zog Posted March 23, 2014 Author Share Posted March 23, 2014 zog, the last measurements with all settings off except the xover and freq shaping for high freq? Yes, while I'm in a screenshot mood I'll paste the "active" screens, all the rest per channel dead flat and zerod out. Would be nice if miniDSP had a way to print all this stuff in an at a glance report, the XML file is a tad unreadable. I'm using the 2.1 advanced plugin, but I use it in stereo LF mode.. Link to comment Share on other sites More sharing options...
henry218 Posted March 23, 2014 Share Posted March 23, 2014 cool. so whats your subjective opinion on the sound? Link to comment Share on other sites More sharing options...
davewantsmoore Posted March 23, 2014 Share Posted March 23, 2014 (edited) Smoothing the hell out of everything seems to be guilding the lily sometimes No not suggesting smoothing.... Just reducing reflections and gating correctly. Then, not taking any more measurements, unless it is for checks and balances. Like I said, that perhaps all a bit down the track. Edited March 23, 2014 by davewantsmoore 1 Link to comment Share on other sites More sharing options...
zog Posted March 23, 2014 Author Share Posted March 23, 2014 cool. so whats your subjective opinion on the sound? My older config and yours are quite close in sound - both sound good. I'd say yours a tad lusher/thicker in sound and my older "evolved with a succession of tweaks" config a bit more cleaner and analytical. Just listening now with a few dB off the tweeter to counter the aforementioned harshness on electric guitar. Dialogue on Thai Variety shows on youtube also rendered very clearly :-) One possible factor here (and I know there are others!) could be (IMHO) the compression driver sounds cleaner than the Lorantz woofer and the older config crosses over into it lower and steeper... Link to comment Share on other sites More sharing options...
henry218 Posted March 23, 2014 Share Posted March 23, 2014 ok good to hear, although im expecting an increase of coherency. how do you listen to them? i mean do you toed them in? Link to comment Share on other sites More sharing options...
zog Posted March 23, 2014 Author Share Posted March 23, 2014 (edited) deleted as I should specify consistently with my earlier post. Edited March 23, 2014 by zog Link to comment Share on other sites More sharing options...
henry218 Posted March 24, 2014 Share Posted March 24, 2014 My older config and yours are quite close in sound - both sound good. I'd say yours a tad lusher/thicker in sound and my older "evolved with a succession of tweaks" config a bit more cleaner and analytical. Just listening now with a few dB off the tweeter to counter the aforementioned harshness on electric guitar. Dialogue on Thai Variety shows on youtube also rendered very clearly :-) One possible factor here (and I know there are others!) could be (IMHO) the compression driver sounds cleaner than the Lorantz woofer and the older config crosses over into it lower and steeper... i wish this was the case, but im not sure such a good idea to run them lower than 800hz. although your measured fs indicated around 400hz. ill try to simulate around 600hz Link to comment Share on other sites More sharing options...
zog Posted March 24, 2014 Author Share Posted March 24, 2014 ok good to hear, although im expecting an increase of coherency. how do you listen to them? i mean do you toed them in? They slightly only toed in like I described in post #5, which which I now realise bad description as the approx 22 deg angle I describe there is relative to the listening position to perpendicular from the drivers. If being flat against the wall is zero, they would be toed inward about 10 degrees. I have moved them off the wall a bit more to about 14cm since the first pics. i wish this was the case, but im not sure such a good idea to run them lower than 800hz. although your measured fs indicated around 400hz. ill try to simulate around 600hz The compression drivers themselves are reported to be able to run very low for home hifi use (reading about the very similar FaitalPro HF140 used in Cornscalas etc), I don't think I've done thorough measurements with the XT1464, though just from casual use it seems fine running lower than the 800hz the specs mention (for pro use?) - yet again, proper measurements outside will shed light here. other stuff: Just noting here for future reference an ARTA case study with the XT1464 horn, in German: http://www.artalabs.hr/AppNotes/PAF212_Dokumentation_ARTA_User_Case_Study.pdf includes directivity heat map for the XT1464 which indicates the WG is OK lower than 800hz A french DIY project in the same ballpark (but JBL woofer) http://www.homecinema-fr.com/forum/diy-enceintes/hr-jbl-2226-cp755ti-xt1464-t29991591.html ( like the natty little horn holder things he has) currently skiming the thread in translated mode in chromium browser... Another active system using the same horn with AE woofers for a retro/modern type thing: http://www.aespeakers.com/phpbb2/viewtopic.php?t=1660 Link to comment Share on other sites More sharing options...
