Measuring, analysing and treating a synth room

[diedushka]

Hey there,

So since the build process of my room is well underway now I have stumbled into some problems. Also, since Im a working professional engineer - I dont really have much time to spend on designing and Sketchup, so I cant actively show a model of where Im at right now.

Basically my room is 5.8x3.2x2.1 in meters. It is a room within a room design with pine framework that has been done with MDF and gypsum boards and very heavily caulked. The floor is granite tiles.

I have built the soffits according to Johns design and made the first measurements. These measurements have revealed something that im quite puzzled about. My soundcard has a very large internal frequency response deviation of around 11 dB. It is an RME Fireface UCX and im using a Presonus PRM1 mic with JBL P708 speakers. You can see the measurement here:

Soundcard_700px.jpg

Im not sure why it is this way. I have an idea there might be some kind of earthing ground loop messing with the interface or something in the power grid.. I watched several videos and read tutorials and posts on REW and most sound interfaces only deviate in the lowest and highest frequencies within several dB. So my question is maybe some of the guys here can give me any ideas of why might my soundcards response is the way it is? Theres an official video using an identical RME soundcard that has almost a flat response all through the frequency range so I am a bit puzzled as to why mine isnt at least similar. You can see the one im talking about in this video: HERE

I measured the empty room with added hangers right under the speaker enclosure. For anyone willing to check them out you can do that HERE.

Also, i have a question: the JBL P708 speakers have quite a bit low end in a seemingly small cabinet. I have another set of speakers that have a bit less presence in the low end and im thinking what should I do here. I guess mounting speakers in a soffit that have a bit more low end presence would be more beneficial since it would help with reflections, correct? And since im going to mangle synths and basically do electronic music I think having more low end is a good thing. Would any of you reccomend the softer speakers with a sub or stronger speakers without one?

Also I have thought about making a slot resonator to attack the problems in the low end of the spectrom that my room showsand came up with this fast sketch of a box. It is supposed to target frequencies between 29 and 92 Hz. Ive calculated that using an excel file that was on a thread here in the forum.

You can download the sketch HERE.

Can someone just tell me if im in the right direction or not? The design goes as follows:
2 slots of 250x50 mm that are 1 mm away from one another;
2 slots of 230x50 mm that are 2 mm away from one another;
2 slots of 200x40 mm that are 3 mm away from one another;
2 slots of 160x40 mm that are 3 mm away from one another;
2 slots of 160x30 mm that are 3 mm away from one another;
1 slot of 160x20 mm that are 3 mm away from the wall.

The depth is 240 mm for all the slots.

Thank you for your help in advance.

[Soundman2020]

My soundcard has a very large internal frequency response deviation of around 11 dB. It is an RME Fireface UCX and im using a Presonus PRM1 mic with JBL P708 speakers. You can see the measurement here:

It's feedback. You didn't do the calibration process correctly, and somehow the output is getting back to the input but NOT through the loopback cable. This problem is documented in the REW calibration procedure itself. You can find it in using "Help":

REW-calibration-feedback-error.jpg

I dont really have much time to spend on designing and Sketchup, so I cant actively show a model of where Im at right now.

Please don't take this the wrong way, but that's not a good approach to building a studio. Design is 80% of the process, and actual construction is 20%. If you spend the time to design it right, then construction will bo much faster, with far fewer complications. It will be cheaper too.

I have an idea there might be some kind of earthing ground loop messing with the interface or something in the power grid..

It's not a grounding issue: It's a feedback issue. There's an unwanted signal path that is returning the output to the input directly.

So my question is maybe some of the guys here can give me any ideas of why might my soundcards response is the way it is?

That is not the real response of your soundcard: it is what comb filtering of feedback looks like.

i have a question: the JBL P708 speakers have quite a bit low end in a seemingly small cabinet. I have another set of speakers that have a bit less presence in the low end and im thinking what should I do here

Which set do you plan to use as your main speakers? The P708's, or the "other" ones?

I guess mounting speakers in a soffit that have a bit more low end presence would be more beneficial since it would help with reflections, correct?

