ProSoundWeb Community

Please login or register.

Login with username, password and session length
Advanced search  
Pages: [1] 2  All   Go Down

Author Topic: mutli Lab Sub Measurements get together in Michigan!  (Read 3887 times)

Timmahh

  • Full Member
  • ***
  • Offline Offline
  • Posts: 131
    • http://www.chromepony.us
mutli Lab Sub Measurements get together in Michigan!
« on: June 05, 2004, 06:42:08 pm »

In the process of building 5 or 6 pair of Labs.  4 pair for a guy, and 1 or 2 pair for our personal use.  When we have 5 pair complete, was wondering if any of the Professionals would be willing to come up to Michigan, Saginaw Bay City area, and do some stacking measurements using Smaart.
   We are thinking of starting with one pair, then 2 then 4 then 6 ect.... and taking measurements along the way to be posted for those out there wondering how they would preform in various set ups, i.e. vertical side by side, horizontal stacking ect.....  Most likely this will be done outside, unless we can find an inside place willing to donate an area large enough and with the open time to do some inside tests.
   Too Tall just posted a message about "guidelines" for use for a single Sub/ pairs/ ect... and the possible results using a modeling program.  Maybe we could work this together to get modeling reading compaired to actual use readings...

I'm sure a few of the Michiganders in this perticular area would like to make it if time permits them too.  But was hopeing some of the more Prominant people involved with this site would/could be willing to come up n lend their expertise on this endevour.   As its been stated on various posts already;
Inquiring Minds Want To Know.

We are thinking the 1st or 2nd weekend of July at this point. Of course there will need to be some co-ordination of minds, time, and the like, not to mention enough power to do this. we have some but nowhere near enough to run 4 - 6 pair. Any thoughts or interest out there?  Everyone would be welcome to attend, but are hoping some of you experts can make it too.
let me know.
Logged
I had to do it cuz I'm Just like that!!!!!
Timmahh!!!

Too Tall (Curtis H. List)

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1591
Re: mutli Lab Sub Measurements get together in Michigan!
« Reply #1 on: June 07, 2004, 12:22:53 pm »

Hi Tim,
   I just read a reply from Mark Seaton on this subject. I wrote the reply below before Mark answered, but neglected to post it till now.
   Go take a look at Mark's last post.
   I agree with his points completely. Following his suggested protocol looks good to me
That said I stand by what I wrote below.
   The finesse comes in when deciding how far to push the speakers while we test them and deciding how useful the measurement is.



Timmahh wrote on Sat, 05 June 2004 18:42

In the process of building 5 or 6 pair of Labs.  4 pair for a guy, and 1 or 2 pair for our personal use.  When we have 5 pair complete, was wondering if any of the Professionals would be willing to come up to Michigan, Saginaw Bay City area, and do some stacking measurements using Smaart.
.



Well I may not be a professional, but I have played one on TV.
I don't have Smaart, but I do own Praxis, which works a little better for investigating speaker performance in any case. (Several types of Distortion measurements including something experimental)
   If you talk to Ken O. he should know I live in Bridgeport. (In any case tell Ken I said "Hey")


Timmahh wrote on Sat, 05 June 2004 18:42


   We are thinking of starting with one pair, then 2 then 4 then 6 ect.... and taking measurements along the way to be posted for those out there wondering how they would preform in various set ups, i.e. vertical side by side, horizontal stacking ect.....  Most likely this will be done outside, unless we can find an inside place willing to donate an area large enough and with the open time to do some inside tests.
let me know.



   I tried this very setup with four of Al's LAB subs some time ago.
We had the worst luck in wiring and equipment snafus you can imagine. That and other problems I won't go into.