davewantsmoore Posted March 24, 2014 Share Posted March 24, 2014 You should run the CD as low as possible IMO... this will depend on the horns low response (polar pattern)... on where you blend it best with the pattern of the woofer.... and thirdly, distortion. Link to comment Share on other sites More sharing options...
henry218 Posted March 24, 2014 Share Posted March 24, 2014 well, i cant do much with the current measurements. ill wait until you have a proper gating, the phase interference just too great to make sense of the crossover Link to comment Share on other sites More sharing options...
zog Posted March 25, 2014 Author Share Posted March 25, 2014 I found this link from the (der!) minidsp site that has the "missing link" http://www.minidsp.com/applications/acoustic-measurements/loudspeaker-measurements it mentions a "secret sauce" GUI component : "hover the mouse over the display and select the "%FS" dropdown" I wasnt aware of.. once I selected this it all starts acting how I've seen it behave in HolmImpulse, etc.. ...now I feel I can drag this lot outdoors, perhaps tomorrow if the weather is good. Link to comment Share on other sites More sharing options...
henry218 Posted March 25, 2014 Share Posted March 25, 2014 u can do indoor measurements effectively as long as you dont include the reflections. Link to comment Share on other sites More sharing options...
henry218 Posted March 28, 2014 Share Posted March 28, 2014 how's the measurements coming along? Link to comment Share on other sites More sharing options...
zog Posted March 28, 2014 Author Share Posted March 28, 2014 (edited) Did some outdoors not long ago, will post shortly.... edit: here they are: I gated at 5ms, these are measured outdoors with the speaker lifted up on milk crates to try and increase distance from the ground. Still not completely happy with these as there was some kind of hole @ 437Hz which I'm not sure what it was. acd2_woofer_only_at_120cm_onaxis_with_CD-outdoors-gate5.txt acd2_CD_only_at_120cm_onaxis_with_CD-outdoors-gate5.txt acd2_CD_and_woofer_at_120cm_onaxis_with_CD-outdoors-gate5.txt acd2_woofer_nearfield-2cm-outdoors-nogate.txt I dont have a huge amount of room outside but it is bigger than the loungeroom, and I hoped the diagonal placement in the yard pointed it away from the brick walls.. a few fruit trees about but I'm hoping their foliage diffused reflections.. Hitech directional measurement rig I made some 0-45 degrees directivity plots @ 2M, I'll post those, both raw drivers (No EQ), my and with mine and Henry's configs added. Edited March 28, 2014 by zog Link to comment Share on other sites More sharing options...
zog Posted March 28, 2014 Author Share Posted March 28, 2014 And the graphs.. I have chosen the scales to try and match the ones in the first post, though its not quite apples with apples as the earlier ones were done with some miniDSP EQ in place for the compression driver only measurements, and this time I could only go to 45 degrees instead of 60 due to the extra size of speaker box covering the scales. The woofer graph I dropped the lower range a bit down as it was being truncated @ 30Hz. All smoothed @ 12/octave, taken at 2M distance on axis, with the mic up high pointed at the middle of the waveguide. compression driver and waveguide, no EQ or crossover except the 76uF compression cap: Lorantz woofer. no eq or crossover Henry's second config: The rough and ready one I have been using, crosses @ 600hz Even though these are outdoors I'm still getting some reflections from nearby objects, but the measurements are what they are.. Link to comment Share on other sites More sharing options...
henry218 Posted March 28, 2014 Share Posted March 28, 2014 (edited) zog, i prefer you apply gating, this measurements are similar to last time, it doesnt work for xover design. nevermind, i see that you have gating. cheers henry Edited March 28, 2014 by henry218 Link to comment Share on other sites More sharing options...
