John Sayers' Recording Studio Design Forum Archive

Yeah, the shelf you have there is fine. The "inner framing" can be your existing wall or corner or whatever. My drawings weren't intended to be specific construction plans. I just wanted to illustrate the general concepts.

Yes, the soffit face is still coupled to the speaker via the wall. But it's a very indirect coupling. A more roundabout coupling path means less vibration transmission.

John has a more traditional soffit design with a box. Maybe you're talking about my suggestion to beef up the speaker cabinet by gluing or clamping extra wood to the sides?

Thomas

Hello Thomas. Hey, sorry I didn't get back to thank you right away. I had total disconnect/meltdown from the net for over a week from simply moving my modem/router/computer net from one room to another. What a nightmare. I can't believe what happened. Anyway, thats another story.

To continue this decoupling issue, I have a few more questions in that regard, but have a few other issues which are related . All which must be solved before cutting the baffle fronts so at this point I need to start my own new thread, which I will do tonight.

Thanks again Thomas, and if you have time, please look at my thread/issues and comment accordingly. Your input is highly appreciated.

Oh, btw, the issue of another "box" around the monitors will be included in my thread.
fitZ

Hi

I notice on the pictures of Dave Blauvelt's sutios he has soffit mounted Dynaudio BM15a's. Has there been any equalisation of these speakers?

I was interested in doing this a couple of years back but was told on this forum that because of the way the speaker is designed you had to have something to reduce (or increase, i can't remeber which!) the bass frequencies of the speaker when soffit mounting..

So, how do they sound? Brilliant i bet. If so I'm definitelky going to do it myself!

Cheers

Oliver

Just a question regarding the bezel, is the bezel required to retain a hard reflective face, or could it be covered in an absorptive material? Would covering it in say 1inch rigid glass be beneficial or do you want your bezel to act as an extension of the speaker?

Thanks for any help, am about to mount my Mackies,

Matt

Hi Matt,

Yes, the bezel acts as an extension of the speaker. So you do want it to be reflective at the wavelengths where it is active. However, the influence of the bezel drops to lower and lower frequencies the further you move away from the center of the speaker. If you play around with my Wall Bounce Calculator you can see this effect. Set Mf to zero to represent the flush mounting case and increase Ls to something like 2000 to clear up the effect of the side wall refection. Now increase the speaker width W to say 50 and then 100cm. You'll notice the left side of the curve gets flatter at lower frequencies. This basically illustrates how the higher frequencies care less and less about the bezel or the wall it's mounted in the further you get from the center of the speaker. For example, if you made the bezel/wall highly absorptive above say 3kHz greather than 25cm out from the speaker center, it would have almost no effect of the speaker response.

So imagine that you gradually increase the thickness of the absorptive material as you move out radially from the center of the speaker. You follow the guide of the Wall Bounce Calculator and you just make sure the bezel/wall is reflective below the frequencies where the curve begins to rise as governed by that imaginary delineation of a larger and larger speaker cabinet baffle. In this case, you won't effect the speaker response at all.

To put it in it's simplest terms, you can make the soffit face more and more absorptive as you move further out from the speaker.

Am I making sense?

Thomas

Hi Thomas,

Thanks for the reply that was exactly what I was after. Just one more question if you have time, i understand that incorporating some sort of filter is desireable in a soffit mount setup. My Mackies feature a 'half space' (speakers close to wall, -2dB shelving) and 'quarter space' (speakers in corner, -4dB shelving) setting to counter the effects of speaker placement. While the data in the manual is limited it appears from the chart that these filters start rolling off as high as 1kHz. Would i be correct in assuming that the second filter would be most suited in a soffit mount situation? Or am i better just to trial all 3 settings (ie normal, half, quarter space)?

Thanks

Matt

Theoretically, you need 6 dB of roll-off if you soffit mount, if I recall correctly, since your soffit acts as an infinite baffle.

Maybe one of the speaker experts can confirm that. I don't recall what frequency the roll-off should start at, but 1 kHz. does sound pretty high! I would suspect that it would be somewhere around 200 to 400 Hz where you'd need to start rolling off, but I'm sure someone like Andre or Ro can give you the exact numbers. I also seem to recall that it has something to do with the width of the baffle, as well.

So those "-2dB" and "-4dB" numbers on your Mackie don't seem to jibe with theory for soffit mounting. I wonder if they were just meant for having the speaker "close" to the wall, and "close" to the corner, rather than exactly in them?


- Stuart -

Hey,
this thread is like a good movie that stopped before it finnished.

Over a year and no reply to the roll off question.
My monitors only have a 3db roll off

Now I am worried and off to search through the forum for the info I need

?

The next stickie down show's you how to DIY a filter

I know it is an old post, but I decided that there were some unanswered questions from this thread about baffle step compensation.

Here is a response graph of a single driver in an enclosure on an infinite baffle (i.e. flush mounted)



If that speaker in that enclosure is mounted on a 7" baffle the bass response will drop like this at about 650hz.



So when the crossover is designed for this speaker a low pass filter is created that will drop the sensitivity of the speaker from 86dB to 80 dB, you lose 6dB of efficiency to flatten out the bass.

So when you flush mount a speaker your bass response will look like this.



Just for comparison sake here is the some driver on an 18" baffle. Notice the difference of where the -6dB shelf occurs, about 250hz.



The formula to determine where the compensation takes place is below.

f3 = 4650/x where x is the baffle width in inches.

So for the 7" baffle we used you have 4650/7=651.43hz.

Thanks Eric! That is REALLY useful stuff! Clarifies some questions I had in the back of my mind, but hadn't gotten around to researching yet.

So I would imagine that's the same equation you'd need to figure where you need to roll off the bass for any given baffle size? For example, if you are going to soffit mount a speaker and will use a parametric equalizer to roll of the bass, then this is the equation you'd use to calculate the roll off frequency, based on the baffle width, correct?

Alternatively, if your speaker has a built-in roll-off control preset at a certain frequency, then you could also use that equation to help decide on the correct baffle width?

Also, the roll off is 6 dB/octave initially, but then shelving, correct?



- Stuart -

Exactly! Since I've been doing so much studying of speaker design, I understand it so much better, I thought I'd pass that on.

thanks for 'reviving' an important topic !

A related question please ...

Lets say, breaking the 'Cardinal Rule', you do NOT have a 'flush mounted' system

However ... the monitor is placed 'close' to the front wall. EQ compensation may [might] be
useful ... are there any 'equations' available to mathematically examine this issue ??

Thanks!

he monitor is placed 'close' to the front wall. EQ compensation may [might] be
useful ... are there any 'equations' available to mathematically examine this issue ??

If your monitor is close to a wall, then you don't just have an EQ problem. What you ALSO have is an SBIR problem, comb filtering, first reflections and other issues. EQ can't do much for any of that.

- Stuart -