The No.1 Website for Pro Audio
 Search This Thread  Search This Forum  Search Reviews  Search Gear Database  Search Gear for sale  Search Gearslutz Go Advanced
RoomEQ waterfall chart - takeaways Bass Traps
Old 2 weeks ago
  #1
Here for the gear
 

Thread Starter
RoomEQ waterfall chart - takeaways

Howdy!

I finally downloaded RoomEQ and did some measurements of my home studio, which is basically a ~16ft x ~10ft x ~8ft room in the basement with homemade absorption and corner bass traps. I want to determine if I should try to treat the room further, but I want to make sure I understand the actual measurements first. I ran the test using a calibrated Beyerdynamic MM1 mic in the listening position of my monitors.

When I look at the attached waterfall, here is what I "think" I am seeing:

1) There seems to be a dip in the frequency response between ~75Hz and ~110Hz

2) Aside from the large dip, the overall frequency response appears a bit "choppy"

3) There seems to be some resonance issues at ~70Hz and ~105Hz as well as at several higher frequencies

Some questions I would love input on are:

A) Given that the resonance issues appear to have levels in the 40-50dB range compared to the overall 70-75dB response, are they truly issues, or is this something that can be ignored?

B) How would you consider the overall frequency response?

C) What can cause such a dip in the 75-110Hz range, and is there anything that can be done about it?

Thanks in advance for any feedback!
Attached Thumbnails
RoomEQ waterfall chart - takeaways-waterfall-11302017.jpg  
Old 1 week ago
  #2
Lives for gear
 
JWL.GIK's Avatar
 

Old 1 week ago
  #3
Here for the gear
 

Thread Starter
Thanks for those links, some very interesting reading there.

I spent several hours moving my bass traps around and also trying different positions for the monitors and listening position. My bass traps are essentially two layers (~5-6in) of 2x4ft Safe n Sound insulation.

With all of the different scenarios I have tried, there is a dip/null around 75Hz-85Hz that does not seem to go away (it simply shifts up or down a few Hertz and even got worse when I brought the speaker up to the front wall). However, the dip/null at the slightly higher frequency (around 105Hz - 120Hz) appeared to improve when I removed my bass traps from the room, which surprised me, and I also got some improvement when moving my listening position closer to the front wall (without moving the monitors much closer than 2.5ft or so).

The ringing at ~70Hz has not been much affected so far, but I did have some success in taming the ~100-105Hz ringing when moving my bass traps to the rear wall. I need to do more work here, though. The measurements below were done without any rear wall treatment, but again, the dips that I am most concerned with were not affected much by my various attempts on the rear wall.

I have attached a few images that I would love some feedback on:

1) "4 Corner bass traps" file. This is the same chart as in the earlier post - I am including it again as I had adjusted the scale slightly. There are big dips at 86Hz and 106Hz but generally, from 72Hz all the way to 130Hz there is a bit of a "valley". The frequency response can be seen in the red line in the file named "Frequency chart" (I used 1/24 smoothing).

2) "No corner bass traps". As you can see, there is still a dip at 83Hz, but there is less of a dip over 100Hz compared to having the bass traps installed. Overall, though, there is still a "valley" from ~70Hz - 130Hz. The green line on the "Frequency chart" corresponds to this measurement.

3) "No corner bass traps listening closer to front wall". Still have a dip around 77Hz, but now there is much less of a dip at the 100Hz plus area, although there is a bit of a dip at 122Hz. The blue line on the "Frequency chart" corresponds to this measurement.

I have also attached a photo from a few days ago which shows the room with the bass traps in the front upper corners and this should give you some idea of the layout. It is not usually this messy and I usually have two panels mounted on the front right (the one covering the closet door is not shown in this pic).

Anyway, do you see anything in these charts that you could draw conclusions from or suggest next steps?
Attached Thumbnails
RoomEQ waterfall chart - takeaways-4-corner-bass-traps.jpg   RoomEQ waterfall chart - takeaways-no-corner-bass-traps.jpg   RoomEQ waterfall chart - takeaways-no-corner-bass-traps-listening-closer-front-wall.jpg   RoomEQ waterfall chart - takeaways-freq-chart-12.5.2017.jpg   RoomEQ waterfall chart - takeaways-room.jpg  

Old 1 week ago
  #4
Gear Guru
 
DanDan's Avatar
Boom

Do you have drums in the room while measuring?
DD
Old 1 week ago
  #5
Here for the gear
 

Thread Starter
I do indeed have drums in there when measuring but they will always be in there so I figured that made sense. It could perhaps be interesting to see what impact they may or may not have, though, so perhaps I should try removing them.
Old 1 week ago
  #6
Gear Guru
 