For outdoors the problems are noise (are you in the country or downtown and next to an expressway) and wind.
   In Michigan finding a calm day out in the country is a real crap-shoot.
   A light wind not only raises the noise floor, it also can easily bottom the microphone diaphragm.
   All that being said I highly recommend outdoors for low frequency measurement.
   Makes scheduling a real problem

Finding an indoor space "large enough" will be a serious problem. To measure down to 20hz we need a MINMUM of one wavelength or about 57 feet in all direction (except down, we will measure in a half-space unless someone is going to spring for a crane) before the first reflection hits the measurement mic and several wavelengths is better. This works out to about 30 feet to a surface at an absolute minimum.
   Also while talking to Tom Danley on this very subject he recommended being at least 20 feet away from a LAB sub array. IOW the bigger the source (in this case the size of the horn mouth as we stack more and more boxes) the further the measurement mic must be away to be out of its near field effects.
So we are looking for a room 80'x 60'x30' and double is better. The ceiling height is usually the limiting factor. For this size room you can forget doing anything but on-axis measurements

   We did the original Michigan Sub woofer shootout in Fenton, Michigan kindly hosted by Joe Heslip the winter before last indoors in a room with about a 12' ceiling and found it to be a real compromise for low frequency measurement.
   
For what you are proposing indoors would not make sense unless we can use the Sliver Dome.

Timmahh wrote on Sat, 05 June 2004 18:42


   Too Tall just posted a message about "guidelines" for use for a single Sub/ pairs/ ect... and the possible results using a modeling program.  Maybe we could work this together to get modeling reading compaired to actual use readings...



It would be nice
   First we need to be clear as to what can and can not be done with a measurement program versus a modeling program.
In the horn modeling program they predict cone movement at a specific voltage input. Unfortunately I know of no way to "directly" measure how far the cone has actually moved and what is going on except in terms of distortion and SPL.
   IOW I can't shine a laser on the cone and tell how far the cone has physically moved referenced to a specific voltage input. This means I can't check the prediction model directly. Especially with a horn we are relying on the modeling program to predict cone movement referenced to volts input and SPL output.
   So we are confined to comparing SPL measured (by Praxis) compared to SPL predicted By McBean or something similar.

I know there is a direct link between SPL, the cone's Sd and how far it is traveling back and forth from center.  But it is a link away from physically measuring how far the cone is moving and if it is exceeding Xmax or more dangerously Xmech (Mechanical travel limit before the driver tears itself apart)

   Perhaps a subtle difference until you start doing real measurement experiments.

   Where the rubber meets the road is when we crank up a big power amp and run the voltage output up to the point where the horn modeling program predicts Xmax..

We can't physically tell how far the cone is moving other then through predicted SPL.

   As we play around up near the limit of the system all we can do is hope distortion shows up before the driver starts to physically hurts itself under long term use.

   For a reference on what is "safe" limit for using LAB subs under different configurations I REALLY wish we could hang a sensor on a driver that charts its real physical movement off center.
   Considering that the preferred failure mode is exceeding Xmech having a direct physical measurement would be best.

   In the mean time perhaps we will find either distortion or compression will show up at Xmax and well before Xmech..
   


Timmahh wrote on Sat, 05 June 2004 18:42


I'm sure a few of the Michiganders in this perticular area would like to make it if time permits them too.  But was hopeing some of the more Prominant people involved with this site would/could be willing to come up n lend their expertise on this endevour.   As its been stated on various posts already;
Inquiring Minds Want To Know.

We are thinking the 1st or 2nd weekend of July at this point. Of course there will need to be some co-ordination of minds, time, and the like, not to mention enough power to do this. we have some but nowhere near enough to run 4 - 6 pair. Any thoughts or interest out there?  Everyone would be welcome to attend, but are hoping some of you experts can make it too.
let me know.


For people in the sound business during the summer weekends are "workdays" (at least we HOPE they are), but perhaps a Sunday might work. for some. For me it doesn't matter. I have other issues.

   So did you guys turn on the tablesaw and make some sawdust yet?
   Once you get some working models you are welcome to bring them out to Bridgeport where I can check their performance for frequency and impedance.

