Well, you can make anything 'flat' with a MiniDSP, provided it is not fundamentally limited in some physical way. For example, there is not enough EQ available to get 20kHz out of a 15" bass driver. Even if you could, you wouldn't want to because factors other than frequency response quickly become overwhelming. The directivity, distortion and decay characteristics render a 15" bass driver useless at 20kHz no matter how much EQ you apply.
My previous exhortation to 'Flat EQ' a modest system was really serious. You can't make a silk purse from a sows ear but a bit of massaging will get you quite close in some areas, but no closer at all in others.
What I mean by that is that 'Flat EQ' will cure many of the problems modest systems might suffer from IN FREQUENCY RESPONSE TERMS.
And that can sound massively better than the same system without 'Flat EQ'. But if the system has a dodgy phase response from unmatched crossover slopes, or ghastly stored energy problems from a ringing cone, or harmonic distortion issues from a poorly designed magnetic circuit, then no amount of EQ can help. These other factors then become the limiting factor of system performance.
So, I don't propose a MiniDSP or other EQ to make a lousy system sound good.
But if you have a system with OK, or good, or excellent performance in most areas other than frequency response, then EQ can help.
Many wide range drivers have a non-flat frequency response and, in my experience, often sound much better when 'flattened out'. Some will cry "Don't use EQ! It Is Evil! Use baffle step correction and a notch filter instead". Well, that's just EQ in another guise. I firmly believe that anything that can be done passively with Rs and C's and Ls (such as the abovementioned BSC) is done better at line level, particularly if you want to avoid burning precious watts in warm resistors.
So, if you plan to actively EQ your system, then try and design your crossover to meet all audio quality requirements EXCEPT frequency response. If you end up with a bloody great bump somewhere, EQ it out and listen. You will soon hear (or even better, measure) if there is something amiss and you can fiddle your crossover to suit.
Whether or not first order slopes are enough depends on a swag of things that are not fundamentally altered by 'Flat EQ'. So, as usual, look at the response and impedance of your drivers for at least 2 octaves either side of your chosen crossover frequency. A passive crossover at very low frequencies is a dangerous thing, as the resonance of both drivers is usually well within their passband, and will make a complete mess of shallow slopes.
EQ is not a substitute for sound loudspeaker design practice. But if you include it IN sound loudspeaker design practice then you have many more (and more efficient) options.
Sorry for the incoherent ramblings but I feel that this is a subject not often brought to light, but worthy of consideration.
Good Luck!
>>> Would it be sufficient to cross everything first order to keep the crossover as cheap as possible, and then use minidsp to make it flat?
That's what i do. Simple cap on the tweeter and then use the sub amps variable gain and frequency response.
I like what blak does too but don't have the gear.
That's what i do. Simple cap on the tweeter and then use the sub amps variable gain and frequency response.
I like what blak does too but don't have the gear.
as Blakkshepeaudio says
EQ is not a substitute for sound loudspeaker design practice. But if you include it IN sound loudspeaker design practice then you have many more (and more efficient) options.
I could not agree more. I am hoping I can get the system "close" using mechanical methods, and then fine tune electrically.
EQ is not a substitute for sound loudspeaker design practice. But if you include it IN sound loudspeaker design practice then you have many more (and more efficient) options.
I could not agree more. I am hoping I can get the system "close" using mechanical methods, and then fine tune electrically.
Okay, so after digesting all that, here's what I've been thinking about. Tell me if this would work: from what I understand the beta 12lta should roll off on OB at around 90hz (Fs/Qts, or 45/~.5 = 90hz), so put it on an OB, even a narrow one, and then use minidsp to adjust for baffle step as well as peaking. I lose 6db+ in output but I have 50 watts so a 90db speaker is okay. Now I have a ~90db driver that goes from 90hz to 8khz, right? Then if I passively cross a tweeter to the 12LTA and use one channel for them, I can use the other channel of minidsp to cross in a woofer, and if necessary I can make these slopes steep. Or I could use a plate amp for the bass drivers as Zilla suggests
Good idea? Bad idea? Does it just suck on my part to be willing to rob the 12LTA of 6db sensitivity?
