Active 3-way open baffle here plus subwoofer system.
The OB consists of Fountek Neo x 3.0, Mark Audio Alpair 12pw, and Alpha 15A. I'm crossing with a Minidsp 4x10 to two Rotel six channel amps. Each amp is bridged to 150x3. So I effectively have two 450w monoblocks pushing each speaker.
Speaker drivers, parts, amps, and minidsp total around $1500. I sold my $9K B&W Diamonds and amps because they weren't even close in performance.
The OB consists of Fountek Neo x 3.0, Mark Audio Alpair 12pw, and Alpha 15A. I'm crossing with a Minidsp 4x10 to two Rotel six channel amps. Each amp is bridged to 150x3. So I effectively have two 450w monoblocks pushing each speaker.
Speaker drivers, parts, amps, and minidsp total around $1500. I sold my $9K B&W Diamonds and amps because they weren't even close in performance.
Except when they're not, as I already explained above.
Are you serious? I didn't make your files sound dull - you did! You're the one who slathered them with EQ, dude!
I might agree with that if it were accurate. Had you read a bit more carefully, you would have noticed that this 3-4 dB curve begins not in the HF region but at 32 Hertz. It's a slope of roughly 0.5 dB per octave, spread over seven octaves.
But why would I want to do that? I've already explained that this setup sounds wonderful with nearly any source material I listen to, without having to micro-manage individual EQ settings for each title, even once!
I understand your position on this. And I also understand the appeal of a pet theory - god knows I've had my share. But your approach, with all these myriad, convoluted individual EQ corrections seems misdirected to me. You've also made some claims on your own website and elsewhere regarding the way commercial recordings are made and mastered that are unsubstantiated at best, taking jabs at industry experts who collectively have likely forgotten more than you or I will ever know about this stuff. It seems to me like you're trying to bend the rest of the universe to fit your straight-line spectrum theory. Last but not least, I find the condescending, infallible tone in your replies to me to be just plain annoying.
My own theory is that the more accurately I can make my system perform, the less trouble I have with variations in software. Conversely, when I find myself needing to adjust for different source material, it tells me the hardware isn't properly dialed in yet. My latest home hardware iteration is bearing this out pretty strongly, and I experienced the same thing with a car system some time back. But that's just me.
-- Jim
I can't draw the same conclusion at the end. If I were to average the EQ for all I listen to I'd still have the curve. It's been a similar curve with other speakers. I think that's what makes it difficult.
There's every DIY guide telling us to make it flat, friends preferring splashy treble (a red herring) and all recordings sounding incorrect.
Paper full-range drivers are a milestone. Here is where sense departed and I became confused when people started saying adding a tweeter is a good thing.
It's no wonder a person would question themself, their speaker building abilities, and perhaps even the rest of the world.. who maybe didn't know any better?
So, my 'house curve' is integrated into my crossover and my EQ is held optional and defaults to flat.
Last edited:
I'm beginning to get a glimpse of the bigger picture here as well. I'm sorry I didn't figure it out sooner.
Take care.
-- Jim
I've made an active system with two 2-way TQWP enclosures and a sub.
The woofers are salvaged from a pair of german 80s speakers and are cut at 2.3 kHz and 100Hz (to eliminate any low end load). They have an Fs of 100Hz so a sub was needed.
Tweeters are HiVi K1.
No additional EQ-ing is implemented since it can easily be done from the source (which is most of the time a computer).
The sub takes care of the rest of the spectrum, below 100Hz.
The first crossover was the ESP 18db/oct, but then I upgraded to the ESP LR 24db/oct.
The woofers are salvaged from a pair of german 80s speakers and are cut at 2.3 kHz and 100Hz (to eliminate any low end load). They have an Fs of 100Hz so a sub was needed.
Tweeters are HiVi K1.
No additional EQ-ing is implemented since it can easily be done from the source (which is most of the time a computer).
The sub takes care of the rest of the spectrum, below 100Hz.
The first crossover was the ESP 18db/oct, but then I upgraded to the ESP LR 24db/oct.
Wow, that's actually news to me. The following is a bit off-topic for the thread (or maybe not...), but very much on that particular topic:
I arrived at a flat sloping curve purely by empirical means. First, I didn't change anything above 100-150 Hz, preferring to restore the missing bass octaves only. Then I ran into issues with balance, so I started to tilt the curve up and down on the right side to get a treble-bass balance (which varies a few dB depending on instrumentation and depth of frequencies played in each succeeding octave. Then I started to fill in missing holes in the bass-midrange-treble octaves and rebalancing. (I use Spectrogram log(df) views now to look at the background noise levels to help set the overall slope and HF EQ levels back to "flat".)
Then I started to find music tracks that were pristine without any real changes required at all (Steely Dan's Aja, MFSL version, is one of those discs that comes to mind, but is certainly not limited to that one example). They all were straight/decreasing ~5.5 dB/octave.
I experimented with different curve shapes and found that the monotonically decreasing straight line target FR curve sounded more natural--comparing A-B at every step--than any non-straight curve, assuming full instrumentation that covers most of the 10 audible octaves (20-20K Hz).
