poptart said:
Your discussion is getting more reasonable, but still off the mark. Your line; "but my point wasn't about reproduction, it's the music" is your best point and I would simply reply that all I am "talking about is reproduction, not the music". I think that you were the one who got this confused not I.
This issue is at the core of every single argument about listening etc. And I never said that "musicians are poor judges of sound" only that they tend to not hear imperfections as well as they could and one can certainly not ASSUME that they are the final judge. If you want to talk about "proof" where is yours?
How else do you explain a concert pianist who could not hear the clipping in his own records? He surely heard it after I pointed it out to him! But he had just glossed over this aspect listening to the performance, which being his own he should have been intimately familiar with.
If you would stop taking things out of context and ty to understand the point that I am trying to make you might come to appreciate how much more difficult the job of speaker design and evaluation is. Its far more than just making a speaker "sound good" - to whom? With what source? If the sound reaching the "producers ears" is not the golden reference, then what is? Thats their job! Sometimes this is the musician and sometimes they know better.
Your point about monitors, is dead on. But then this means that the monitors need to be accurate and nuetral too, which is, by ANY producers or recording engineers standard, the ideal. The target is not a fixed one - continuous improvement - but the final goal is - accuracy. Anything else is "false".
Idealy there would be no need for mixing because that could mix up some spacial ques that should be present at the location of the recording. Of course not all speaker can reproduce these spacial ques, and rely on the playback room to create the effect instead.poptart said:
This makes a massive assumption that (a) the piece can be recorded all in real time via a 2 mic arrangement and (b) that 'spatial cues' are all that relevant in the enjoyment of the piece.soongsc said:
Whilst either may be true some of the time, they are far from universal.
This is one reason why I try to get my hands on live unamplified recording of performances that I have listened to live.Brett said:
Can you be more specific about it? I see no more design tradeoffs than problem with matching directivity of the front lobe and leaving back lobe to corrupt power response. Could little bit attenuation of the mid driver output in crossover region improve power response and leaving direct sound balance nearly untouched?
MethMan said:
Matching the polar response to the waveguides polar response requires a new waveguide; the enclosure is not acceptable for Pro applications, which also have a real problem with the backwave; there is simply no effective way to do the LF portion - below 200 Hz - with a dipole due to its low efficiency - thus you have a problem with a sharp change in the power response going from the dipole to the monopole. The edge diffraction issue concerns me.
Bottom line is that these problems ensue without a clear-cut advantage.
Josh, Earl, AJ, Klaus, thanks very much for your posts about the OS wave guide equation, the spreadsheet, and the indication of how to get it into a CAD program. Very helpful! 😎
cool
cool
Dr. Geddes here is a plot of the DDS Eng-1:
http://htguide.com/forum/showpost.php4?p=392802&postcount=13
Would your 10" WG have directivity control any lower than this? Or are they pretty comparable? How about the 12"?
http://htguide.com/forum/showpost.php4?p=392802&postcount=13
Would your 10" WG have directivity control any lower than this? Or are they pretty comparable? How about the 12"?
gedlee said:
I agree about PA use but in a home environment 80 cycles at high sensitivity is easy, and I find the benefits are worth the size- ie it is defiantly possible to build a dipole with sensitivity of your Summa if not higher. It just won't go as low without eq but 80 cycles is a breeze
augerpro said:
This would be control done to about 3 kHz in my opinion. Below that it begins to widen, becoming pretty wide at 2 kHz. and below.
The bottom line is the bigger the mouth the lower the control.
I wouldn't call those curves "impressive", about what I'd expect.
To compare they would have to be the same driver measured using the same system. I don't usually trust other peoples measurements.
Agreed. I'd be surprised if you could get a dipole of the same size within 6 dB of the sensitivity at 80 Hz. I have spent a lot of time looking at (and trying) the various options in a package of that size (~18" wide baffle) and the best I can get is about 100-101 dB (@1m w/ 2.83 V from 8 ohm nominal). That is with just the right driver in a good sized ported box tuned high (2.5 cubes tuned to 75 Hz) to provide just the right balance between the baffle step rolloff and the enclosure's peak around tuning and it's still -3 dB at 80 Hz (so I can cascade it with a 4th order BW to get a net 8th order LR). No dipole setup comes close in terms of senstivity. If you add more drivers or make the baffle huge to get the sensitivity you (1) make the speaker much larger in size and (2) limit the bandwidth over which the directivity is effectively controlled. If you're trying to run the driver all the way up to 1k and have decent polar behavior, then there's definately no other option that gets you there.
