Speaker measurement with WM-61A - diyAudio
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 24th August 2010, 12:05 PM   #1
zdr is offline zdr  Belgium
diyAudio Member
 
zdr's Avatar
 
Join Date: Nov 2005
Location: Brussels
Default Speaker measurement with WM-61A

I am trying to setup my DIY measurement system with WM-61A modded capsule and Linkwitz's mic amp. I built it sometime ago but gave up due to Speaker Workshop complexity.

I have now installed Arta and trying to get this up and running again. I have SB Live! lying somewhere, and for now have set it up using built in Realtek sound card.

I tested the mic by simply recording my voice and playing it back, and it sounds ok to me. Then I tried frequency response test (FR1) with quite low level pink noise played from GC to Hatt-MKIII and this is what I got.

I could not play pink noise loud in the middle of night, but I even don't know how loud should I play, where to place the mic, what should I expect as result, if mic and amp are working OK, etc. In short, too many unknowns in one equation How this graph looks to more experienced diyers?
Attached Images
File Type: gif response.gif (32.0 KB, 388 views)
  Reply With Quote
Old 24th August 2010, 12:46 PM   #2
just another
diyAudio Moderator
 
wintermute's Avatar
 
Join Date: Aug 2003
Location: Sydney
Blog Entries: 22
Hi Zdr, it would probably help to know how the measurement was done, Without knowing this it is difficult to comment.

I'm not familiar with ARTA, but the fist thing that strikes me is that it doesn't look like a typical gated measurement, or perhaps it is but the gating has been set to include some reflections, hence the "fuzzy" look at the higher frequencies.

It looks too clean to be a raw in room measurement, but it does look like it includes room interations.

Certainly I would say for a first try you are on the right track.

As a starter I would suggest placing the speaker in the middle of the room (removing any hard furniture if you can. try and get it 1/2 way between the flloor and the ceiling and as far from any other objects as possible. put your mic on axis one meter away.

As far as levels go, one way to set it is to put your multimeter (on AC) on the output of the amp when you are playing the signal, Set the levels to get 2.8V and you will probably be ok (remove the multimeter after setting the levels as It can effect the results). Wear hearing protection too!! I was surprised just how loud 2.8V is!!

Even though it is a different program, you could check out the Holm Impulse threads here as well as there are plenty of tips about doing measurements in them.

Tony.
__________________
Any intelligence I may appear to have is purely artificial!
Some of my photos
  Reply With Quote
Old 24th August 2010, 04:03 PM   #3
zdr is offline zdr  Belgium
diyAudio Member
 
zdr's Avatar
 
Join Date: Nov 2005
Location: Brussels
Thanks Tony, I was just wondering about that fuzzy look Speakers are sitting in a tiny room of 4m2, right next to the rear walls and on the table. I will try next to give them much more space and set the output level as you suggested.
  Reply With Quote
Old 24th August 2010, 04:28 PM   #4
cuibono is offline cuibono  United States
diyAudio Member
 
cuibono's Avatar
 
Join Date: May 2006
Location: City of Angles
It sounds like your just trying to test you mic for functionality - and it seems to be good. Your graph looks normal-ish to me, and if your voice sounds fine, I'd say your good to go.
  Reply With Quote
Old 24th August 2010, 08:04 PM   #5
zdr is offline zdr  Belgium
diyAudio Member
 
zdr's Avatar
 
Join Date: Nov 2005
Location: Brussels
This is the test of Linkwitz Pluto, 1m distance. Flat down to 20Hz? I don't think that can be accurate...
Attached Images
File Type: gif response2.gif (37.6 KB, 326 views)
  Reply With Quote
Old 24th August 2010, 08:04 PM   #6
zdr is offline zdr  Belgium
diyAudio Member
 
zdr's Avatar
 
Join Date: Nov 2005
Location: Brussels
Forgot to add - this is now with Holm Impulse... Arta is still fuzzy.
  Reply With Quote
Old 24th August 2010, 10:58 PM   #7
diyAudio Member
 
5th element's Avatar
 
Join Date: Oct 2002
Location: England
Arta is fuzzy because you're not measuring with it correctly.

You need to switch it into impulse mode and make a measurement using an MLS signal.

If you do that you will end up with an
impulse response on the screen and looks like the first graph in the attached image.

Notice the arrows.

The yellow one shows the start of the impulse response.

