ProSoundWeb Community

Please login or register.

Login with username, password and session length
Advanced search  

Pages: 1 2 [3] 4 5   Go Down

Author Topic: Mixer summed mono voltage?  (Read 15491 times)

Art Welter

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 2201
  • Santa Fe, New Mexico
Sub Measurements
« Reply #20 on: August 05, 2014, 02:30:01 PM »

1) Do you know of a reasonably priced limiter or speaker management device that monitors and controls actual amp voltage output, peak and rms?
2)Anyway, the way the dip at 62 hz on 1 and 2 boxes disappeared with 4 boxes blew my mind.  I went back and tried 3 boxes, and found the dip gone there too.  I remember Tom saying the design was for a minimum of 6 boxes...wow, now I wish I had two more.  I sure wont use less than 3 together...unless I find these measurements to be bogus.  Your thoughts?
Mark,

1) Eminence markets a unit called D-fend that is put between amp and speaker that controls the voltage output, peak and average. It can be (must be) set for the specific parameters of the speakers used, but once set is virtually bulletproof.
2) Looks like the 2 box output at 62 Hz is simply a typo (I have done the same more than once..), most of your other readings average around a +6 dB per doubling of cabinets and power as is normal, though from 2-4 is generally a bit less, perhaps due to amplifier current limiting. As Ivan points out, your test conditions are responsible for some of the frequency response you measured, the dip at 60 Hz is larger than the usual Lab Sub response deviation.

As far as how many cabinets you use, that should simply be determined by the SPL level you require for the gig. Every room (or partial room like your deck measurements) requires it's own EQ to deal with room modes and proximity to boundaries, so the relatively small differences in EQ required between 1 and 4 cabinets are just another variation to deal with.

Art
« Last Edit: August 05, 2014, 02:36:02 PM by Art Welter »
Logged

Mark Wilkinson

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1104
Re: Sub Measurements
« Reply #21 on: August 05, 2014, 05:09:09 PM »

Mark,

1) Eminence markets a unit called D-fend that is put between amp and speaker that controls the voltage output, peak and average. It can be (must be) set for the specific parameters of the speakers used, but once set is virtually bulletproof.
2) Looks like the 2 box output at 62 Hz is simply a typo (I have done the same more than once..), most of your other readings average around a +6 dB per doubling of cabinets and power as is normal, though from 2-4 is generally a bit less, perhaps due to amplifier current limiting. As Ivan points out, your test conditions are responsible for some of the frequency response you measured, the dip at 60 Hz is larger than the usual Lab Sub response deviation.

As far as how many cabinets you use, that should simply be determined by the SPL level you require for the gig. Every room (or partial room like your deck measurements) requires it's own EQ to deal with room modes and proximity to boundaries, so the relatively small differences in EQ required between 1 and 4 cabinets are just another variation to deal with.

Art

Art,
1)  Yes, I've been looking for user reports on the d-fend...doesn't seem to have sold too well and seems to be sold out everywhere except Amazon Home and Garden Shop...really !
Should I give it a try?
2) Yes thanks, typo...should have been 94.3
3) As you and Ivan point out, boundaries and reflections make the kind of measurements I'm trying perilous.... but I did get lucky and figure out the crazy dip at 63 hz for 1 and 2 labs as per yesterday's chart...

I knew the dip couldn't be a reflection cancellation because the suckout was everywhere I walked around and listened to.  I am using steady sine waves with time to roam around and listen.
(the freakin reflections are what made me give up trying to learn Smaart indoors about 10 years ago lol )

It turns out, there must be a passive radiator suckout from the adjoining Labs that aren't being driven.
The middle two in my bank of 4 get sucked out at 63hz, either singly or as a pair when the others aren't driven.
With subs numbered 1-4 left to right, yesterday I chose (by bad luck :)  #2 as typical for one sub, and 2 & 3 together as typical for 2 subs.
Today I remeasured pairs and quickly found the only 2 sub combo that would dip at 63hz was 2 & 3.
So I removed 1 and 4 from the bank, from physical proximity, and bingo...no dip for the pair, or singly!

Here's the results of today's test .. #4 is closest to house, hence the gradual increase in levels with tests that lean towards the higher combos.
Sorry for the false alarm on the subs...the novice learns...

Oh, and John...waiting to hear from A&H....
Logged

Ivan Beaver

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 9534
  • Atlanta GA
Re: Sub Measurements
« Reply #22 on: August 05, 2014, 05:16:15 PM »



It turns out, there must be a passive radiator suckout from the adjoining Labs that aren't being driven.
.
Any time you have woofers nearby that are not hooked up-they will cause notches (suckouts) in the response.

Shorting the speaker terminals helps-but the best thing is to move them a long distance away-at least relative to the distance to the mic position.

I have spent quite a bit of time with the D fend circuit.  It works quite well-when programmed properly-that usually requires careful measurement of the speaker.  They do have a guideline that is a good start.  However there are some maximum input levels that can damage it when exceeded.
Logged
A complex question is easily answered by a simple-easy to understand WRONG answer!

Ivan Beaver
Danley Sound Labs

PHYSICS- NOT FADS!