gedlee said:
I'll be using the DE250, with most likely a 10" from DDS or yourself, so I was just curious if I could expect yours to hold control as well or better than the DDS. I wanted to cross an B&C 8NDL51 to these at around 1.5k Hz LR4 but now that I see the plots of the DDS I'm thinking that may be too low.
gedlee said:
Well, depends how the footprint is measured. I use 12 sixteen ohm tens (91 db / watt /m each) wired in parallel on a board (do the math - it's very sensitive) that is ony 26 inches wide and 32 tall for bass. They are nearly flat without eq because of the .8 QTS. It's a three way. Probably 6 to 9 db more sensitive then the Summa at 80 cycles. I base this on comparing to a JBL 2226H in a reflex with baffle step correction of 3 db. I use them 60 to 500 cycles. The drivers were 18 dollars a piece and the board and wire didn't cost to much.
😉
Magnetar said:
What did I miss? I don't remember seeing any picture about this...
Oh, sorry for the OT.
CLS said:
What did I miss? I don't remember seeing any picture about this...
Oh, sorry for the OT.
They are in the raw, I have some new measuring gear here and I'm working on the midrange still. I'm starting to get the hang of using the software. Microphone placement is ultra critical.. I'm learning. Right now it's a large format phenolic compression driver with a small format aluminum for 3K up. I'll post something in the OB thread I started when it's done. I only added one woofer compare to the old OB system - the bass is where I want it.
Magnetar,
Can you give us the T/S parameters on those 10s? When I model that size baffle with 12 10s in The Edge it shows me -6 dB at 80 Hz relative to half-space. Supposing your 10s are ~91 dB 1w (half-space) and the 12 add perfectly, you're looking at a 10.8 dB efficiency gain, putting you at a max of about 102 dB w/ 1w to the set. When you throw in the 6 dB of loss (relative to half space) at 80 Hz from going to dipole, you're back down to 96 dB there with 1w on the set.
Furthermore, assuming these are in a 3x4 grid, in the 3 dimension (w/ 11" center-center spacing), not including dipole effects (just from cones) you hit a 90 degree beamwidth at about 350 Hz. In the 4 dimension under the same assumptions you hit a 90 degree beamwidth at 260 Hz. Lower these numbers some to account for the dipole directivity and examine the center to center spacing required from the midbass to midrange and IMO you're looking at a setup not really usable above 200-250 Hz.
I don't consider this to be comparable performance to 1 $100-150 15 in a smaller footprint with the same or slightly higher sensitivity at it's lowest point, twice the usable (directivity wise) bandwidth, and the ability to be placed fairly close to a corner. As a bonus, ported and tuned high the 15 probably has lower cone excursion over all the 12 10s dipole, but I'd imagine at normal levels it's negligible in both systems anyways.
Can you give us the T/S parameters on those 10s? When I model that size baffle with 12 10s in The Edge it shows me -6 dB at 80 Hz relative to half-space. Supposing your 10s are ~91 dB 1w (half-space) and the 12 add perfectly, you're looking at a 10.8 dB efficiency gain, putting you at a max of about 102 dB w/ 1w to the set. When you throw in the 6 dB of loss (relative to half space) at 80 Hz from going to dipole, you're back down to 96 dB there with 1w on the set.
Furthermore, assuming these are in a 3x4 grid, in the 3 dimension (w/ 11" center-center spacing), not including dipole effects (just from cones) you hit a 90 degree beamwidth at about 350 Hz. In the 4 dimension under the same assumptions you hit a 90 degree beamwidth at 260 Hz. Lower these numbers some to account for the dipole directivity and examine the center to center spacing required from the midbass to midrange and IMO you're looking at a setup not really usable above 200-250 Hz.
I don't consider this to be comparable performance to 1 $100-150 15 in a smaller footprint with the same or slightly higher sensitivity at it's lowest point, twice the usable (directivity wise) bandwidth, and the ability to be placed fairly close to a corner. As a bonus, ported and tuned high the 15 probably has lower cone excursion over all the 12 10s dipole, but I'd imagine at normal levels it's negligible in both systems anyways.
Magnetar said:
You use so many drivers and the combo is very sensitive, BTW what amp do you use to drive them?
ttan98 said:
Normally an old beefed up Crown PSA2 - around 700 WPC into 2 ohm. Lots of current and control