The white one shows the end of the impulse response.

The grey arrow shows the first serious reflection from something and is likely to be the closest hard reflective surface between the mic and the loudspeaker.

Arta allows to you to gate the measurement and let in, or keep out, as much influence (reflections) from the room as you want.

Left click allows you to set the start point, on the second graph, this is the first grey arrow, and as you can see it's set at the start of the impulse.

The end point you place with right click. I decided to use the right hand grey arrow as my end point.
The corresponding frequency response is displayed below.

Note that the FR is still slightly fuzzy, this is because I have a sub optimal setup for measuring as there are hard reflective surfaces all around the measuring area.

Now note the position of the red arrow. Below it is a green line that goes just beyond 300hz. This is showing that due to the short gate length we can only get accurate data to around 300hz.

Now lets try extending the gate end point to include the first reflection.

As you will notice in the last graph the frequency response is a lot fuzzier, this is because we've included the reflection. Also note at the white arrow the same green line has now moved down closer towards 200hz.

As the gate length increases so does the low frequency boundary, but is it accurate? No of course not, it's including the reflection and you don't know if the bumps in the response are a reflection or really produced by the loudspeakers.

To remove reflections you can perform what's known as a near-field response. To do this you place the microphone next to the loudspeaker cone and measure, make sure nothing is clipping though!

This will create a much cleaner impulse response and allow you to set the gate much longer. This gives you very accurate measurements but doesn't take into consideration baffle effects, such as diffraction. These are useful for making low frequency measurements.

To make measurements that include how the baffle affects the sound you have to perform a far field response. As we have seen though, these have limited low frequency accuracy. The common thing to do is make both far and near field measurements and splice the near-fields accurate low frequency onto the far-fields high frequency response. You will need to most like add in the simulated effect of baffle step.
Attached Images
File Type: jpg impulse.jpg (371.7 KB, 311 views)
__________________
What the hell are you screamin' for? Every five minutes there's a bomb or somethin'! I'm leavin! bzzzz!
  Reply With Quote
Old 25th August 2010, 03:57 AM   #8
just another
diyAudio Moderator
 
wintermute's Avatar
 
Join Date: Aug 2003
Location: Sydney
Blog Entries: 22
Hi ZDR,

The flat to 20 Hz is a side effect of the gating. The gating in that measurement means that frequencies below about 1.3Khz should be ignored, as they have no resemblance to reality . If you move the gating out to just before that 60cm blip you should get a bit more low freqency resolution. The Gating indicator should move lower in freq which will show you where the new cutoff point (below which the fr data is suspect).

Tony.
__________________
Any intelligence I may appear to have is purely artificial!
Some of my photos
  Reply With Quote
Old 25th August 2010, 09:12 PM   #9
zdr is offline zdr  Belgium
diyAudio Member
 
zdr's Avatar
 
Join Date: Nov 2005
Location: Brussels
I played around a bit more. Here are the responses from Arta on two Plutos, 1m but low volume. Fuzziness is gone with "octave smoothing" setting on. I have some worrying differences around 1k which are pretty consistent.
Attached Images
File Type: gif pluto response.gif (28.9 KB, 263 views)
  Reply With Quote
Old 2nd September 2010, 12:09 PM   #10
zdr is offline zdr  Belgium
diyAudio Member
 
zdr's Avatar
 
Join Date: Nov 2005
Location: Brussels
Default good cheap sound card?

Is there a cheap good sound card that works well with Arta or any other measurement sw out there? I tested my built in AC97 which comes out with very low THD but high noise from mobo, resulting in too high THD+N = 0.18%. SB Audigy I had lying around produced some ridiculously high figures for THD alone, and old SB live threw my Windows 7 into BSOD.

So here I am, looking for good alternatives.
  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
School project wm-61a question paintrodsey Solid State 5 23rd March 2008 11:14 PM
Panasonic WM-61A, where to buy in Australia? Foxx510 Parts 0 23rd April 2007 04:02 AM
WTB Panasonic WM-61A neilyboy84 Swap Meet 3 31st January 2007 03:41 PM
Stereo binaural WM-61A with preamps AdamZuf Everything Else 21 30th December 2006 04:34 AM
finished WM-61a measurement mic scone Multi-Way 3 14th March 2005 08:52 AM


New To Site? Need Help?

All times are GMT. The time now is 01:21 AM.


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