Art Welter

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 2201
  • Santa Fe, New Mexico
Re: Sub Measurements
« Reply #23 on: August 06, 2014, 01:44:03 PM »

Art,
1)  Yes, I've been looking for user reports on the d-fend...doesn't seem to have sold too well and seems to be sold out everywhere except Amazon Home and Garden Shop...really !
Should I give it a try?
Today I remeasured pairs and quickly found the only 2 sub combo that would dip at 63hz was 2 & 3.
So I removed 1 and 4 from the bank, from physical proximity, and bingo...no dip for the pair, or singly!
Mark,

I have not actually tried the D-fend yet, but have an earlier model available if you are interested.
Proper settings of limiters would be more cost effective since you are running multiple amp channels, but for  safeguarding a cabinet against people that like to see red, D-fend seems perfect.

Since you seem to enjoy testing, another interesting test would be to compare the response of a pair of your subs with a pair of shorted cabinets flanking on either side, so the front of the stack is 135" wide by 45" tall. You will probably find about a 4 dB increase in level over the single pair, if you measure away from boundaries.

The "cool" thing about that +4 dB gain is it requires no additional power, therefore no thermal compression, which is a weak point on the Lab 12 in a small compression chamber.

Art
Logged

Mark Wilkinson

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1104
Re: Sub Measurements
« Reply #24 on: August 06, 2014, 05:16:18 PM »

Mark,

I have not actually tried the D-fend yet, but have an earlier model available if you are interested.
Proper settings of limiters would be more cost effective since you are running multiple amp channels, but for  safeguarding a cabinet against people that like to see red, D-fend seems perfect.

Since you seem to enjoy testing, another interesting test would be to compare the response of a pair of your subs with a pair of shorted cabinets flanking on either side, so the front of the stack is 135" wide by 45" tall. You will probably find about a 4 dB increase in level over the single pair, if you measure away from boundaries.

The "cool" thing about that +4 dB gain is it requires no additional power, therefore no thermal compression, which is a weak point on the Lab 12 in a small compression chamber.

Art

Thanks Art,

I went ahead and ordered a D-Fend to see how it works.

The flanking subs setup sounds incredible !  Do the 2 "flankers" have mouths pointing away from, or into the active pair?  And where do they position to the active pair?  ie even with the fronts, or mid, or backs?  Is the +4 db gain across the whole sub range?
 
I know what corner stacking the Labs can do above and beyond the 1/4 space gain...can't wait to try this..
Logged

Mark Wilkinson

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1104
Re: Mixer summed mono voltage?
« Reply #25 on: August 08, 2014, 02:40:19 PM »

This seems wrong, since there should be a 6dB difference between TRS output and XLR unless their user advice is incorrect.

Apparently mono output is post fader, and there is a rear panel selector switch to alternate between mix 1 or mix 2.

JR

A&H says they don't know what the XLR output should be under the conditions I give without a unit to test, but that there should definitely be a 6db difference between TRS and XLR.
And yes, the mono out is post fader and with a switch to choose mix 1 or mix2.......my switch is wired backwards....may be indicative of further wiring issues...?
Probably wont get service though...too many workarounds and everything else appears to work great
best, mark
Logged

Ivan Beaver

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 9534
  • Atlanta GA
Re: Sub Measurements
« Reply #26 on: August 08, 2014, 08:05:26 PM »

Thanks Art,

I went ahead and ordered a D-Fend to see how it works.

The flanking subs setup sounds incredible !  Do the 2 "flankers" have mouths pointing away from, or into the active pair?  And where do they position to the active pair?  ie even with the fronts, or mid, or backs?  Is the +4 db gain across the whole sub range?
 
I know what corner stacking the Labs can do above and beyond the 1/4 space gain...can't wait to try this..
The mouths should be positioned so that the whole thing looks like a big horn.  The exits should "flare" towards the outsides-so the "narrower part" should be towards the center.

What this does is to help change the acoustic impedance that the drivers see with respect to the air.  The cabinets help each other.  This can increase the gain and smooth out the response.
Logged
A complex question is easily answered by a simple-easy to understand WRONG answer!

Ivan Beaver
Danley Sound Labs

PHYSICS- NOT FADS!

Mark Wilkinson

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1104
Re: Sub Measurements
« Reply #27 on: August 09, 2014, 08:29:35 AM »

The mouths should be positioned so that the whole thing looks like a big horn.  The exits should "flare" towards the outsides-so the "narrower part" should be towards the center.

What this does is to help change the acoustic impedance that the drivers see with respect to the air.  The cabinets help each other.  This can increase the gain and smooth out the response.

Thank you Ivan,

Looking down on the subs, does the little diagram below picture what you mean?  If so, best angle  to use?  Could you just add plywood flares?

Also, I remember you said you spent a good deal of time with the D-Fend... Any advice/ comments for the settings below for a single Labsub with a bridged QSC capable of at least 3400 watts into 4 ohms...ie plx 3402, pl 340, etc
Double thanks !!!
Logged

Ivan Beaver

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 9534
  • Atlanta GA
Re: Sub Measurements
« Reply #28 on: August 09, 2014, 12:41:32 PM »

Thank you Ivan,

Looking down on the subs, does the little diagram below picture what you mean?  If so, best angle  to use?  Could you just add plywood flares?