Too Tall

Logged
Too Tall
        Curtis H. List    
             Bridgeport, Mich.   
        I.A.T.S.E. Local # 274 (Gold Card)
        Lansing, Mich
Independent Live Sound Engineer (and I'm Tall Too!)

John Sheerin

  • Newbie
  • *
  • Offline Offline
  • Posts: 15
Re: mutli Lab Sub Measurements get together in Michigan!
« Reply #2 on: June 07, 2004, 01:03:35 pm »

>We can't physically tell how far the cone is moving other then through predicted SPL.

Curtis,
Use an accelerometer.  I did that recently (thought I mentioned that to you in email), and it worked quite well as compared to a laser.  It works especially well considering the price and that you can't point a laser at the rear of the cone in most typical situations.  I used a Measurement Specialities ACH01 that I bought from Digikey a few years ago for about $20.  It needs a preamp but the data sheet gives a simple circuit for building one.  You could work out a chart of what accelerometer voltage gives xmax at various frequencies if you want a real-time indicator.  The things come calibrated, although you do need to account for the voltage gain of the preamp.

John
Logged

Tim Padrick

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 5008
Re: mutli Lab Sub Measurements get together in Michigan!
« Reply #3 on: June 07, 2004, 11:50:44 pm »

How about a teeny camera and an LED light source?

As far as using a laser goes, how small an apeture can it measure through?  Perhaps one could mount the unit in the access plate and look at the cone movement through the vent tunnel?

Timmahh

  • Full Member
  • ***
  • Offline Offline
  • Posts: 131
    • http://www.chromepony.us
Re: mutli Lab Sub Measurements get together in Michigan!
« Reply #4 on: June 07, 2004, 11:57:17 pm »

this makes me think of the Madison knockoff with the plexi cover, what if we could get a thick piece of plexi, like 1/4 or 1/2 in, or maybe even a thick piece of glass just to measure the cone movement. i would think I could find an associate in my area, willing to get me a piece cut in a circle. use it just for measurement. make it air tight. ect.....
thoughts
Logged
I had to do it cuz I'm Just like that!!!!!
Timmahh!!!

Brad Litz

  • Jr. Member
  • **
  • Offline Offline
  • Posts: 96
Re: mutli Lab Sub Measurements get together in Michigan!
« Reply #5 on: June 08, 2004, 12:32:43 am »

You need one of these Laser Vibrometers Shocked

http://www.kef.com/technology/new_uniq/laser.html
Logged
Brad Litz

Too Tall (Curtis H. List)

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1591
Re: mutli Lab Sub Measurements get together in Michigan!
« Reply #6 on: June 08, 2004, 11:00:46 am »

Timmahh wrote on Mon, 07 June 2004 23:57

this makes me think of the Madison knockoff with the plexi cover, what if we could get a thick piece of plexi, like 1/4 or 1/2 in, or maybe even a thick piece of glass just to measure the cone movement. i would think I could find an associate in my area, willing to get me a piece cut in a circle. use it just for measurement. make it air tight. ect.....
thoughts


   You guys are killing me here!  LOL
   I don't have a Laser measuring device and don't see one showing up here in the near future.
   Even if it did I wouldn't know how to run it.
I was just trying to make the point that I don't know how to measure Xmax directly during operation. As far as that goes I have never measured it while the driver is still. I rely on prediction programs.
   Nice idea anyway Tim. If a Laser expert shows up with his (or her) very own laser we will discuss it further.
Too Tall

Logged
Too Tall
        Curtis H. List    
             Bridgeport, Mich.   
        I.A.T.S.E. Local # 274 (Gold Card)
        Lansing, Mich
Independent Live Sound Engineer (and I'm Tall Too!)

Too Tall (Curtis H. List)

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1591
Re: mutli Lab Sub Measurements get together in Michigan!
« Reply #7 on: June 08, 2004, 11:54:24 am »

John Sheerin wrote on Mon, 07 June 2004 13:03

>We can't physically tell how far the cone is moving other then through predicted SPL.

