The receiver hasnt any problem with 4ohm speakers, the Epos are, and my earlier Dali was 4ohm. Yeah I know 4ohm and 4ohm are not the same always.
I thought 1,5-1,6kHz was to low for the 27TDFC, but I havent tried so you know better! But is that with higher xo slopes or?
I'm testing now with 2.order LR@2400Hz, in a 11L closed box. And the CA26 in a 50L vented box. It doesnt sound bad at all but it could be better.........
This is what I have purcased, and can play with:
2x 27TFFC
5x 27TDFC
6x CA18RNX
2x CA26RFX
2x L26RFX/P
Maybe I should try a WMTMW ?
Well I know what my wife would say!
I thought 1,5-1,6kHz was to low for the 27TDFC, but I havent tried so you know better! But is that with higher xo slopes or?
I'm testing now with 2.order LR@2400Hz, in a 11L closed box. And the CA26 in a 50L vented box. It doesnt sound bad at all but it could be better.........
This is what I have purcased, and can play with:
2x 27TFFC
5x 27TDFC
6x CA18RNX
2x CA26RFX
2x L26RFX/P
Maybe I should try a WMTMW ?
Well I know what my wife would say!
What you mean by 2nd order LR is acoustic LR2, isn't it? Acoustic LR2 is not easy to design. You'll need a stepped or slanted baffle, or a waveguided tweeter (e.g., John Zaph Krutke's Waveguide TMM). I used the TDFC crossed at 1.55 kHz with LR4 slopes with no distortion or compression issue. Zaph also used TBFC/G crossed at 1.45 kHz for his BAMTM design.
I wouldnt do 24db filter and I would be very cautious with very low xo points
Here are some inspirations with MTM waveguide supported tweeter ... seems very viable with low order slopes and higher xo point
Even more skilled people seem to get better sound placing xo point as high as possible rather than as low as possible ... making low order slopes viable
I dont know the seas drivers, but I cant imagine them to be any worse than other drivers in this respect
http://www.troelsgravesen.dk/C20WH.htm
also see other Vifa C17 projects
http://www.troelsgravesen.dk/Diy_Loudspeaker_Projects.htm
Here are some inspirations with MTM waveguide supported tweeter ... seems very viable with low order slopes and higher xo point
Even more skilled people seem to get better sound placing xo point as high as possible rather than as low as possible ... making low order slopes viable
I dont know the seas drivers, but I cant imagine them to be any worse than other drivers in this respect
http://www.troelsgravesen.dk/C20WH.htm
also see other Vifa C17 projects
http://www.troelsgravesen.dk/Diy_Loudspeaker_Projects.htm
tinitus said:
An appropriate xo point depends on multiple factors. There's no absolute guideline about this. The CA18 will benefit from a low XO freq because of its nonlinear distortion characteristic: http://www.zaphaudio.com/6.5test/compare.html. Crossing low with a 6.5-7" midwoofer has another advantage: better off-axis performance, because midwoofer's beaming starts low. And the TDFC's low-end performance is exceptional. Very low distortions. With a waveguide, the TDFC will have no problem being crossed at 1.8 to 2 kHz in an LR2 design.
-Jay
Nope, though the CA18RNX's nonlinear distortions are not that bad in the group, it's certainly not better than high performers. For example, compare it to the Ushers and Peerless HDS. Significantly worse.
And remember that a midwoofer's rolloff in a crossover starts much lower than a nominal XO freq. The CA18's distortions are high in the 800 Hz to 1.5 kHz range. If you use a low xo point, these distortions will be better reduced. Coupled with the TDFC's low low-end distortions, the system will have very low distortions in the mids and upper midrange. This is basically what I did in my Dayton RS180/Seas 27TDFC 2-Way design.
And remember that a midwoofer's rolloff in a crossover starts much lower than a nominal XO freq. The CA18's distortions are high in the 800 Hz to 1.5 kHz range. If you use a low xo point, these distortions will be better reduced. Coupled with the TDFC's low low-end distortions, the system will have very low distortions in the mids and upper midrange. This is basically what I did in my Dayton RS180/Seas 27TDFC 2-Way design.