Not really, no. The soffits will help to extend the low end a bit, and will also eliminate a number of other problems caused by having a speaker in the room. Putting them in soffits technically means that they are NOT in the room any more, so therefore the problems disappear. But that has nothing to do with reflections! Putting your speakers in soffits that are properly designed as part of an RFZ style room will ALSO help be re-directing first order reflections around the mix position. But that has nothing to do with low-end response, or artifacts. These are entirely unrelated issues: soffits help with all of them, but there's no relationship between them.

And since im going to mangle synths and basically do electronic music I think having more low end is a good thing.

If you mean that having speakers that cover more of the spectrum down to lower frequencies, then yes, that is desirable. But if you mean speakers that have enhanced low end response (a "bump" in the frequency response), then no, that's not a good thing. The frequency response in a control room should be flat, or close to flat, across the entire rang of frequencies that you need to work with, and that your speakers can produce.

Would any of you reccomend the softer speakers with a sub or stronger speakers without one?

What frequency range do you need to cover? Which of the two sets of speakers has the highest quality and best specifications for YOUR room? If the highest quality ones are suitable for your room, and also cover the frequency range you need to cover, then use them as they are. If not, then add at least one sub, and choose that sub carefully so that it does extent the range down low enough, and also fits in with the main speakers.

Also I have thought about making a slot resonator to attack the problems in the low end of the spectrom that my room showsand came up with this fast sketch of a box.

To be honest, I would not take that approach to treating your room. Helmholtz resonators are VERY difficult to tune correctly, especially at very low frequencies. You also seem to be attempting to treat all of your modal problems with one tuned device, but that is impossible. Modal issues can ONLY be treated at the points where the pressure component is at its maximum, or close to maximum, and considering that your modes occur in three different directions, there is no single place in the room where you could treat all of them. For example, ou can only treat the 0,x,0 axial modes on surfaces that are involved in those modes, so putting your device on the rear wall would do nothing at all for those modes. And if you put it on one of the side walls, then it would do nothing for the x,0,0 and 0,0,x axial modes.

Also, the internal volume of the resonator needs to be about 1% of the entire room volume. Your box has a volume of 0.1 m3, but your room has a volume of nearly 40m3. You would have to make that box at least four times as large to get a useful effect.

What I would suggest is to first build general porous absorption traps, such as superchunks, corner traps, hangers, or some such, in the usual obvious locations, as well as treatment on your first reflection points, a cloud, etc. since you WILL need those anyway, then do some further REW testing to see if there are any remaining issues that could possibly need Helmholtz resonators.

That would be my approach to treating your room.

However, with no design to analyze, and not even a clear design philosophy, it isn't possible to say any more about what you might or might not need. I would need to see the complete, finished design, with all aspects in place, in order to be able to say anything additional.

- Stuart -

[diedushka]

It's feedback.

Thank you, Stuart. This helps a lot.

Design is 80% of the process, and actual construction is 20%.

Yes I know, its all here in the forum.. But im so eager to just go in and just start mangling them knobs..

Which set do you plan to use as your main speakers? The P708's, or the "other" ones?

Theats the whole deal - Im not sure which ones to choose as my mains. The P708s have a stronger lower end, but to this day (they have been released for almost a year now) there is not frequency response data or official measurements to actually know if the low end isnt boosted or not. The website says they are using some new technology that legit reproduces the lowest frequencies better than any other monitors in the market. A bald statement indeed: "These proprietary drivers allow the 708P to deliver two to three times the output of existing studio monitors." So im quite puzzled at this point.

The other set of speakers is a custom made pair that i have. These were measured and tested in an anechoic chamber and come with their own frequency response measurements and technical specifications. I can swear by their flatness and nice stereo image. But they are not as prominent in the low end - around 12 db less. It is also pretty audible when you compare them side to side. The customs also seem to have a more crisper highs. So I kind of wanted to mount the customs, but then the JBLs came along and seemed to play much louder and have more low end and that kind of messed with my head a bit.
Also, the JBLs state that their monitors go down to 35 Hz at -10 dB and the customs go to around 38 Hz at -6dB.

If you mean that having speakers that cover more of the spectrum down to lower frequencies, then yes, that is desirable.

That is exactly what I meant.

The frequency response in a control room should be flat, or close to flat, across the entire rang of frequencies that you need to work with, and that your speakers can produce.

