Question for Thomas :)

[John Sayers]

Mate - as I remember it to test a speaker's frequency response the old way was to bury the speaker in the ground in an open paddock with the front face of the speaker flush with the ground (hence the term flush mount) - a mic was then hung over the speaker and the response measured. The idea being that the ground absorbs any sound generated in the opposite direction of the front face of the speaker and the open air returns no other reflections.

Nowadays thay use an anechoic chamber.(for those who don't know this term and anechoic chamber is a room with extensive absorption wedges that absorb ALL the sound - i.e. a reverb time of Zero and no reflections) see attached photo.

They now place the speaker in the chamber and put a mic in front and measure the response.

Surely therefore the chamber is acting like the flush mount in that none of the rear projected sound is reflected back (i.e. like the ground) and the plane in front of the speaker also acts like the open air with no additional reflections.

Thus am I wrong in concluding that the printed frequency response of a speaker system as presented by a manufacturer, measured in a chamber, is in fact the response of the speaker in a soffit mount??

cheers
john

[barefoot]

John,

The first test method you mentioned is still used extensively, especially among smaller companies (like my own ) who can't afford an anechoic chamber -yet. It's called a "half-space" or "infinite baffle" measurement and it's a great technique. Actually the ground acts as a reflector rather than an absorber, creating an effectively infinite speaker baffle. The sky provides a near perfect anechoic space. The only downsides are ambient noise and the occasional rain delay! This is the measurement that most effectively represents the direct field response of a soffit mounted speaker. You can also approximate this measurement in an anechoic chamber which has one solid reflective wall.

The measurement setup in the picture is different. It's a true anechoic "full-space" measurement. This method most effectively represents the direct field response of a free standing speaker.

As far as the specs published by many speaker manufacturers, I tend to look at them with a great deal of skepticism. One would expect that the measurement technique used to obtain the published response curves would match the speaker's intended application - i.e. half-space measurements for speakers intended to be soffit mounted and full-space measurements for free standing speakers like nearfields. However, especially when it comes to some of the budget monitor brands, it's anybody's guess how they cooked up some of their "data". I can tell you this for sure. Data smoothing is used very liberally by many studio monitor manufacturers.

I can also tell you that virtually every nearfield monitor incorporates some sort of baffle step compensation. This flattens the free field response, but creates a very significant mismatch in the high and low frequency power (ambient) response. Unless the monitor specifically provides a setting for soffit mounting, I wouldn't trust that any of the EQ or "acoustic space" settings will result in a flat soffit mount response.

Hope this helps!:D

Thomas

[John Sayers]

Thanks for the reply mate - so how do they get the frequency response of a monitor like the Mackie that has a rear port?? In an anechoic chamber there would be no rear reflections to measure?:?

cheers
john

[barefoot]

Speaker drivers (or any other wave emitters) want to radiate omni directionally if their diameter is significantly smaller (about 1/5 or less) than the wavelengths they are emitting. If they are mounted in a large baffle, however, the baffle will force the otherwise omni directional wavefront to propagate only in the forward hemisphere, +6dB louder since the energy that would have gone in the rear hemisphere is added to the front. (You can angle the baffle in further and this becomes a horn, again forcing the otherwise omni directional wavefront into an even smaller solid angle). However, there are even lower frequencies whose wavelengths are large compared the baffle's dimensions. These frequencies will simply wrap around the baffle as if it weren’t even there. At these frequencies the driver still radiates omni directionally despite the baffle. Such is the case with the Mackie passive radiator.

A passive radiator (PR) only contributes to the output at the very lowest frequencies of the speaker's range. The Mackie PR produces little to no output above 100Hz. The wavelength at 100Hz is about 3.4 meters. This is at least 10 times as large as the largest dimension of the PR. This also happens to be at least 10 times as large as the baffle the PR is mounted in. At these wavelengths, then, the only acoustic effect the cabinet has is to isolate the rear of the speaker or PR cone from the front of the cone. Otherwise, these sound waves act as if the cabinet weren't even there and radiate equally in all directions.

So you don't need a rear wall for the low frequencies to reflect from. They radiate in all directions. The high frequencies on the other hand can only radiate in the forward hemisphere due to the speaker baffle. And like we discussed above, they will be boosted by +6dB because the energy that would have otherwise headed in the rear hemisphere is added to the front. Hence, we come back around to the whole free standing versus flush mounted argument. Anechoicly a free standing speaker has half of the bass energy escaping out the back, or half of the treble energy is being forced to the front - depending on how want to look at it. To flatten out the anechoic response you then have to either boost the bass output, or cut the treble. This is fine in an anechoic chamber, but when you put the speakers into a real room with real reflections the reverberant response is now all out of balance.