Curtis,
Use an accelerometer.  I did that recently (thought I mentioned that to you in email), and it worked quite well as compared to a laser.  It works especially well considering the price and that you can't point a laser at the rear of the cone in most typical situations.  I used a Measurement Specialities ACH01 that I bought from Digikey a few years ago for about $20.  It needs a preamp but the data sheet gives a simple circuit for building one.  You could work out a chart of what accelerometer voltage gives xmax at various frequencies if you want a real-time indicator.  The things come calibrated, although you do need to account for the voltage gain of the preamp.

John


Hi John,
   I own that very one.
Stated calibration is 8.4mV/g @ 1kHz.
So if the "g" is gram how will that relate to "Pascal"????
The calibration files for Praxis are in "MV/PA"
One is a static weight referred to gravity and the other pressure over a surface, right?

   I have played around with it some on cabinet walls. I have never attached it to a driver.
   I can see where it would be useful real time, but I wouldn't know how to figure out how to relate it to a specific Xmax. IOW relating it to a specific movement off center in millimeters for the driver.

   I have had some problems with it. Using it with my desktop computer and Praxis I have trouble with a big peak at 60Hz.
Praxis comes with a calibration file for its use, though I can't find it right now.
Something else for me to try on my old Gauss 15"
Too Tall




Logged
Too Tall
        Curtis H. List    
             Bridgeport, Mich.   
        I.A.T.S.E. Local # 274 (Gold Card)
        Lansing, Mich
Independent Live Sound Engineer (and I'm Tall Too!)

John Sheerin

  • Newbie
  • *
  • Offline Offline
  • Posts: 15
Re: mutli Lab Sub Measurements get together in Michigan!
« Reply #8 on: June 10, 2004, 08:52:14 am »

Hi Curtis,
Peak at 60hz sounds like a grounding issue...

The 'g' in the mV/g is the acceleration of gravity - you know, like saying astronauts experience 12g's (or whatever it is).  One g is about 9.81 m/s^2.  You can relate acceleration to displacement by dividing acceleration by (2*pi*f)^2.  In my measurements, I measured rms voltage off the accelerometer (well, really the preamp) and had to multiply that by 2*sqrt(2) to get to a 0-peak value.  I'm not sure where the extra two came from...

As far as using it with Praxis, I have no experience there, but I would assume you could make it tell you what the voltage input is vs. frequency.  I did all my processing in Excel, but I was also using a signal analyzer to take measurements - it's easier to get an absolute reading, and it's also easier to cart around outside as compared to a desktop computer.  Anyway, you could either export your data or process it in Praxis - LMS will let you divide by j*omega, for example.  That would be useful to see what the cone displacement curve looks like in order to set a high pass filter and / or limiter if you measure at a high signal level.  You could also investigate multiple boxes.  Just fyi, I measured my HL10 box as having very high excursion at low frequencies as opposed to the predicted value as limited by a small sealed box.  This was with dual drivers ala the Lab horn, and I have yet to figure out what's going on.  I have some more measurements in mind, though...

If you want to get a real time reading of xmax with a sine wave input, I would just hook up you accelerometer to a DVM and work out what the voltage would be for xmax at that particular frequency.  As you get closer to that voltage, you would know how close you are to xmax.

John
Logged

Too Tall (Curtis H. List)

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1591
Re: mutli Lab Sub Measurements get together in Michigan!
« Reply #9 on: June 11, 2004, 12:08:45 pm »

Hi John,
What did you use to glue the accelerometer to the driver without harming the driver. It goes for quite a ride on a sub woofer driver.
Too Tall
Logged
Too Tall
        Curtis H. List    
             Bridgeport, Mich.   
        I.A.T.S.E. Local # 274 (Gold Card)
        Lansing, Mich
Independent Live Sound Engineer (and I'm Tall Too!)
Pages: [1] 2  All   Go Up
 

Page created in 0.141 seconds with 20 queries.