DanDan's Avatar
Boom

Ringing Toms could look exactly like modes. You could just throw heavy towels or such over them but that won't kill the bottom heads. If you want to identify modes, try the Sine Generator. Tick the box frequency follows cursor. Slide over your suspected modes. It will either light up the room into a modal pattern of peaks and nulls or drive one of those drums crazy.
DD
Old 1 week ago
  #7
Lives for gear
You have trapping from the waist up. I would extend your corner traps to floor. But perhaps we should hold off until you do some waterfalls with the drums out of the room.
Old 1 week ago
  #8
Here for the gear
 

Thread Starter
That signal generator is quite neat - it was very ear opening to set it to the lower null frequency and move around the room and realize how much the level changes in different spots. I used both my ears and an SPL meter. I did not notice any issues arising from the drums when playing around with the frequency of the signal generator, but it might still make sense to remove the drums completely just to be sure. I also need to spend some time on the resonating frequencies evident in the waterfall charts. I am hoping I can set the signal generator to ring at those frequencies and maybe using an SPL meter mounted at the listening spot, move some bass traps around the room to see if I can reduce the ringing at the listening spot.

One thing I noticed was that at the null frequency, in the right rear corner, there was a large dip, whereas there was no such dip in the left rear corner. I also noticed along the right side, there were null spots not noticeable on the opposite left side. The left side wall faces the outside of the house and thus there is drywall, insulation and a concrete wall (the room is underground) on that side. On the right hand side, however, there is drywall, no insulation and bordering rooms. In this type of situation, is the solution basically to move the speakers and listening spot to you find the least inconsistent spot, or is there a way to reduce the nulls by using absorption? Would the absorption need to be on the inside wall/corners where the nulls are, or on the outside wall where the nulls do not appear to exist, or is this not a straightforward answer?

I will try to spend more time on this tomorrow, but would love any further pointers you guys might have so I can make good use of my otherwise limited time!
Old 1 week ago
  #9
Gear Guru
 
DanDan's Avatar
Map

Focus on the largest and longest modes. Make a map of the hotspots and nulls and keep it. This will always advise you where to and where not to place sources or receivers. The frequency follows cursor tick box on the sig gen will play exactly the ridge you are looking at.

DD
Old 1 week ago
  #10
Here for the gear
 

Thread Starter
Well, I thank you both for suggesting the drums might be an issue! Even though I could not sense that they were resonating, the decay times look a lot better after I removed the drums from the room as is evident in all three waterfall charts below. Although I will need to live with the drums in the room, I might be able to address the resonating with some heavy blankets as suggested or by other means, but at least I know (or think I know) it is not a room issue.

I measured the room today with 4 panels on the sides at the reflection points and the drums removed from the room (see photo below). I tried moving my 4 bass traps around (all on the left side along the outer wall, all along the right hand side on the inside wall, corners, and combinations thereof without a whole lot of luck. What I did find, however, is a small improvement in the lower frequency null when I put floor to ceiling bass traps on the front wall behind the speakers, and that the null became worse when I put floor to ceiling bass traps in the front corners. The three waterfall charts below show:

1) Brownish/yellow chart - shows response with two regular 3in absorption panels behind the speakers on the front wall

2) Red chart - shows response with floor to ceiling bass traps in the front corners

3) Green chart - shows response with floor to ceiling bass traps ion the front wall behind speakers - see the picture below

As you can see in the frequency chart, the red line (corner bass traps) has the worst null and the green (floor to ceiling traps behind speakers) shows a few decibels improvement. Unfortunately, I am not seeing a way to eliminate this null so far.

I should point out that there is a cavity in the wall behind the right speaker. You might be able to see it in the photo - there is a shelf inside of it behind the left bass trap. It is about 30in wide and equally deep. I wonder if that might be an issue?

It is very evident that at the measured null frequency, there are peaks and nulls around the room, but I honestly don't know what else I can try at this point, especially since putting anything in the corners seems to make matters worse. Any further ideas?
Attached Thumbnails
RoomEQ waterfall chart - takeaways-4x-3in-panels-reflection-points-one-3in-panel-behind-each-speaker-12.6.2017.jpg   RoomEQ waterfall chart - takeaways-4x-3in-panels-reflection-points-floor-ceiling-6in-traps-front-corners.jpg   RoomEQ waterfall chart - takeaways-4x-3in-panels-reflection-points-floor-ceiling-6in-traps-behind-each-speaker.jpg   RoomEQ waterfall chart - takeaways-frequency-response-12.6.2017.jpg   RoomEQ waterfall chart - takeaways-room-setup-12.6.2017.jpg  

Old 1 week ago
  #11
Gear Guru
 
DanDan's Avatar
Good

Old 1 week ago
  #12
Here for the gear
 

Thread Starter
The interesting thing is that when I placed the speakers up against the wall the other day, I got worse dips than when they were farther out from the wall. However, I will try it again with the speakers right up against the floor to ceiling traps just to see as it might give different results with those large traps there.