I kept trying different curves and trying to retain the non-straight curves on the discs, then I found out that the flat curve always sounded better. There are only a handful of tracks--less than 10 or so--that do not follow that rule of thumb out of the 5000+ tracks that I've remastered to date. That's a fairly strong empirical relationship in my experience. Only the particular slope for each track changes, and where the LF roll=off should occur--from no roll off at all down to below 20 Hz for organ performance, to bass at 40 Hz (normal 4-string basses: upright or electric), to piano at about 40-70 Hz (depending on composition) and a capella and other solo instrument tracks that roll off at or above 150 Hz, etc., etc.
So none of this is really theory, but only observations based on a large number of trials and empirical data gathering. The observations that it was a straight led me to think about the reasons why, so I observed that it's in log-log space and realized that there also was a correlation to pink noise spectra, albeit not exactly at the same slope curve (my rule of thumb is 5.5 dB/octave, or 18 dB decade +/- 3-4 dB over the entire audible bandwidth (yes, 3 or 4 dB across the entire audible bands for a house curve is a lot, I've found), and that is something that has borne itself out.
Some of the early tracks that I did exhibit a learning curve, but now, not so much (and thousands of remastered tracks later). I should go back and delete a few of those early EQ curves to avoid the situations like I just experienced, above.
Returning you to your regularly scheduled programming.
Chris
Last edited:
It is? Yes, I could be on the wrong thread it's easy to forget.
The joy for me is finding out that we've all been there and it's not just me who doesn't prefer the flat and uniform SPL/f.
I can relate to that. I'm using a 3-way active XO setup using two FirstWatt B4s. Speakers are B&W Matrix 802s with the LF crossover removed, and a JL F112 sub. I initially tried to adjust for a flat response but I liked better a little + 2 to +3 bump in the LF. No problem on the HF. Very happy indeed.
Active is definitely the way to go.
In my opinion, crossover below 1 kHz should always be active - quality passive components for these frequencies get big and expensive fast and it's simply more cost effective to go active (good 20...50 W per channel stereo amp modules are what, 20...50 € these days? A single low DCR aircore coil can cost more). Another reason to consider active is, like me and my fellow audio hobbyists often find, an amp good for bass is not so good for midrange/treble, and vice versa. Active solution can get the best out of both amps.
Crossovers at higher frequencies are a mixed bag - simple 2. order passive crossover at 3 kHz and above can get away with small and cheap components. But if you need higher orders and EQ, an active solution can be better both sonically and for wallet.
As for me, for now I've got an analog active crossover for bass and mid separation and passive for mid/high. In my experience, mid to tweeter crossover often needs asymmetrical slopes, non-textbook Q-s, frequency over/underlap and other trickery to get it right and a simple 4. order L-R analog active won't do the trick. On the other hand, DSP would and it's on my upgrade list.
Crossovers at higher frequencies are a mixed bag - simple 2. order passive crossover at 3 kHz and above can get away with small and cheap components. But if you need higher orders and EQ, an active solution can be better both sonically and for wallet.
As for me, for now I've got an analog active crossover for bass and mid separation and passive for mid/high. In my experience, mid to tweeter crossover often needs asymmetrical slopes, non-textbook Q-s, frequency over/underlap and other trickery to get it right and a simple 4. order L-R analog active won't do the trick. On the other hand, DSP would and it's on my upgrade list.
well I did not finish my OBs (shown previously); that project is currently on hold 
,but as a placeholder I put together LXMinis so I am officialy active now. I am doing a preamp for my F5/ACA combo to use it with the Minis (currently running on Emotiva); even though now that I have looked at Siegfried's XO I realize that for my hearing capabilities I could have probably gotten rid of his one EQ boost for which he pushed the Full range channel down 10dB and then I could have used my A amps even without the preamp.
,but as a placeholder I put together LXMinis so I am officialy active now. I am doing a preamp for my F5/ACA combo to use it with the Minis (currently running on Emotiva); even though now that I have looked at Siegfried's XO I realize that for my hearing capabilities I could have probably gotten rid of his one EQ boost for which he pushed the Full range channel down 10dB and then I could have used my A amps even without the preamp.
For 35 years I never used a crossover.
I tend to use full range drivers.
Or a full range driver with a sub in parallel with no crossover.
I only recently came across a crossover when I built a twin tweeter box.
It came with a 2u2 to go in series with the tweeters.
I cant say it has caused any problems.
I tend to use full range drivers.
Or a full range driver with a sub in parallel with no crossover.
I only recently came across a crossover when I built a twin tweeter box.
It came with a 2u2 to go in series with the tweeters.
I cant say it has caused any problems.
I think active and dsp certainly has it's use, but i'm wondering why everybody like those crappy cheap chipamps and da convertors. Doing DSP right needs also a lot of money, and is often more expensive on low power models than passive. Good DA does not cost a fortune, but still a bit, and the standard dcx or minidsp is not good for me. Idem with amps. Those cheap tpa and other chipamps many of you use sound like ****. That won't do it for me. Idem with those cheap dsp's. Minidsp HD is ok (altough the ad/da is still the weak point), and real good dsp's cost a quiet a lot so passive is cheaper and for me at least, easier to implement right.
I would only use active for higher power systems or studio monitors (where a flat response is mandatory). For small power hifi, passive is often way better and a more or less flat response is enough, and you only have to have one amp so you can afford better amps.
But as always, each to his own. As long as you are happy, your system is good...
I would only use active for higher power systems or studio monitors (where a flat response is mandatory). For small power hifi, passive is often way better and a more or less flat response is enough, and you only have to have one amp so you can afford better amps.
But as always, each to his own. As long as you are happy, your system is good...
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.