Yes, but the thing is that this wont be a control room, per se. More of like a producing room. There will be 4 keyboard stands in the room that will house my collection of synths and some other equipment so this will already mess up the acoustics to be close to a control room standard because of reflections. All im aiming here is to have a decent response within a 10 dB range. If I can get into 6 dB that would be awesome. Less that that I think is not possible in this room, because I would not be able to sacrifice the space..

What I would suggest is to first build general porous absorption traps, such as superchunks, corner traps, hangers, or some such, in the usual obvious locations, as well as treatment on your first reflection points, a cloud, etc. since you WILL need those anyway, then do some further REW testing to see if there are any remaining issues that could possibly need Helmholtz resonators.

I think you are right.. This might be just be getting way ahead of myself. Indeed, I will first sort my soundcard out and check the measurements i made once more. After that I will do the absorbent stuff (703 or 705 similar products between the studs in my studio) and some superchunks.
I do have a cloud in mind just as you have mentioned, but I also understand its a bit too early to talk about it.

What would you need to be able to better provide some advice? I am way into the building process and doing a detailed design of whats already inside would be time consuming and tedious.

[diedushka]

Oh, also one thing:

Im ordering the insualtion to fill the spaces between the studs and I was thinking - should I mount one specific type of insulation or should I use different types and play around with them to see what happens? Also, I have read that some users have been able to achieve nice goals with FRK insulation. Maybe I should use different density and coated/uncoated and measure to see what combination has the best result?

[Soundman2020]

Yes I know, its all here in the forum.. But im so eager to just go in and just start mangling them knobs..

Ahh, yes... sadly, you have been infected with the "home studio builder" disease... It affects your mind, and forces you to want to just "get the darned place finished", so you can finally track and mix in peace. There is no known cure for this disease. Once you have it, there's nothing that can be done...

Seriously, slow down a bit! It's really not a good idea to rush into building without having a solid plan in place for the entire studio.

For example, what are your plans for HVAC? What air flow rate do you need? What air flow velocity? What is the sensible heat load that your HVAC system must deal with? And the latent heat load? And what is the static pressure? What are the cross sectional areas of the registers, the interior of the silencers, and the ducts?

If you have not yet calculated all of that, then you should not be building yet!

The P708s have a stronger lower end, but to this day (they have been released for almost a year now) there is not frequency response data ...

I found this for the 708i, which is basically the same speaker but passive, not active:

JBL-LSR708i Spin.gif

The 708p should have pretty much the same response, I would expect. I'm a bit curious about the sudden dips at around 600 Hz and 11k. Not sure what those are about: unlikely to be crossover issues. Also curious about the peak at 65 Hz, although I suspect that's intentional, to give them a little extra low-end "punch". Or it might just be something related to the reflex port...

The website says they are using some new technology that legit reproduces the lowest frequencies better than any other monitors in the market. A bald statement indeed: "These proprietary drivers allow the 708P to deliver two to three times the output of existing studio monitors."

Bold indeed! And very unusual, coming from someone like JBL. It's basically meaningless, since they didn't say "two to three times" of WHAT! Total power output? SPL? Power at a certain frequency? The claim has no meaning...

Also, the JBLs state that their monitors go down to 35 Hz at -10 dB, yeah, but that's not a useful reference. The manual also says that the -3 dB point is 41 Hz. Still quite respectable, for an 8" driver.

But they are not as prominent in the low end - around 12 db less.

-12 dB at what frequency?

but then the JBLs came along and seemed to play much louder

"Louder" is not necessarily "better". As long as your speakers can perform perfectly at 85 dBC SPL, and can also hit 100 dB when you need it, then you should be good. 85 dB is the reference calibration level for cinemas and studios, and is also the level that many engineers mix at, or try to keep below.

There will be 4 keyboard stands in the room that will house my collection of synths and some other equipment so this will already mess up the acoustics to be close to a control room standard because of reflections.

Not necessarily! With careful design, the issues can be minimized. You can still get good control-room acoustics if you need it.

Also... and this might seem a little harsh, but if this is not a control room and does not need excellent acoustics, then why would you blow US$ 4,000 on a pair of speakers that you don't need? After all, if the room is not going to be much good, then why put in excellent speakers that won't be able to perform to their maximum?

If you already have them, then it makes sense to get the room designed so that it will complement the speakers as well as possible, maximizing performance...

All im aiming here is to have a decent response within a 10 dB range. If I can get into 6 dB that would be awesome.