(I just don't see the difference, apart from needing more power, between this and using a really big baffle. If you can afford baffle step loss then you can make the baffle as narrow as you want, can't you?)
Good idea? Bad idea? Does it just suck on my part to be willing to rob the 12LTA of 6db sensitivity?
(I just don't see the difference, apart from needing more power, between this and using a really big baffle. If you can afford baffle step loss then you can make the baffle as narrow as you want, can't you?)Correct me if I am wrong, but the info on the mini dsp website says that it operates in the digital domain. It takes an analog signal, converts it to digital at 24bit/48kHz, adds correction to the signal, then converts it back to analog (24bit/48kHz), all for $125. Having gone through several upgrades to my digital front end, I am confident that the resulting sound quality is 'entry-level'. I am sure that it improves the sound of many systems. Used for EQ, it will remedy many frequency response linearity issues. However, I would not trade detail, transparency, dynamics and harmonic richness (available from a high quality digital source, as well as from vinyl) for response linearity... unless the latter were really bad. Speakers are typically prone to the greatest response deviations of any of our sound system components. I try to choose drivers with response anomalies that I can live with or easily correct for at line level or through simple physical modifications. IME, too much 'corrective action' does indeed flatten the sound. Still, these sound too fun to not play with. I see one in my computer room systems future...
Regards,
Bob
Regards,
Bob
Okay, so after digesting all that, here's what I've been thinking about. Tell me if this would work: from what I understand the beta 12lta should roll off on OB at around 90hz (Fs/Qts, or 45/~.5 = 90hz), so put it on an OB, even a narrow one, and then use minidsp to adjust for baffle step as well as peaking. I lose 6db+ in output but I have 50 watts so a 90db speaker is okay. Now I have a ~90db driver that goes from 90hz to 8khz, right? Then if I passively cross a tweeter to the 12LTA and use one channel for them, I can use the other channel of minidsp to cross in a woofer, and if necessary I can make these slopes steep. Or I could use a plate amp for the bass drivers as Zilla suggests
Good idea? Bad idea? Does it just suck on my part to be willing to rob the 12LTA of 6db sensitivity?
Your plan makes sense. A first order tweeter crossover at HF is a time honoured approach. As the MiniDSP HF channel will be driving your LTA and your tweeter, you can EQ them both together and take care of any sensitivity mismatch as well as the rising response of the LTA, within reason of course.
As to the merits of EQ vs bigger baffle, you need input from those more experienced with open baffles than I.
However, don't get confused with gain, level and sensitivity. The speakers sensitivity, and hence its need for power to achieve a certain SPL, does not change when actively EQed. The change is made in the amount of power delivered to the speaker at different frequencies, by altering the level going into the amp. At line level you also have variable gain available which allows you to set an overall level after the EQ is applied.
Correct me if I am wrong, but the info on the mini dsp website says that it operates in the digital domain. It takes an analog signal, converts it to digital at 24bit/48kHz, adds correction to the signal, then converts it back to analog (24bit/48kHz), all for $125. Having gone through several upgrades to my digital front end, I am confident that the resulting sound quality is 'entry-level'. I am sure that it improves the sound of many systems. Used for EQ, it will remedy many frequency response linearity issues. However, I would not trade detail, transparency, dynamics and harmonic richness (available from a high quality digital source, as well as from vinyl) for response linearity... unless the latter were really bad. Speakers are typically prone to the greatest response deviations of any of our sound system components. I try to choose drivers with response anomalies that I can live with or easily correct for at line level or through simple physical modifications. IME, too much 'corrective action' does indeed flatten the sound. Still, these sound too fun to not play with. I see one in my computer room systems future...
Regards,
Bob
Yes, they are an A/D-process-D/A device. And, yes, they are a lot of fun to play with.
SQ wise, it depends on the overall resolution of the system, and the areas of performance that are critical to the individual listener. I believe that frequency response flatness is just one important characteristic of a system, and that FR correction should only be used if more is gained than lost. I have designed speakers that lost their 'magic' when over EQed, and it is easy to imagine a very high-res system revealing digital artifacts that mangle the sound worse than 2dB of treble lift in a driver.