zog Posted March 28, 2014 Author Share Posted March 28, 2014 zog, i prefer you apply gating, this measurements are similar to last time, it doesnt work for xover design. nevermind, i see that you have gating. cheers henry If I export the impulse response or wav or some other format (or even just the REW mdat saved session file) can I get it to you in a raw enough way that you can gate as you see fit? Anyway I can gate at another time if 5msec isnt suitable, or even gate and re-upload the original indoor measurements I provided. Link to comment Share on other sites More sharing options...
henry218 Posted March 28, 2014 Share Posted March 28, 2014 the last measurements files was unusable for some reasons, i can see in my omnimic software but cant process it on PCD, Phase extractor and Freq blender. Link to comment Share on other sites More sharing options...
zog Posted March 28, 2014 Author Share Posted March 28, 2014 the last measurements files was unusable for some reasons, i can see in my omnimic software but cant process it on PCD, Phase extractor and Freq blender. Aha.. I'd copied these over to my Linux box to work on in the Linux version of REW... Unix text files not MS-DOS... here they are again, converted... acd2_woofer_only_at_120cm_onaxis_with_CD-outdoors-gate5.txt acd2_CD_only_at_120cm_onaxis_with_CD-outdoors-gate5.txt acd2_CD_and_woofer_at_120cm_onaxis_with_CD-outdoors-gate5.txt acd2_woofer_nearfield-2cm-outdoors-nogate.txt Link to comment Share on other sites More sharing options...
davewantsmoore Posted March 28, 2014 Share Posted March 28, 2014 If I export the impulse response or wav or some other format (or even just the REW mdat saved session file) can I get it to you in a raw enough way that you can gate as you see fit? Anyway I can gate at another time if 5msec isnt suitable, or even gate and re-upload the original indoor measurements I provided. Gate the measurement so it is as long as possible without including any reflected sound Or post impulse/rew, and then people can work with it themselves. Link to comment Share on other sites More sharing options...
henry218 Posted March 28, 2014 Share Posted March 28, 2014 (edited) zog, the tweeter and nearfield woofer measurements are good, but the woofer farfield have some issues on 420hz (sharp dip). i did the xo, you can have a go with this W : BW 6db @200hz T : BW 6db @1500hz with EQ @1650hz -3db Q:1 TF @ 680hz, with approximate 30db null. this should get you started for lower point xo and good phase alignment. this xo didnt address the 420hz dip which i think its a measurements artifact cheers Henry Edited March 28, 2014 by henry218 Link to comment Share on other sites More sharing options...
zog Posted March 28, 2014 Author Share Posted March 28, 2014 Thanks, I'll try shortly!Yes the 420 Hz dip I think is an artifact from it being up on the crates/turntable, I took it down and it was gone. 1 Link to comment Share on other sites More sharing options...
davewantsmoore Posted March 28, 2014 Share Posted March 28, 2014 Yes the 420 Hz dip I think is an artifact from it being up on the crates/turntable, I took it down and it was gone. If possible you would use near field data to avoid these issues. .... but near field data is only valid up until the frequency = 10950 divided by the speaker (or baffle) width in centimetres. This means that above that frequency (quite low)... you need to make proper free field measurements away from boundaries (ie. no reflections) These are the trials of measuring large woofers :-( 1 Link to comment Share on other sites More sharing options...
henry218 Posted March 29, 2014 Share Posted March 29, 2014 (edited) zog, try this, it may be better than the last one (as i omitted the 420hz dip ; as you confirmed to me that it was an artifact) W : BW 18db (3rd order) @350hz T : BW 12db (2nd order) @700hz INVERTED (assuming you supplied me with all correct polarity @ measurements - check it with measurements after applying the xo) EQ @1650hz -3 to -4db to taste Q: 0.8 cheers Henry Edited March 29, 2014 by henry218 1 Link to comment Share on other sites More sharing options...
henry218 Posted March 29, 2014 Share Posted March 29, 2014 here is another one W : BW 18db (3rd order) @450hz T : BW 12db (2nd order) @800hz INVERTED (assuming you supplied me with all correct polarity @ measurements - check it with measurements after applying the xo) EQ @1650hz -3 to -4db to taste Q: 0.8 play with this 3, let me know which sounds best, based on the sim there's one that have the best phase alignment. 1 Link to comment Share on other sites More sharing options...
Recommended Posts