Also, I remember you said you spent a good deal of time with the D-Fend... Any advice/ comments for the settings below for a single Labsub with a bridged QSC capable of at least 3400 watts into 4 ohms...ie plx 3402, pl 340, etc
Double thanks !!!
If using 4 subs I would stack them up 2 over 2-laying with the large flat sides down.  So the whole block would be 90wide x45tall.

Yes I have spent a lot of time with the Dfend.

The first thing I do is to determine the real parameters of the speaker they are to be used with.

Things such as measuring distortion per freq (to determine the point of the excursion limiting), measuring the point of power compression and so forth.

Without having real data-I am hesitant to suggest something-because if something should "go wrong", then I could be blamed for "Well Ivan said to do this" type of thing.

To even determine the voltage based on rated power requires an actual impedance measurement (NOT DC resistance) to put into the calculation.

But my "gut feeling is that your voltage are to high-based on the "simple numbers" and not measurement.

The attach and release times on the all pass are to short.

When I was considering it, I spent a lot of time measuring the actual speaker under consideration.

I came up with some pretty good settings that amazed people (myself included) with what I could do to a loudspeaker and it not tear up.  Including taking a 300 watt 8" full range speaker-driven by a 4000 watt amp into full clip (ie the clip light never went off) with music-then add a 8KHz sine wave (to simulate feedback) that was a high enough level to drive the amp to full clip and then add a 15dB boost down at 40Hz to try to overexcursion it.

It could run like this for hours with no damage.  Now I did not say it sounded good-but rather that I could not find a way to tear it up.
Logged
A complex question is easily answered by a simple-easy to understand WRONG answer!

Ivan Beaver
Danley Sound Labs

PHYSICS- NOT FADS!

Mark Wilkinson

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1104
Re: Sub Measurements
« Reply #29 on: August 10, 2014, 06:21:29 PM »

If using 4 subs I would stack them up 2 over 2-laying with the large flat sides down.  So the whole block would be 90wide x45tall.

Yes I have spent a lot of time with the Dfend.

The first thing I do is to determine the real parameters of the speaker they are to be used with.

Things such as measuring distortion per freq (to determine the point of the excursion limiting), measuring the point of power compression and so forth.

Without having real data-I am hesitant to suggest something-because if something should "go wrong", then I could be blamed for "Well Ivan said to do this" type of thing.

To even determine the voltage based on rated power requires an actual impedance measurement (NOT DC resistance) to put into the calculation.

But my "gut feeling is that your voltage are to high-based on the "simple numbers" and not measurement.

The attach and release times on the all pass are to short.

When I was considering it, I spent a lot of time measuring the actual speaker under consideration.

I came up with some pretty good settings that amazed people (myself included) with what I could do to a loudspeaker and it not tear up.  Including taking a 300 watt 8" full range speaker-driven by a 4000 watt amp into full clip (ie the clip light never went off) with music-then add a 8KHz sine wave (to simulate feedback) that was a high enough level to drive the amp to full clip and then add a 15dB boost down at 40Hz to try to overexcursion it.

It could run like this for hours with no damage.  Now I did not say it sounded good-but rather that I could not find a way to tear it up.

Thanks for the continuing help. I'm gathering there is a need to find the real world edges to protect what I'm hearing or smelling haha ....and then making D-Fend (or limiter) settings based on measuring those edges ....

But to try know where to look for the edges, I've been running impedance sweeps on single boxes.  And then on 2, 3, and 4 boxes in series, to see how acoustic coupling effects their impedance.
Also ran it on 4 in series / parallel just to check vs readings vs 4 boxes in series, average per box
   
The lowest dip I could get on any one box was 2.6 ohms at 42hz, but my tester is 1/3 octave so it could sure be a little lower for a 42hz sine wave. 2 boxes gained .3 ohm and 4 boxes took it 3.1 ohms.  I was happy to see any rise.
I mean, I'm gonna get in thermal trouble where current is highest, right?

So shouldn't I set RMS limiting based on a 3 ohm load, and play with raising it till I measure compression or distortion  ....particularly in the low 40's hz?

And for 4 boxes, I'm willing to call them 5 ohm nominal....graph below....
I'm thinking of using the 5 ohms for initially choosing/calculating peak voltage settings? Not sure how to really tell when to limit peak till I tear something up....
Peak is about mechanical failure, right?

Is mechanical distortion more likely to be audible at a particular frequency based on impedance, (or other criteria)?

Art, thx for your reply in the subwoofer forum.  I'm trying to reconcile the number of people who have reported using bridged amps on single labs (when in a bank), with wattage typically in the 3000 @ 4 ohm category. 
Maybe they just never got hammered with low 40 hz music? ... assuming I'm thinking straight....
Logged

ProSoundWeb Community

Re: Sub Measurements
« Reply #29 on: August 10, 2014, 06:21:29 PM »


Pages: 1 2 [3] 4 5   Go Up
 



Site Hosted By Ashdown Technologies, Inc.

Page created in 0.019 seconds with 24 queries.