I do focus on 3.order and I still dont agree as I believe in "smooth" linear and frequency independant distortion and I really dont think that low distortion is to be the end of all ... to me its more important how the complete speaker reproduce the distortion on the recordings
But it seems rather pointless to go that far if we havent worked practically with this specific driver
But it seems rather pointless to go that far if we havent worked practically with this specific driver
Here's CA18RNX's harmonic distortion sweep:
Don't you see high 3rd and 5th order harmonics in the 700 Hz to 1.5 kHz range? Note that higher order harmonics are more audible than lower order ones. For example, a 5th order harmonic that is 10 dB lower than a 3rd order harmonic isn't necessarily less audible. And a third order harmonic at the same level as the second order one should be very noticeable.
Compare the CA18's performance to the Peerless HDS 830883's:
And to even better Usher 8945P's:
Significantly different.
And nonlinear distortions of a driver, provided that it has good linear distortions, tell much about its performance. A system's distortion characteristics are directly related to individual drivers' distortions unless their high-distortion regions are attenuated by the crossover. This is why I suggested using a low xo freq for the CA18, to improve the system distortions.
-Jay
Don't you see high 3rd and 5th order harmonics in the 700 Hz to 1.5 kHz range? Note that higher order harmonics are more audible than lower order ones. For example, a 5th order harmonic that is 10 dB lower than a 3rd order harmonic isn't necessarily less audible. And a third order harmonic at the same level as the second order one should be very noticeable.
Compare the CA18's performance to the Peerless HDS 830883's:
And to even better Usher 8945P's:
Significantly different.
And nonlinear distortions of a driver, provided that it has good linear distortions, tell much about its performance. A system's distortion characteristics are directly related to individual drivers' distortions unless their high-distortion regions are attenuated by the crossover. This is why I suggested using a low xo freq for the CA18, to improve the system distortions.
-Jay
Very interesting.
I thought it would be worse with a low xo for the tweeter rather than "to high" for the mid, related to distortion. But there something new to learn every day
I have two questions; is this also true with high SPL, or will the 27TDFC getting worse when its "pushed to the limit"?
And what about the xo, a fourth order isnt that a bit hard to get right for a newbie like me?
I thought it would be worse with a low xo for the tweeter rather than "to high" for the mid, related to distortion. But there something new to learn every day
I have two questions; is this also true with high SPL, or will the 27TDFC getting worse when its "pushed to the limit"?
And what about the xo, a fourth order isnt that a bit hard to get right for a newbie like me?
Norcad said:
According to my experience, the 27TDFC is a very solid performing driver. With my 1.55 kHz XO in a very narrow baffle, I have no compression or distortion issue when I crank the volume up. Notice that Zaph used the 27TBFC/G (essentially the same tweeter except dome material) with 1.45 kHz LR4 XO for his BAMTM design. He's a guy who measures everything he can think of.
As for your second question, no, it's the other way around. An acoustic LR 2nd order design is more difficult to implement than LR4. This is why the LR4 design is so popular. But your driver choice also makes LR2 feasible as Zaph did for his Waveguide TMM. And you probably know the difference between electrical 4th order and acoustical 4th order? Usually LR4 can be achieved by an electrical 2nd order network. And to achieve good phase tracking, people don't use symmetric LR4 slopes, but use asymmetric ones with either shallower LP or HP slope.
If you're interested, here's a method I use to design a crossover: http://www.geocities.com/woove99/Spkrbldg/DesigningXO.htm
I don't use a measurement system, but know how to achieve what I want, using the method I described in my page.
-Jay
Look at Mark K's distortion test of the TDFC: http://www.markk.claub.net/Testing/Tweeter3/seas_comparison.htm
Notice that he drove the tweeter to 106 dB at 0.25m with 1 kHz test tone to observe very good harmonic distortion numbers. This means that if you use the TDFC crossed at 1.5 kHz with LR4 slope and push it to 106 dB at 1 kHz at 0.25 meter, then the system's overall SPL at 0.25m will be about 120 dB. It equals to 102 dB at 2 meter away. This is very loud. And the 106 dB at 0.25 m is apparently not the TDFC's limit considering observed low distortions.
