Preferable Frequency Response!!

[avare]

Originally Posted by Glenn Kuras

Or just cover up the desk with some REALLY thick blankets. Might not be perfect, but better then moving the desk.

How much work to go through is a relative thing. He was willing to cut up the desk for acoustics. With that level of dedication, I decided the most realistic thing to recommend was to move the desk right out of the picture.

Yes, my first thought was to cover it with absorbent material, but then I thought about his dedication. Why half way to save 30 minutes of labour when he is so dedicated? If he is willing to cut up the desk, that requires much more time than just moving it.

Andre

[barefoot]

My first concern is that these measurements might not be telling you much. I'm assuming they're just steady state frequency responses? If so, they reveal little about which response features are due to the speaker's on-axis response, speaker's power response, early reflections, ambient room response, room modes, etc. In order to take informed steps to remedy the situation you first need to understand what is going on. And to better understand what is going on requires more specific measurements (gated response for example) and the background to analyze the data.

One thing I can guess from the curves you've posted is that you have a significant floor-to-ceiling room mode. I'm guessing this because that 65Hz null seems to be independent of the mic position in the horizontal plane in all three curves. Also 65Hz corresponds to a half-wavelength of 8.8ft (2.6m) which is a very common ceiling height. And a half-wavelength mode would produce a null (rather than a peak) halfway between the floor and ceiling, very close to ear height. It's all consistent. I bet if you put the mic on the floor, that 65Hz null will flip to a peak. Give it a try.

You see, one curve or a few curves is just a start. You need to look at the data with an understanding of how sound waves "work" and then imagine mechanisms that might explain what you see. That will lead you to different types of measurements to confirm or deny your hypotheses. Without this sort of informed detective work you're really just groping in the dark.

Hope this helps,
Thomas

[davea]

Originally Posted by avare

Great stuff! The picture helps a lot.

First Reflections

Looking at the picture, you definitively have the same side reflection points (like left speaker left wall to listening position) covered. However it almost looks like you do not have the opposite side (left speaker right wall to listening position) covered. Please clarify this.

The Desk

Before you cut up the desk, take it out of its position and repeat the measurements. This will show much negative reflections are coming from it.

Mic Height

Just to confirm, you tested with the mic at 42"?

Bass Trapping

From looking at your test results, bass trapping is not a high priority at this time. We wrote already about the null at the low end. The rest of the frequency range is the highest priority.

Keep the questions coming. The more you know, the better your space will be.

Andre

Hi Andre,

1) Yep, both sides are treated for first reflections

2) The mic distance from the floor is 4'2" which is around 50". The distance from the mic to the ceiling is around the same. The ceiling above the mic is 4" of rockwool.
Also you were right in saying that the 60hz null was caused by the side walls.

3) When you say enough bass trapping do you mean no more mid/high traps too or are they considered the same thing ? As you can see I got all corners covered but still have space for more mid/high traps.

Thanks again Andre

[davea]

What I have been doing today is moving traps around the room to see how they affect the overall response. I found that moving the mid high traps on the side walls do change the response but not nearly as much as slight shifts of the speaker in relation to the mic and vice versa.

It seems that you get to a certain point where you have to settle for the best compromise. The room has its dimensions and you just end up wrestling with it, shifting the problems from one node to another.

What I am also going to do is remove the desk from the room and see how much of an influence it has on the overall freq response.

I shall post the result up when done.

thanks again for all your tips!!! Any more will be more than welcome.

[Glenn Kuras]

Originally Posted by avare

How much work to go through is a relative thing. He was willing to cut up the desk for acoustics. With that level of dedication, I decided the most realistic thing to recommend was to move the desk right out of the picture.

Yes, my first thought was to cover it with absorbent material, but then I thought about his dedication. Why half way to save 30 minutes of labour when he is so dedicated? If he is willing to cut up the desk, that requires much more time than just moving it.

Andre

good point!!

[davea]

Hi All,

Was waiting for my mate to turn up to help move the desk only to find out he was held up at work. So the next best thing was just to put towels at the reflective points on the desk. Did not expect to see much of a difference but .....