Any idea why I could be getting worse dips when I have corner trapping than without? Everything I have read suggests that corners should be treated, but in my case that seems to make matters worse.
Old 1 week ago
  #13
Lives for gear
 
akebrake's Avatar
 

Idea?

Quote:
Originally Posted by sleipnir View Post
...Any idea why I could be getting worse dips when I have corner trapping than without? Everything I have read suggests that corners should be treated, but in my case that seems to make matters worse.
"Everything" in a small room interacts. Less SPL in one spot might increase pressure in another spot (and the other way around...)

Can you show us the rear of the room? Is it possible to turn "the room" 180°?
Eg. no cavity behind the speaker. No door to the right.

Best
Old 1 week ago
  #14
Here for the gear
 

Thread Starter
Unfortunately, the rear wall of the room has double doors as well as a door to a bathroom on one side, so I think I am somewhat limited in which general area I need to have my desk and speakers (I don't think I want to sit at 90 degrees i.e. up against the long wall with the speakers firing across the short dimension, although is this something worth trying?)

I have attached a photo below for reference of the rear wall with no treatment.

I ran a few more measurements today. The chart in blue represents a measurement with two bass traps floor to ceiling behind the right speaker with nothing on the rear wall.

The chart in red represents a measurement with three 3in panels floor to ceiling behind the right speaker with 4 bass traps covering the french doors and part of the wall left of the french doors.

The chart in purple represents a measurement with four bass traps floor to ceiling behind the right speaker covering more than the width of the cavity and three 3 inch panels covering the french doors. I have attached a couple of photos showing the panels in this setup.

From what I can tell, the last setup has a slightly better waterfall chart, but the lower notch is only marginally improved. I do think that covering the cavity is an overall improvement vs. the measurements done earlier this week, but it does not seem to be enough to reduce the notch sufficiently. I will probably build a freestanding floor to ceiling trap wide enough to cover the cavity.

With the cavity covered with 3in panels floor to ceiling and an SPL meter set up in the listening position, I also moved around the room and placed bass traps in various places and did not happen upon any placement that increased the levels at the listening position. I was only able to make matters worse. To some extent, just bringing the bass traps into the room seems to have negatively affected the levels at the LP by a few decibels.

Last night I also tried moving the speakers to the rear wall and placing panels next to and as flush as possible to the front of the speaker (to the extent I could). I did not have any success making the response any better. I have attached a pic - obviously this setup is not exactly the same as mounting the speaker in a wall, but it was an attempt to reduce low frequency reflections off the front wall (DanDan - thanks for the link above, it had some great info in it)

Anyway, I am starting to think this lower null is something I might just have to deal with, but I don't know if I want to accept that!
Attached Thumbnails
RoomEQ waterfall chart - takeaways-rear-wall-french-doors.jpg   RoomEQ waterfall chart - takeaways-2-bass-traps-floor-ceiling-behind-right-speaker-nothing-rear-wall.jpg   RoomEQ waterfall chart - takeaways-4-bass-traps-covering-french-doors-left-3-three-inch-panels-behind-right-speaker.jpg   RoomEQ waterfall chart - takeaways-4-horizontal-bass-traps-behind-right-speaker-3-panels-french-doors.jpg   RoomEQ waterfall chart - takeaways-3-inch-panels-covering-door.jpg  

RoomEQ waterfall chart - takeaways-bass-traps-covering-entire-opening-wall-cavity.jpg   RoomEQ waterfall chart - takeaways-attempt-place-panels-same-plane-front-speaker.jpg  
Old 1 week ago
  #15
Gear Guru
 
DanDan's Avatar
Closer

I'm a bit lost. Did you try your speakers close to the Front Wall, almost touching, no trap between speaker and wall?
DD
Old 1 week ago
  #16
Here for the gear
 

Thread Starter
I have indeed tried that but got worse results. I should point out though, that when I say up against the front wall, the right speaker really is not up against anything due to the cavity on that side, so only the left speaker really is truly against the wall (albeit is at an angle pointing toward the listening position). I also tried with the speakers against the front wall and surrounding the speakers with panels somewhat in the same plane as the front of the speaker with not much luck (that was the photos included in the prior post). That was an attempt to mimic placing the speaker "in" a wall.