+/-10 dB is typical for home studios with moderate treatment. +/- 6 dB is achievable with careful design and careful construction. A room I'm working in in another thread is already at +/- 10, and has only done about a quarter of the room treatment ...

After that I will do the absorbent stuff (703 or 705 similar products between the studs in my studio) and some superchunks.

705 is too dense: it doesn't go down low enough in frequency. Use 703, or even 701. Did you build your walls inside-out? Do be aware that doing the entire room with 703 around all the walls will make it very dead and unpleasant, while still leaving large issues in frequency response. That's the starting point for the "kill it then revive it" tuning plan. As long as you are aware of that, then that's fine.

I am way into the building process and doing a detailed design of whats already inside would be time consuming and tedious.

... but necessary, nevertheless....

What would you need to be able to better provide some advice?

PM me, and we can talk about it. Or PM John himself: it's his forum, not mine! You should always contact John first, if you need professional advice.


- Stuart -

[diedushka]

For example, what are your plans for HVAC? What air flow rate do you need? What air flow velocity? What is the sensible heat load that your HVAC system must deal with? And the latent heat load? And what is the static pressure? What are the cross sectional areas of the registers, the interior of the silencers, and the ducts?

Im afraid all of that has already been done. I had a professional HVAC guy look into the room. He did the calculations, bought everything and put it in. I, of course, have stressed him very much about the isolation from the outside so he did silencer boxes and lined them with some special duct lining that goes into air ducts. Some sort or 703 typo product that has a black cover on one side. Also very quiet IN and OUT fans were put inside. The air is flowing very smoothly and you cant hear a single sound from the outside.

-12 dB at what frequency?

At around 30 Hz. There is a picture with both of those speakers measured in my first post here. One is way above the other down at 30 Hz and then they even out at around 90 hz.

You can still get good control-room acoustics if you need it.

Yes, that is preffered ))

A room I'm working in in another thread is already at +/- 10, and has only done about a quarter of the room treatment ...

Yes, I have seen. What do you think could be achieved in that small room? I would also aim for as good sound as possible, but I know I cant get VERY good because of the size of the room But I think well work something out to at least a good level

That's the starting point for the "kill it then revive it" tuning plan. As long as you are aware of that, then that's fine.

Yes, that was my idea for the room

You were also correct about the feedbak loop. I went in and did some more adjustments on my soundcard and got a decent response:

Soundcard_700px.jpg

I also measured both speakers at once and separately:

NEW-Measurement.jpg

[Soundman2020]

I went in and did some more adjustments on my soundcard and got a decent response:

Why are there two curves in there? There should only be one...


- Stuart -

[diedushka]

I went in and did some more adjustments on my soundcard and got a decent response:

Why are there two curves in there? There should only be one...


- Stuart -

I think I left the default where theres frequency response and phase.

[diedushka]

705 is too dense: it doesn't go down low enough in frequency. Use 703, or even 701. Did you build your walls inside-out? Do be aware that doing the entire room with 703 around all the walls will make it very dead and unpleasant, while still leaving large issues in frequency response. That's the starting point for the "kill it then revive it" tuning plan. As long as you are aware of that, then that's fine.

I have read several posts, that 4" of 705 FRK is actually better than 703 or 701. I also downloaded the technical data for OC products and their measurement results for FRK show a much bigger absorption for 2 inch thick insulation panels. Check it here:

OC700-Absorption.JPG

So why should I use 701 exactly, when it says 705 has better absorption in the low end. Maybe im not understanding something..

[Soundman2020]

I have read several posts, that 4" of 705 FRK is actually better than 703 or 701

You are not comparing apples to apples! And you are not using the right product for the right application.

Since you say your room is already built, and now you want to treat it, I'm assuming that you are talking about what type of insulation you should use for your TREATMENT, particularly bass trapping, not what type you should use inside the walls for ISOLATION. Those are two very different things.

You mentioned FRK: Do you realize what that means? You do realize that "FRK" stands for "Foil Reinforced Kraft", and refers to OC-70x that has been faced on one side with kraft paper that is covered with a thin layer of metal foil? That is NOT comparable to the OC-70x product without the facing. Different products, different acoustic properties, different acoustic uses.