Notice that he drove the tweeter to 106 dB at 0.25m with 1 kHz test tone to observe very good harmonic distortion numbers. This means that if you use the TDFC crossed at 1.5 kHz with LR4 slope and push it to 106 dB at 1 kHz at 0.25 meter, then the system's overall SPL at 0.25m will be about 120 dB. It equals to 102 dB at 2 meter away. This is very loud. And the 106 dB at 0.25 m is apparently not the TDFC's limit considering observed low distortions.
They really dont look better to me, but drivers like 18W(SS) and XG18(Vifa) does look better, at least to me
But its worth to mention that Krutke still hears some roughness in the 4khz region on the XG18
I think you interpret the measurements wrongly
Well, enough of that ... we are simply not aiming at the same goal I guess that yours are low distortion and mine are ultimate sonics
Good luck to you!
But its worth to mention that Krutke still hears some roughness in the 4khz region on the XG18
I think you interpret the measurements wrongly
Well, enough of that ... we are simply not aiming at the same goal
I guess that yours are low distortion and mine are ultimate sonics Good luck to you!
Perhaps, you don't know how to read Zaph's harmonic distortion sweeps? The Usher 8945P is one of the best 7" in the DIY market---with respect to overall nonlinear distortions it is even a bit better than the ScanSpeak Revelator. The Peerless HDS Exclusive is also among the best---but not better than the Usher or the ScanSpeak. These points are what both Zaph (read his description about the Usher and the Peerless in the 6.5" test page) and Mark K agree with.
Also look at Mark K's (more comprehensive) distortion test: http://www.markk.claub.net/Testing/woofer8/Usher_new_7_tests.htm
As for the XG18, note that Zaph and Mark's tests don't agree: http://www.markk.claub.net/Testing/XT18/Vifa_XG18.htm
Ultimate sonics? I honestly don't know what you mean by it
Also look at Mark K's (more comprehensive) distortion test: http://www.markk.claub.net/Testing/woofer8/Usher_new_7_tests.htm
As for the XG18, note that Zaph and Mark's tests don't agree: http://www.markk.claub.net/Testing/XT18/Vifa_XG18.htm
Ultimate sonics? I honestly don't know what you mean by it
I think its absolute pointless to choose the best drivers and putting them together expecting them to work perfectly
In my world its first and mostly a matter of determining where and how the drivers work together, and thats not just a matter of using them where they are at their best ... many have failed and wasted money thinking like that
It surprices me that I dont seem to find distorsion measurements on the finished speakers .... I would think that to be more relevant
In my world its first and mostly a matter of determining where and how the drivers work together, and thats not just a matter of using them where they are at their best ... many have failed and wasted money thinking like that
It surprices me that I dont seem to find distorsion measurements on the finished speakers .... I would think that to be more relevant
tinitus said:
I really don't know what you mean here. I didn't simply say that a certain driver is the best for any purpose. That's why I wanted to show you data and let you judge. Of course, we have to consider linear, nonlinear distortions in the range for the dirvers to be used, their power handling, box modeling, baffle modeling, and so on. Provided that other factors look good, nonlinear distortions (of course, in the range we want to use the drivers for) are very important and often overlooked factors in choosing drivers, crossover points, and slopes. I don't know what I did wrong in using the data to show my points.
At this point I really want to ask you what data you use to choose drivers in order to achieve your "ultimate sonics?" Perhaps your intuition? Or simply believe in others' listening impressions in their particular applications?
Look at Zaph's ZD5 and Bargain Mini pages. You will find the system harmonic distortion measurements there. Also take a look at Mark K's RS225/RS28A design page. You'll find another there. The reason why these people do the distortion test of so many individual drivers is that they want to use the data to choose appropriate drivers for their speaker designs.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.