I played around with the low frequency +/- 6dB boost/cut on the monitors (Focal Solo) hoping that it could help even out the low end, but to no avail. When boosted, the levels of all low frequencies increased except at the null which remained pretty much at the same level. When I cut 6dB, the low frequency levels decreased including at the null.
Old 1 week ago
  #17
Gear Guru
 
DanDan's Avatar
Cavity

Aaah, a cavity. Can you fill that with fibre?
Also I see a lot of big traps around the speakers. Better in the corners I would guess.
DD
Old 1 week ago
  #18
Here for the gear
 

Thread Starter
Filling the cavity would be an option, but I kind of need the space as storage (I have a shelf within the cavity). I am thinking based on the measurements so far, that having a freestanding floor to ceiling panel in front of the shelves could be beneficial. However, I might try removing the shelf, put as many traps inside the cavity that I can fit, place the speakers against the front wall and measure just to get a sense for whether filling it might be superior to the trap covering the front.
Old 1 week ago
  #19
Lives for gear
 
akebrake's Avatar
 

L/R Differences

Quote:
Originally Posted by sleipnir View Post
... I am thinking based on the measurements so far, that having a freestanding floor to ceiling panel in front of the shelves could be beneficial....
Left/ Right Symmetry in a room is important. I guess the cavity/ shelves and R (gypsum) wall will influence the Right speaker a lot more than the Left.

Always measure L and R separate in order to find the cause of the problems.

Best

EDIT: PS If you upload your mdats we can look a little deeper

Last edited by akebrake; 6 days ago at 01:02 PM.. Reason: added offer
Old 6 days ago
  #20
Here for the gear
 

Thread Starter
I had not thought of testing the speakers individually. I tried it, and indeed the response is not symmetrical between the two! I guess I will need to try and get the response as close as possible between the two which adds another challenge.

Per your suggestion, I have attached an mdat file which contains the raw data for one of the better (or less bad) looking charts where I tested the room with both speakers. This is the setup where I had 4 bass traps lined up horizontally across the cavity from floor to ceiling and three 3in traps covering the French doors on the rear wall. I have reattached the waterfall to this post for reference. I can provide mdats for any other measurements if you feel it would be insightful.

In this mdat file are also two sets of data for the left and right speakers when I covered the cavity with wood. I have also attached the waterfalls as jpg files here. As you can tell, the response differs from left to right.

I am thinking about building a door made out of 3/4in plywood to cover the cavity. On the back, I could frame in floor to ceiling 3in Safe n Sound insulation that I have on hand. This would allow me to continue to utilize cavity for the shelving I have and also provide a reflective surface similar (albeit it not identical) to the drywall covering the rest of the front wall. Further, since there is insulation behind the drywall on the front wall, having insulation on the back of the door would perhaps be more consistent than just placing traps in front of the wall like I have done in the previous experiments. I am hoping this door could help even out the front wall relative to the left and right speakers.

Anyway, I would then also plan to either build or buy bass traps for the corners. At that point, I might be able to get slightly improved measurements from which to make further tweaks.

I could potentially open up the right wall and insulate it, but I would need to somehow convince myself that it would lead to an overall improvement. There is no way of making the left and right walls identical since the left wall has concrete behind it and the right wall two doors and adjacent rooms, but maybe there is room for improvement.

I created a scale drawing of the room. It might be easier to get a sense for the room from this versus the photos which never capture the entire room at once and thus make it hard to get a sense for relative dimensions.

Anyway, thanks for taking the time to read my post! If you have any further insights, please do share!
Attached Thumbnails
RoomEQ waterfall chart - takeaways-4-horizontal-bass-traps-behind-right-speaker-3-panels-french-doors2.jpg   RoomEQ waterfall chart - takeaways-left-speaker-wood-covering-cavity.jpg   RoomEQ waterfall chart - takeaways-right-speaker-wood-covering-cavity.jpg  
Attached Files
File Type: pdf Studio layout.pdf (98.7 KB, 11 views)
Old 6 days ago
  #21
Here for the gear
 

Thread Starter
Looks like my mdat file might have been too big (I did not get an error when I uploaded it, but it did not show up in the post). I broke the file into three, one for each of the three waterfalls attached in the post immediately prior to this one.
Old 5 days ago
  #22
Gear Guru
 
DanDan's Avatar
Symmetry

While I agree with akebrake, always if I remember correctly, LF symmetry is not really on the cards here.
If the current orientation is set, then I would keep an open mind as to where anything can be. When confronted with an asymmetrical room I often end up with the speakers off- set to mitigate the effects. Given that we have two ears at different locations and LF response can differ dramatically over 8 inches, I would go for the best summed response at both ears. Particularly trying to avoid nulls.
So I commend an utterly open mind in this very asymmetrical space.
e.g. One could fit one speaker 'flush' in the cupboard, the other touching the wall. Odd but why not?
Or go off centre, trying to make the LF response of each speaker individually least bad, AND the sum least bad. They interact.

