Go Back   Home > Forums > Loudspeakers > Multi-Way

Multi-Way Conventional loudspeakers with crossovers

Please consider donating to help us continue to serve you.

Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving
Reply
 
Thread Tools Search this Thread
Old 4th February 2014, 07:53 PM   #11
ScottG is offline ScottG  United States
diyAudio Member
 
Join Date: Feb 2003
Location: US
A few things I'd add:

1. Lift the speaker up more than half a meter (usually with books) and raise the mic by the same amount.

2. Add a lot more pillows around the loudspeaker and between the mic.

3. Start with shorter gate (3ms), and expand on it until you start running into trailing noise (and back-off from that as appropriate).

Remember the longer the gate (that's clean), the higher the resolution (or the less "averaging" that's taking place) at lower freq.s within that window.

Here is Dan's info. on REW and gating:

audio blog: Gating loudspeaker measurements

Here is ARTA's info. on gating:

http://audio.claub.net/tutorials/FR%...ing%20ARTA.pdf

and on measurements in general:

http://www.artalabs.hr/AppNotes/AP4_...d-Rev03eng.pdf

http://www.artalabs.hr/AppNotes/AP6_...EngRev1.01.pdf
__________________
perspective is everything

Last edited by ScottG; 4th February 2014 at 07:58 PM.
  Reply With Quote
Old 4th February 2014, 08:01 PM   #12
diyAudio Member
 
Join Date: Jan 2002
Location: Wellington, New Zealand
search on ground plane measurement techniques. This effectively doubles or quadruples the first reflection point (meaning a lower gate frequency). Downsides is artificially high efficiency (is it +3dB) due to ground plane loading plus you lose baffle diffraction effects. It might be useful anyway as a "2nd opinion" measurement technique (I try and measure multiple ways to confirm behaviour).

You could raise the speaker off the floor (midpoint between floor / ceiling) to gain a few miliseconds. Based on software though - this may not increase resolution or decrease measurable frequency, if the software (like speakerworkshop) measures in "bands".

If you can - try and use a balcony outside with a mic on a boom.

What xover point are you targetting? That will give you a good measurement goal.

composit measurement of a nearfield woofer and farfield gated tweeter will work but remember you lose relative phase (separate measurements with mic being moved) so need to re-extract minimum phase and enter driver offset (since VCs are not time aligned?) when modeling your xover
  Reply With Quote
Old 4th February 2014, 08:08 PM   #13
diyAudio Member
 
TMHutson's Avatar
 
Join Date: Dec 2010
Location: SoCal
I have a seos12/dna360.
Here is how I measured...

I did a nearfield on the woofer just for the heck of it. Don't need it for the crossover.
But first, I put the microphone on the tweeter axis. 3 Feet away.
Measured the tweeter with no xover.
I did not touch ANYTHING, including volume of signal output.
I then measured the woofer.
If you know how to do it, I would measure the tweeter first with detect time zero.
After you get the measurement, then time zero lock and press "use" on "last detected".
This is on Holm... Not sure on REW. My computer can't handle any more programs.
I would check it out if I could...

This will allow you to set your acoustic offset to perfect.
One benefit of having non baffled horn! And you don't have to enter the offset in PCD7 once you physically align them. One less thing to enter...
I know you are going active, but I figured I would mention PCD anyways.

What woofer are you using?
I am using a crappy Cerwin Vega.
I want to upgrade in the future.
Curious what your measurements are once you get it all ready for digital crossover.
I have the same seos12/dna360, so I can post up measurements just to compare.
  Reply With Quote
Old 4th February 2014, 08:23 PM   #14
Defo is offline Defo  Norway
diyAudio Member
 
Join Date: Nov 2008
Here are the measurements zoomed into a 10 ms window. What I'm suprised of, is that the farfield and nearfield impulse response of the woofer is nearly the same...
I will try to raise the speaker, add more pillows, and raise the gain, and see how that changes the measurements.

My target crossover point is 1600hz in order to match directivity with the woofer (B&C 10HPL64).
The horn is the SEOS 15 btw, with B&C 250. Voice coils are time aligned, or are at the same vertical plane at least.
Attached Images
File Type: png HF IR.png (167.9 KB, 76 views)
File Type: png LF farfield IR.png (136.3 KB, 75 views)
File Type: png LF nearfield IR.png (142.5 KB, 73 views)

Last edited by Defo; 4th February 2014 at 08:29 PM.
  Reply With Quote
Old 4th February 2014, 08:51 PM   #15
diyAudio Member
 
TMHutson's Avatar
 
Join Date: Dec 2010
Location: SoCal
Looks like there is your first big reflection just after 2m.


I would look into measuring the acoustic centers as they don't always match the voicecoil center... Then you know you are perfect.

Now that you are measuring, save everything. You are going to get crazy with all the measuring.
Even with just one little speaker, I have tons of measurements from one short session.
I was trying tape on tweeter edges, removing grilles, different baffles, different stuffing. I went nuts. And learned a lot. I am trying to learn about as much as I can as to what effects the frequency response and what is worth as far as modification. I don't want to have to eq much at all.