[Glenn Kuras]

Originally Posted by davea

Hi All,

Was waiting for my mate to turn up to help move the desk only to find out he was held up at work. So the next best thing was just to put towels at the reflective points on the desk. Did not expect to see much of a difference but .....

[davea]

Now I really want to see how much the desk is contributing to the overall response !!!!! Will move it out of the way and then post the results when done

[SRS]

Originally Posted by davea

Hi All,

Was waiting for my mate to turn up to help move the desk only to find out he was held up at work. So the next best thing was just to put towels at the reflective points on the desk. Did not expect to see much of a difference but .....

Interesting huh. I have experienced similar results with the same type of application of towels... (previous post) Nice to put the graphed results up to actually see it too. Good for others to see what those desk/console reflections do to the upper mids/highs. The direct sound and those first reflections off the desk are VERY close to one another, causing these peaks and nulls for sure. Cancellation at it's best. Thanks for posting that. Moving the desk completely will definitely have a significant effect. Just make sure you place your monitors at the same height, angle and distances from one another and the side and front walls. Otherwise, your test results will surely be skewed in such a way as to not have a good idea of just what you really have.

[davea]

Originally Posted by barefoot

My first concern is that these measurements might not be telling you much. I'm assuming they're just steady state frequency responses? If so, they reveal little about which response features are due to the speaker's on-axis response, speaker's power response, early reflections, ambient room response, room modes, etc. In order to take informed steps to remedy the situation you first need to understand what is going on. And to better understand what is going on requires more specific measurements (gated response for example) and the background to analyze the data.

One thing I can guess from the curves you've posted is that you have a significant floor-to-ceiling room mode. I'm guessing this because that 65Hz null seems to be independent of the mic position in the horizontal plane in all three curves. Also 65Hz corresponds to a half-wavelength of 8.8ft (2.6m) which is a very common ceiling height. And a half-wavelength mode would produce a null (rather than a peak) halfway between the floor and ceiling, very close to ear height. It's all consistent. I bet if you put the mic on the floor, that 65Hz null will flip to a peak. Give it a try.

You see, one curve or a few curves is just a start. You need to look at the data with an understanding of how sound waves "work" and then imagine mechanisms that might explain what you see. That will lead you to different types of measurements to confirm or deny your hypotheses. Without this sort of informed detective work you're really just groping in the dark.

Hope this helps,
Thomas

Thanks a lot for your thoughts Thomas !! Basically I use Fuzzmeasure which sweeps a sine wave through the freq spectrum from 20hz to 20khz.
I have tried to make my control room as symmetrical as possible and have used traps where needed and then used monitor and mic placement to get the freq response as smooth as possible especially from 200hz below.

Since posting the original freq graphs I have managed to smooth the response out much more. Looking at the waterfall output, all seems uniformed without excessive low energy buildup and RT20 times across the freq range is quite uniformed.

The next step is sawing my desk in half which I feels is really skewing the mid high response as demonstrated by placing towels over the reflective points.

I have not heard about what you talked about regarding "gated response". If you could explain what that is and how its achieved including the benefits I would be very grateful. Thanks again.

[avare]

Marvellous Davea!

Great progress you are making. For the benefit of other people reading this thread now and in the future, could you detail the changes you made since the you made the frequency plots.

Have you done some additional testing that have not detailed regarding the desk? If it is needed, then yes cut it in half. I have doubts if such extreme actions are needed though. Monitor position and selective acoustic treatment should provide the same results.

Keep the great stuff coming!

Andre

[davea]

Originally Posted by avare

Marvellous Davea!

Great progress you are making. For the benefit of other people reading this thread now and in the future, could you detail the changes you made since the you made the frequency plots.

Have you done some additional testing that have not detailed regarding the desk? If it is needed, then yes cut it in half. I have doubts if such extreme actions are needed though. Monitor position and selective acoustic treatment should provide the same results.

Keep the great stuff coming!

Andre

Hi Andre,

The carpenter is coming around tomorrow. I am sure it seems drastic but I would think the less you have in the way the better. We will find out soon I shall post the details, waterfall and freq response once its done. I shall also post the original response so people can see how it has changed. I originally started in one room which was divided into a control room and VO booth. Since then I have changed the spare bedroom into the VO booth so I could have as bigger control room with symmetry. All will be posted by the mid next week. Cheers