Acoustic measurements guidance |
Post Reply |
Author | |
Foca_Dacian
Registered User Joined: 05 February 2011 Location: Romania,Cluj Status: Offline Points: 118 |
Post Options
Thanks(0)
Posted: 10 February 2018 at 2:40am |
I am new to measurements and I'm a little overwhelmed with the different types of measurements and need some guidance. I have installed REW, I have an external sound card, condenser omnidirectional measurement mic, I followed the steps in the REW help section, calibrated the sound card,made all the necesary connections and took my first measurements of a bi-amped 12 inch+1.4 inch top using this type of measurement: Each driver at a time and then the two playing together. The whole measurement setup was: ECM 999 superlux mic on a stand 1 meter from the loudspeaker aimed at the acoustic center, connected to a Steinberg UR22 on input 1, from output 2 on the UR22 I made a loop and fed it to channel two,from channel one output I sent signal to one channel on a DBX Driverack pa2 , from that to the High out and to HF amp in and from mid out to MF amp in. Made the necesary level calibrations using pink noise and the level meters in REW ,clicked on measure and voila, i got a graph. I have seen the peaks on the HF section and corected them with the PEQ in the DBX, and measured and PEQ and measure getting it as flat as I could. Same thing with the mid section, that had a surprisingly fast roll off after 200 hz,probabily some cancelation due to the position or to much fonoabsorbant in the box,lowering tuning and LF efficiency ( just an asumption, please correct me if I'm wrong.) I have not done any gain staging between the DSP and amps, no limmiters set, just the xover point that I chose @ 1,5 Khz on both drivers with a 24 dB BW slope.(I made this decision due to the fact that the HF voicecoil is 2.5 inches( HF144 Faital) and the Beyma TD385 horn I'm using probabily sound bad lower becouse it is a 230mm x 230 mm 1.4 inch exit horn, again an asumption ,don't know if I'm right or wrong due to very little experience in X-overs.And I set the level using the input attenuators on the amps for the HF to match the woofer in SPL on the measurement graph. When I listened to a few tracks the sound got much better then before, but still needs work. Questions are, is the type of measurement that I have made a measurement of the initial sound coming from the speaker without room intereference, or I have measured the speaker and room all together? When I time and phase align , should I use delay , until the phases overlap as mush as possible@x over point? What type of measurement do I have to use for this initial setting of the speaker(phaseing,setting levels and xover point, and what type of measurement should I use when I want to see the frequency response of the room I'm trying to tune the whole system in. I know in theory what I have to do, but I don't understand how I should measure it. The system I'm trying to tune consists of 2 x MTL46 loaded with FaitalPro 18HP1010(4 ohm) , 2 x reflex tops using FaitalPro 12PR310 ,FairalPro HF144 and a Beyma TD385 Horn. Amps are, Mach Audio MA900 , 2 x 240W@ 8 ohms on the HF , Mach Audio MA1300 , 2 x 350 W@ 8 ohms on the MF, and Ecler DPA2000, 2 x 940 W @ 4 ohms on the LF. I know that subs will always have to be align, depending on positioning from tops. Some help would be grately apreciated, also point me to some books that i could read to better understand what I need to do. Have a great day! P.S. Sorry for the long post.
Edited by Foca_Dacian - 10 February 2018 at 2:41am |
|
carlosdelondres
Registered User Joined: 21 June 2012 Location: London Status: Offline Points: 243 |
Post Options
Thanks(0)
|
I am no expert, but I think you are measuring too far away for indoors. Also the mic wants to be on the floor to cut reflections. I have found a good starting point is to measure the response as close as possible to the drivers, also measure port response, with no filtering applied. Remeasure with filtering, keeping mic position the same. Then you start to have a handle on the speaker and filter response and how they interact. Of course boundary gain will throw off lf but that can be looked at later...
REW will generate eq filter suggestions to bring you to a flat or other predefined response curve, I find I tend to prefer the sound without much or any corrective eq apart from hf compensation but ymmv. Once you have some xovers, filtering set up and gains matched you can see your acoustic crossover points. Measuring the combined response is obviously not possible at very close distances, but you can still set up some delay (I use group delay at acoustic xover centre frequency as a starting point) Use the phase overlay function in REW and measure and tweak the delay until the phases traces align through the xover reason as closely as possible. With luck the rate of phase shift and direction aren't too different to get a good match for a reasonable chunk of the band. Of course, no path length difference from different acoustic sources is accounted for. So now you need to move on to further field measurements where this is minimised, and repeat the process, if possible outside well away from any obstacles. Also start to think about off axis response! Etc etc. Well, expect someone will come along and correct errors in the above soon, look forward to enlightenment Edited by carlosdelondres - 10 February 2018 at 9:33am |
|
toastyghost
The 10,000 Points Club Joined: 09 January 2007 Location: Manchester Status: Offline Points: 10919 |
Post Options
Thanks(0)
|
Do it outside, or learn to window your measurement:
https://drive.google.com/file/d/0B_65c8z6pbn2SGZWbVN5cnVpamM/view?usp=drive_web Here’s a good guide on choosing crossover point for HF to mid: https://drive.google.com/file/d/0B_65c8z6pbn2UXNMY28xWDhWR0k/view and here are some slides from a talk given by Charlie Hughes at AES: http://www.excelsior-audio.com/Publications/AES137_PD16_Measuring_Loudspeaker_Systems_rev06a.pdf The tool only responds to your questions. It does not tell you what to do. Before making any changes you should test multiple times both on and off axis (-6dB points) both horizontal and vertically, and then verify your changes at the same points to ensure you aren’t just fixing one spot. And then the same again if you choose to add more boxes - either an array (of which there are several kinds) or subs/kicks etc. |
|
Foca_Dacian
Registered User Joined: 05 February 2011 Location: Romania,Cluj Status: Offline Points: 118 |
Post Options
Thanks(0)
|
Thank you for the replies!I'll post my results after some time of study.
|
|
GeoffHill
New Member Joined: 20 February 2018 Status: Offline Points: 1 |
Post Options
Thanks(0)
|
As said in the above replies you are certainly measuring both the speaker and the room.
I would only use a microphone against or near the floor if you are using the floor as a ground plane. With a tall and relatively narrow cabinet lying on its side this can be very effective if the area is big enough to prevent nearby reflections - but these MUST BE windowed out or they will affect the response seriously |
|
Post Reply | |
Tweet |
Forum Jump | Forum Permissions You cannot post new topics in this forum You cannot reply to topics in this forum You cannot delete your posts in this forum You cannot edit your posts in this forum You cannot create polls in this forum You cannot vote in polls in this forum |