not that easy to find midranges that have such a flat freq curve like these Faitals,

Well, don't mistake flat response with resonance-free. Almost all high spl drivers got resonances somewhere but I'd rather have a f-response peak at any time instead of a distortion and/or decay problem. The bad thing is, you can't see the latter ones in the datasheets. But I thank you for starting to understand I don't write this stuff to annoy you, I really want to help.

Maybe this would fit too within gain correction tresholds.. or .. dunno.

Uhm, no, that's not really a good fit, look what you'll get if you put it in a sealed enclosure:

attachment.php


You could ofcourse equalize it, lowshelf +3dB at ~120Hz isn't too bad but that's only possible the easy way with a DSP. But the spl is still a lot too low, it got the 95dB only from 400-700Hz, the rest is at 92dB, just one single dB more than the 6". So, no matter how you look at it, it's not a good choice either. I don't know which speakers you can get over there or what they cost, suggesting drivers is a bit hard because of that.

2x12" instead of 1x18" ? No-go. My girlfriend :female: told me size matters :mischiev: and I believe her. :tongue: I like strange or insane projects and here I just put my weight on "extraordinary", "unusual", "rare", "huge", "cool", "wanted it since my childhood", etc.. - it's pretty much uncommon in my country to see 18" drivers in a living room, especially not in a cylinder. Now those who visit me, will see them. Call it psychological warfare :devily: Apart from it, it challenges me too during overall speaker design. I don't like easy going stuff, that leads nowhere. Cheers. :judge:

Well, I guess you haven't tried to enjoy two then. Okay, I get it, that's mainly for the show but wouldn't it be a bit disappointing if you show off with such a big thing and then it doesn't go very deep?
 

Attachments

  • dayton PA460-8 BR 250l.jpg
    dayton PA460-8 BR 250l.jpg
    73.3 KB · Views: 215
  • FaitalPro 8PR200FE.jpg
    FaitalPro 8PR200FE.jpg
    80 KB · Views: 209
Last edited:
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.