Looking at the nicely done diagram, would it be possible to place the speakers at the Long West Wall?
This would place the speakers at an LF boosting concrete wall, a long way from both side walls, with the back wall being light drywall with voids behind.

DD

Last edited by DanDan; 5 days ago at 04:33 PM..
Old 4 days ago
  #23
Lives for gear
 
akebrake's Avatar
 

Heavy walls, light walls

Quote:
Originally Posted by sleipnir View Post
... I broke the file into three, one for each of the three waterfalls attached in the post immediately prior to this one.
Thanks for the mdats.

All measurements so far have big dips. Left spkr close to front wall shows the modes more easily. I have included them (coloured bars) along the Frequency scale. (Just insert Room dimensions in REW EQ tab). see the last fig

I’ve found REW Room Sim interesting for comparing different spkr and listener positions. E.g. You can simulate how much the FR will change, a lot or just a little when moving things.

It's very handy to move the speaker and listener around in the computer...


The challenge is to estimate the average (20-200 Hz) LF absorption of the boundaries and DIY traps...


In stead of ”a shot in the dark” why not try a simulated winner?

I simulated 4 different scenarios:
1. Diagonal (spkr in heavy corner) order to investigate the modes of the room.
2. Spkr at heavy front wall
3. Spkrs equally spaced from the heavy front left corner
4. Spkrs at Long Wall as suggested by DD.

Which one is best?
Attached Thumbnails
RoomEQ waterfall chart - takeaways-sim-diagonal-jpg.jpg   RoomEQ waterfall chart - takeaways-sim-short-jpg.jpg   RoomEQ waterfall chart - takeaways-sim-corner-jpg.jpg   RoomEQ waterfall chart - takeaways-sim-long-jpg.jpg   RoomEQ waterfall chart - takeaways-coinc-modes-.jpg  

Old 3 days ago
  #24
Here for the gear
 

Thread Starter
DanDan & Akebrake,

Thanks for getting back to me and coming up with suggestions, I really appreciate your taking the time!

I am away from home for a few days, but I will do some more testing later this week when I get back. I did not know about the room sim feature, that is pretty neat. Akebrake, as far as the surface absorption coefficients you used, are those typical for the materials in question?

I am curious to test both the diagonal (across the corner) speakers and centered along the long (heavy) wall and see what I measure. I think it is worth trying the the speakers offset towards the left of the front wall as well.

I will let you know once I have taken more measurements!
Old 12 hours ago
  #25
Lives for gear
 
akebrake's Avatar
 

Quote:
Originally Posted by sleipnir View Post
DanDan & Akebrake,

Thanks for getting back to me and coming up with suggestions, I really appreciate your taking the time!
My pleasure!

Quote:
Akebrake, as far as the surface absorption coefficients you used, are those typical for the materials in question?
No, that’s just my ”guesstimate” figures (realistic?) inserted for this calculation.

Wall plus some kind of trapping may give a boundary an (average) alpha≈ 0.5
Maybe a bit too high... Comments are welcome!
Real measurements will tell.

IMHO Room Sim is very educational in order to get a rough idea of how FR relates to:
Different room modes (axial, tangential, oblique) clustered or isolated, first reflections, pos of spkr and listener, room dimensions, average absorption etc.

BTW, as with ALL kind of simulations...
A number of parameters have to be known in advance for a detailed simulation.
The accuracy of a (perfect) simulation software can’t be better than the data we put into it.

Best
Topic:
Post Reply

Welcome to the Gearslutz Pro Audio Community!

Registration benefits include:
  • The ability to reply to and create new discussions
  • Access to members-only giveaways & competitions
  • Interact with VIP industry experts in our guest Q&As
  • Access to members-only sub forum discussions
  • Access to members-only Chat Room
  • Get INSTANT ACCESS to the world's best private pro audio Classifieds for only USD $20/year
  • Promote your eBay auctions and Reverb.com listings for free
  • Remove this message!
You need an account to post a reply. Create a username and password below and an account will be created and your post entered.


 
 
Slide to join now Processing…
Thread Tools
Search this Thread
Search this Thread:

Advanced Search
Forum Jump
Forum Jump