As far as I know, 701 is NOT available in the FRK version, so you cannot compare it with 703FRK or 705FRK.

Now look at your chart again, comparing only apples with apples:

At 125 HZ, 701 plain has a coefficient of 0.22, 703 plain has a coefficient of 0.17, and 705 plain has a coefficient of 0.16. Therefore, 701 is the best product for low frequency absorption, followed by 703, and 705 is the worst. Just like I said.

That's for the plain version (unfaced). Now let's do the same exercise for the faced version:

There is no 701FRK, so we cannot compare that. But we can compare 703FRK and 705 FRK:
At 125 HZ, 703FRK has a coefficient of 0.63, and 705FRK has a coefficient of 0.60. Therefore, once again, 703FRK is the best product for low frequency absorption, and 705 is the worst. Just like I said.

The general rule os that for any given class of insulation, lower density will perform better for absorbing lower frequencies, and higher density will perform better for absorbing higher frequencies where low frequency absorption is NOT desired.

Having said all that, the chart you show is not valid for your application! It shows only the results for Type-A mounting, which means the product was tested up against a hard, solid surface. That's not what you need for low frequency bass trapping. Rather, you should be looking at the results for ASTM E-405 mounting, which more closely reflects a bass trap application. Here's the full chart for all OC-70x products, in both mounting types:

OC-703-specs.jpg

As you can see, the best overall bass trapping performance is 4" OC-701 with ASTM E-405 mounting. That has a coefficient of 0.87 at 125 Hz: Second best is OC-703 with 0.75 and last is OC-705 with 0.59. Once again, the lighter, less dense product is greatly superior for low frequencies.

If somebody wanted to go with only 2" (although I'm not sure why anyone would use such a thin panel for low frequencies, but let's just assume....) then here's the situation:

At 125 Hz, 2" of 701 mounted E-405 has a coefficient of 0.44, 2" of 703 mounted E-405 has a coefficient of 0.40, and 2" of 705 mounted E-405 has a coefficient of 0.39. One more time, 701 wins out as the best product (lowest density), 703 is second best, and 705 comes in last. It is too dense.

So, the case is clear: if you want the best low frequency absorption using an OC-70x product, then 4" of OC-701 mounted ASTM E-405 is the clear winner. It beats everything else, even 2" of 703FRK or 2" of 705FRK, regardless of how it is mounted.

So why should I use 701 exactly, when it says 705 has better absorption in the low end.

Because that is not what the chart shows! Take another look, and make sure you are comparing the SAME product, not different products.

Maybe im not understanding something..

You are misreading the data. You cannot compare the characteristics of two DIFFERENT products (once faced, one not faced) and arrive at the conclusion that the difference is due to the higher density! The full table clearly shows that the difference is NOT due to the higher density, and that in fact it is better to use a LOW density product for bass trapping. That's what I said originally: The LOWER density product is better for bass trapping. That's true for the FRK faced class of products, and also for the unfaced class of products. In each class, the lower density provides better performance.

You should always make sure you are using the full data set when comparing products, not just part of it, and always make sure that you are comparing "apples to apples": in other words, comparing products that are basically the same, and mounted in the same way. 703 FRK is nothing at all like 703 plain, and 703 with A-mounting (on the surface) is nothing at all like 703 with ASTM E-405 mounting (16" air gap behind).

In reality, this is not really related to the actual density at all: rather it is related to the Gas Flow Resistivity of each product. GFR is a measure of acoustic impedance. Lower GFR in general is better for low frequency absorption, and higher GFR is better for high frequency absorption when the lows should not be touched. Fortunately, there is a rough relationship between GFR and density for each individual class of products, but NOT between classes. Each class stands on its own

- Stuart -

[diedushka]

Damn,

You schooled me there quite properly, Stuart.. Thank you for that You are indeed right about me misreading data. I have read some post of Ethan Winers I think that provided emasurements of different OC boards in the room and given that table I downloaded from OC website I assumed that the film somehow works in favor to density..

Also, what may have gotten me was that usually denser stuff in resonant devices corresponds to lower frequency. I guess that got me too Anyway, thank you for your help here man! And thanks for sharing your graph.

You are very right: I am using this insulation for absorption, not for isolation. The room is in a pretty good acoustical place so I dont need much if any treatment to proof my room from someone elses noise or vice versa.

Thank you!