Looks 1st order needs very wideband devices to work right so guess it best to let it be and live with drivers inherent sound : )
Below is a 40Hz-17kHz audio-band summed of two 1st order band-passes and while they sum perfect on axis (axis subject below) yellow show something ala if RS225 had break up bloom removed and sum get very bad.
Another point to be aware of for 1st order is the vertical tilted lobe inside wide XO region that only gets perfect using zero ctc distance coaxial driver. Below show tilt for 50mm steps of spacing with 400Hz XO point and the rightmost one is probably close to performance for build in this thread, the other one shown reflect if cabinet if flipped 180º around.
Attachments
Member
Joined 2009
Paid Member
It's non-ideal perhaps, but I think there are plenty of other factors that may be more important - I'm thinking room boundaries here - the floor, ceiling and walls. At 400Hz/2.8m you may well have a front-wall interaction and floor interaction to worry about ??
the key to making a 1st order XO work is to use DSP to put the drivers exactly on the correct slope. As long as you can EQ the drivers to the 1st order slops out to about an octave or two beyond the XO port, the drivers will sum correctly.
I did this with the RS225HF-8/10F-8424 at 500Hz and it worked OK -- the impulse response was OK, but it sounded hollow and muddy. A Harsch XO sounded much cleaner, even though the impulse response was no better than the BW1.
Bob
I did this with the RS225HF-8/10F-8424 at 500Hz and it worked OK -- the impulse response was OK, but it sounded hollow and muddy. A Harsch XO sounded much cleaner, even though the impulse response was no better than the BW1.
Bob
Member
Joined 2009
Paid Member
The Harsch xo probably sounded better because the steep 4th order low pass on the woofer stops any cone breakup from coming through - hence less muddy. My RS225-8 with first order passive sums perfectly with first order passive high pass on the 10F. I don't have a problem with muddy sound.
Xo design (10F/8424 used where it says FF85WK):
Here is my 1st order xo measurement:
Here is impulse and step:
Xo design (10F/8424 used where it says FF85WK):
Here is my 1st order xo measurement:
Here is impulse and step:
Last edited:
I suspect it is because the two different XO's behave very differently off-axis. We rarely see polar plots, and since the XO's were designed at 1/2m, one expects different results at 2.5m in-room.
Bob
No, I don't thinks so. I know what woofer break-up sounds like, and this isn't it. I was able to put the RS225 dead on a BW1 curve for 40dB's with just three PEQ's.
Muddy probably isn't the work most would choose. Its that undefined and ringy bass that I associate most often with hi-Z SE amps. I also did a straight LR4 XO with no PEQ's at 500Hz, and it also sounded cleaner than the BW1 even though the impulse was a mess.
Bob
Don't know if below is observed actual situation but for verticals 1st order beam energy sum whether drivers are flipped around or not downwards or upwards for very wide XO region, Harsh also tilt little bit in XO region but as with forward beaming LR below XO region is much more narrow. First room reflections will probably be very different for each scenario, if it can be coupled to ringy bass don't know try flip speaker 180º around : )
Attachments
Started a new thread for my build:
http://www.diyaudio.com/forums/multi-way/302788-2-3-way-reference-build.html
- Status
- Not open for further replies.