I propose the MiniDSP more as a tool to aid in developing systems.
I have one system that was transformed utterly by EQ (from a 3 to a 6)
and another that went from sounding slightly shouty to sounding slightly dull (from a 5 to another, different 5).
It seems to me a great way to get the benefits of wide range drivers while addressing their limitations. I can't wait for someone to try 'Flat EQi'ng out the famous Lowther shout. Does the system sound better or worse? What is it about the Lowther that people like, and why do some hate it? Strip away the anomolous FR and find out, I say.
Can't wait to learn about your results. Right now I'm dreaming of 12LTA paired with this tweeter and some appropriate low bass solution...
Hey Zilla! Mine are getting a rejig as well.
'Acceptance & Support' is being withdrawn by my better half so a rehash is in order. All the same drivers in a "smaller cabinet please"
I think I'm gonna start a new thread. Where though? 'multiway' or full range section?
This thread is turning into a really useful mine R&D with the 12LTA - something I wish I had when I started out with this driver.
Good luck boys n girls!!
'Acceptance & Support' is being withdrawn by my better half so a rehash is in order. All the same drivers in a "smaller cabinet please"
I think I'm gonna start a new thread. Where though? 'multiway' or full range section?
This thread is turning into a really useful mine R&D with the 12LTA - something I wish I had when I started out with this driver.
Good luck boys n girls!!
Another question: If you don't intend to use the 12LTA for deep bass, wouldn't it be possible to put it into a much smaller sealed cabinet, like about 40L? WinISD predicts a Qts bump but it might be useful in serving as BSC. Bass would only extend to 70 or 80Hz in this configuration, but low enough that a single driver could be used for omnidirectional sub-bass. (Guys, I had a dream about this last night. I'm truly getting obsessed.)
JRKO, I would keep any more threads about the 12LTA in this forum because I doubt the multi-way people would be all that interested. Even with helper drivers above and below, these projects are definitely in the spirit of full range!
JRKO, I would keep any more threads about the 12LTA in this forum because I doubt the multi-way people would be all that interested. Even with helper drivers above and below, these projects are definitely in the spirit of full range!
Yes, that's what the website says (24bit/48kHz). It looks like you get a lot of functionality and performance for the money, just not enough for some applications. What would be nice is the same functionality, but with no conversions. Digital in from the source (CD player, music server, etc.), response correction, digital out to your high quality DAC. Just dreaming...
Bob
Zilla I'm interested to see your OB setup. I have to say that after playing for a few days mine started to sound a lot better. I cut out the dust caps and put a little foam behind the wizzer . This with a baffle step filter made the speakers sound good. Today I put together a quick and dirty pair of boxes 6 cf. sealed. Well, as sealed as they can be with the holes left by the dust cap removal. I put one driver facing forward and one rearward. No baffle step filter and descent bass. This is getting good enough for me to think they could stay. Let us know what you find....Evan
Better yet is the combination minidsp and minidigi.
miniDSP/DIGI Combo | miniDSP
At $160 it is in the same boat as the minidsp. It can be a great value in the right system, and may perform well above its price level, but it's still probably not something one would likely install in front of their Ayre, Benchmark, DAC Magic, Audio Note DAC or (insert your favorite dac). Sometimes the best value really can be fairly expensive.
Bob
Better yet is the combination minidsp and minidigi.
miniDSP/DIGI Combo | miniDSP
At $160 it is in the same boat as the minidsp. It can be a great value in the right system, and may perform well above its price level, but it's still probably not something one would likely install in front of their Ayre, Benchmark, DAC Magic, Audio Note DAC or (insert your favorite dac). Sometimes the best value really can be fairly expensive.
Bob
Absolutely. Like everything it is horses for courses.
Everything in a system throws in some dirt and it is easy to see the MiniDSP out of its league in a clean high end system. It is certainly not universally applicable.
I might get this discussion into a thread on its own.
A separate thread is a good idea. I looked around to see where one might put such a thread and found several related threads on the 'Digital Line Level' forum here (on diyAudio). Looks like a real gold mine! If you do start one, let us know.
Regards,
Bob
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.