By flush mounting you force all of the frequencies, high and low, to radiate only in the forward hemisphere. No boosts or cuts are necessary to flatten out the semi anechoic half-space response and the reverberant response also remains flat.

[There are important psychoacoustic reasons to distinguish the direct field (anechoic) response from the reverberant field response, but that's a whole other story.]

Thomas

[John Sayers]

Thanks mate - that clears it up for me

cheers
john

[Chaz]

So Thomas.....

I am currently using JBL LSR28P's….. which have a rear port….. for both mixing and mastering.

According to your design experience..... what is better for this specific monitor..... free-standing or soffit-mounted? (btw.....I see that the studio John designed for Sjoko has the same LSR28's, which ARE soffit-mounted in front of the CR.)

I have also read on Steven Klein's site that he is against soffit-mounting monitors..... which is also reflected in his studio designs as well.

I am in the process of designing a complete studio at a new location using John's SAE information and would like to understand the soffit vs. free-standing debate a little better. And frankly….. all of this is very confusing to me, as I am trying to make both an informed and intelligent decision. It seems like the more research I do, the more confused I become regarding this issue.

I do have a pair of JBL 4312's (front port) that I will probably have re-coned for the new CR, but at the present time, I have to design the room around the LSR28P's.

Your insight, help and guidance regarding this matter would definitely be appreciated! The same goes out to John Sayers and Steve (knightfly).

BTW..... I really appreciated your thread on Reflection Free Zone (RFZ) control room design with flush mounted monitors. What is the application for free-standing monitors and how would the front wall be treated?

I know, I know....I am so full of questions. I better take it one question at a time. However, I did find some interesting comments you made about rear ported monitors in your July 29th post found here..... http://www.johnlsayers.com/phpBB2/viewtopic.php?t=448 . Still....

[barefoot]

Chaz,

I wasn't familiar with Steven Klein's site. I read his statement about soffits being used as a method of gaining volume. While volume may have been a consideration back then, this isn't the fundamental principle behind flush mounting. Flush mounting has two major benefits. It aids in the creation of an RFZ without the necessity of a highly absorbent front wall. It also extends the front speaker baffle, forcing the long wavelengths to radiate only in the forward hemisphere the same way the short wavelengths are forced to.

If you download my Wall Bounce Calculator http://www.johnlsayers.com/phpBB2/viewtopic.php?t=610 you can start to get a feel for the issues with free standing speakers and front wall reflections. You can also see that you can't simply just flush mount a speaker intended to stand free. You need a shelf filter to correct for the rise in bass response.

Sjoko flush mounted his LSR28's, but I doubt his setup is without issues. Here is my guesstimate of what his response looks like. The flush mounting probably has effectively plugged up the rear port. This would change the low end response from a 4th order roll off at 46Hz (green curve) to a 2nd order roll off somewhere in the region of 70Hz or 80Hz (yellow curve). This new higher cutoff response combined with the bass response step induced by flush mounting probably yields something similar to the blue curve.

By plugging up the port and flush mounting he probably has as much or more low end response as before, but now he has a big low end bump as well.

If you really want to flush mount then I suggest an open back design similar to what I described in this thread http://www.johnlsayers.com/phpBB2/viewt ... highlight= combined with a shelf filter to flatten out the response.

Hope this helps.

Thomas

[Chaz]

Thanks Thomas for clearing up some things for me.

The reason why I was asking about soffit-mounting vs. free-satnding the LSR28's is because I was having a difficult time understanding how one, like Sjoko, could flush-mount a rear-ported monitor without having any problems as a result; especially when the LSR28 is designed as a free-standing monitor.

So.... The confusion is now resolved and I must design the room, for a free-standing monitor setup.

I know I can soffit-mpount the JBL4312's, as they have a sealed back with the port in front. But, as I mentioned, I have to design the new studio right now for the LSR28's.

I guess I need to address my next question to John and/or Steve with regards to designing a front absorbant wall fror free-satanding rear-ported monitors. I will do so in the "Construction" forum area.

Thanks again, Thomas!