Measure with and without pillows just to see if it is even helping.

After you get all set with your measuring set up, don't forget to try different stuffing in your cabinet.
I was surprised how incredible one speaker sounded, and how incredibly flat it measured in midrange up to about 1.7k after I removed the back of the cabinet and stuffed the heck out of it. I left it open back. It had some polyfill in the regular cabinet, and adding a bit more didn't help.
I took the back off and stuffed a complete wool blanket in there.
Sure, it killed the bass, but for a midrange it would be incredible.

Have fun playing, now.


You will notice in the nearfield that the response below about 500-ish is really nice with very little to no reflections due to low volume of signal. But the highs starting at around 1k or so has weird responses due to the mic picking up competing signals from each side of the outer edge of cone. So you will have weird cancellations. This is why nearfield is not good for highs. Mainly for bass.

There is a good thread about Jeff Bagby and another guy who made a nice nearfield-farfield splicing program to allow you to have both on one FR graph. I have not gotten that far yet in my design.

*edit.

The tweeter is much more precise and the highs reveal more, so I set the gate with the seos impulse response.
  Reply With Quote
Old 4th February 2014, 08:52 PM   #16
ScottG is offline ScottG  United States
diyAudio Member
 
Join Date: Feb 2003
Location: US
Quote:
Originally Posted by Defo View Post
I will try to raise the speaker, add more pillows, and raise the gain, and see how that changes the measurements.

..also move it away from the wall and the chair in the picture.
__________________
perspective is everything
  Reply With Quote
Old 4th February 2014, 09:08 PM   #17
diyAudio Member
 
TMHutson's Avatar
 
Join Date: Dec 2010
Location: SoCal
I wanted to pull my speaker farther from back wall, but there was a ceiling fan that hung really low in middle of room. I hate that thing.
Front baffle is 5 feet from back wall.
Center of baffle is 5 feet from wall on each side.
Later, I made a table to raise the speaker up a little higher to take advantage of 10 foot ceilings.
Attached Images
File Type: jpg measuring CV redux.JPG (38.3 KB, 73 views)
  Reply With Quote
Old 4th February 2014, 09:14 PM   #18
Defo is offline Defo  Norway
diyAudio Member
 
Join Date: Nov 2008
Should the placement of the mic be on axis with the tweeter?
Or so that the length from each drivers acoustic center, to the microphone is equal?

I reckon the latter will be more correct for the midrange area, which is the area of interest here (1600 hz XO).
  Reply With Quote
Old 4th February 2014, 11:17 PM   #19
diyAudio Member
 
TMHutson's Avatar
 
Join Date: Dec 2010
Location: SoCal
Yes to first one, as that is how you are going to listen to it. On axis with tweeter.
Also, when I put the measurements in PCD, I don't have to enter the acoustic offset, as I measure it at 0 and can enter the listening (and mic) distance along with driver vertical alignment. It will sim different listening distances.
You don't listen to the woofer on axis, so why measure it on axis?
But I am new to this, so others will chime in, I am sure.

You are going active, so you need to know how to measure the acoustic offset.
I think it's better to measure farther away (actual listening position) as the two drivers become more aligned as they get farther away. No one listens to most floor standing speakers from 3 feet away (especially a big waveguide or horn). Small nearfield mixing monitors are different, obviously.
Another nice thing about active, is that you can measure at listening position to account for room interaction.
  Reply With Quote
Old 5th February 2014, 03:42 PM   #20
Defo is offline Defo  Norway
diyAudio Member
 
Join Date: Nov 2008
Made some new measurements with elevated gain, longer distance (1m), and experimented with and without pillows. I get the smoothest impulse response using pillows, so I stuck with that.

I've set the gate to 3.8m, just before the reflection on the impulse response.
What boggles me, is that the response is so much nicer without a gate... Makes it hard to trust the results I get with the gate set to 3.8m...
Attached Images
File Type: png HF IR.png (138.2 KB, 51 views)
File Type: png HF no gate.png (78.9 KB, 25 views)
File Type: png HF 3.8m gate.png (79.7 KB, 24 views)
File Type: jpg 20140205_173124_PerfectlyClear_0001.jpg (218.9 KB, 24 views)
  Reply With Quote

Reply


Hide this!Advertise here!
Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
Loudspeaker measure and crossover design 05cordie Software Tools 22 29th June 2013 05:32 PM
Using praxis to measure transfer function of active crossover pjwd Full Range 0 2nd May 2010 07:23 AM
Easiest and fastest way to measure a crossover's transfer function? 454Casull Analog Line Level 3 26th August 2009 07:52 AM
crossover question Measure at 2500? ben62670 Multi-Way 6 1st January 2007 03:18 PM
Speaker Workshop - using "Measure -> Passive Components" to measure ESR ? percy Multi-Way 0 12th March 2006 09:18 PM


New To Site? Need Help?

All times are GMT. The time now is 02:52 PM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright 1999-2014 diyAudio

Content Relevant URLs by vBSEO 3.3.2