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
Measurements Comparison Dynamics Plugins
Old 19th December 2010
  #31
SAC
Registered User
 

That's fine!
You're getting it!
As amish mentioned, typically you will set your time scale to include a little more than twice the room length to allow for the first or second order round trip reflections to be included.

What becomes critical is that, after all is said and done, that the measurement is done correctly.

Let me express one concern. In the various plots we have alternately seen a gain scale where the 20 dB down region relative to the direct signal level has either nearly excluded everything - indicating a VERY dead room, and plots that include almost everything!

Obviously, in the final analysis, this will have to be ironed out, but we should not be seeing this issue in a re-windowing of the same measurement....so.... (I don't know what is going on in the measurement process - and while our concern over gain is relative and not absolute, I have no idea how the dBfs scale corresponds to dB SPL - so for simplicity sake, i am assuming it to be the same and simply ignoring the scale units - ahhhhhhhh!!!! Note, this is something that should be remedied if a platform is to be used for acoustic use as opposed to simply electronic (digital) gain staging.)

In this last post with the gain where it is, only the nearfield speaker and speaker mount regions, and perhaps the desk, are really significant issues. You have one or two small remnants that are not down 20 dB and then only by 1-3 dB. (But remember our concern over the real levels)

I suspect the room is not really that dead.

But going back with a similar view with an increased time scale set to , say, 35 ms that will accommodate roughly a 12-15 foot deep room, and assuming that the room is not that dead, you should see the early arriving spikes that are less than 20 dB down in level from the direct signal. these are the specific reflections that should be addressed.
Old 19th December 2010
  #32
Lives for gear
Quote:
Originally Posted by amishsixstringe View Post
I was asking you the depth of your room to decide how we can fix the fact that you don't have any late arriving reflections at all. I'm sure SAC would like to see that changed, as would I.

Neil
My room dimensions are 16.74' (5.10 meters) for the longest wall 10.5' (3.20 meters) the short wall and 8.86' (2.70 meters)

I don't understand what are you talking about late arriving reflections..
Old 19th December 2010
  #33
Lives for gear
 
amishsixstringe's Avatar
 

In the popular model known as "RFZ", there is a very specific order of events in time.

1. Initial impulse happens.

2. No early reflections can come back to the listening position above 20dB (SPL). Early is defined as 20ms.

3. At 20 ms (really 17-25ms) there should be a very noticeable return reflection at around -12dB below the initial impulse. This is a latter arriving reflection. It usually comes from the back wall. Often this is a diffuse sound field. The diffuse field is a much more dense 'reverb-like' tail that happens. Think of the difference between dropping 4 marbles on a metal plate at the same time verus dropping 200 marbles on a metal plate at the same time.

The later arriving signal aides with spacial imaging and is much more pleasing to listen to than a completely dead room. The 20 ms gap (ISD) is free of reflections allowing for the listener to hear the signal clearly before the latter arriving reflections occur.

There's a way better way to explain all this, I'm sure, but I suck at words.

Neil
Old 19th December 2010
  #34
Lives for gear
Quote:
Originally Posted by SAC View Post
That's fine!
You're getting it!
As amish mentioned, typically you will set your time scale to include a little more than twice the room length to allow for the first or second order round trip reflections to be included.

What becomes critical is that, after all is said and done, that the measurement is done correctly.

Let me express one concern. In the various plots we have alternately seen a gain scale where the 20 dB down region relative to the direct signal level has either nearly excluded everything - indicating a VERY dead room, and plots that include almost everything!

Obviously, in the final analysis, this will have to be ironed out, but we should not be seeing this issue in a re-windowing of the same measurement....so.... (I don't know what is going on in the measurement process)

In this last post, only the nearfield speaker and speaker mount are really significant issues. You have one or two small remnants that are not down 20 dB and then only by 1-3 dB.

Ok SAC, sorry but I'm a bit tired, I mean I really spent 3 weeks 8 hours a day doing measurements, traps, moving them, moving libraries, books inside of them, etc.. so I'm not very reactive

That's the first time I'm using ETC so I've posted many measurements, but are all the same two measurement, I just changed the scale as amishsixstringe suggest me

So, in the end, which are the biggest problems?

I've calibrated the system while measuring
I would not say it's accurate because I've calibrated room eq at 84 db with a spl meter and then, as you can see, room eq goes higher than 84 db, I also read Ethan saying that's normal in one of his article

So, sorry because I think is mostly because now I'm not reactive (kind of tired) and also my english is not that great, but could you please reassume what should I do now?

Sorry... :-/

What are your doubts about my measurement process, ask me I'll tell you how I've done them
Old 19th December 2010
  #35
SAC
Registered User
 

Don't apologize! I think its great you're wrestling with the monster!

You'll be an expert in no time!

The most difficult part of the process is getting a handle on the measurement process and platform interface itself.

Without getting into this topic now, don't worry about the 20 ms ISD time. to over simplify, that figure comes from an analysis of concert halls - large acoustical spaces.

That is normally not achievable in a small acoustical space, and ISDs of ~10 ms are more common.
This value will likewise be adjusted dependent upon the space and especially the live space in which the material is actually being recorded, as you want the CR ISD to be longer than the live room ISD in order that you hear the entire recorded space before the CR acoustics impinge on the source you are mixing.

My doubts? I don't know exactly what the process is - that is my doubt. How many speakers are being drive during the measurements, type of mic (omni, etc) - all of the process issues pertaining to making proper impulse response measurements can be included here. Note, i am not questioning you, but something is happening to cause the gain flip flop mentioned earlier that should be resolved....

But relax and take your time! there is no rush, and you are well on your way to getting this down!

And as you get everything ironed out, we will still be around to help you with the actual reflection identification process. Each step in its time...
Old 19th December 2010
  #36
Lives for gear
 
amishsixstringe's Avatar
 

I think what SAC was trying to say is that even though you're presenting the same measurement and only changing the window sizing, it sometimes appears to give different values. Only slightly, but noticeably. It's a software issue, not your problem. I think it will be fine for what we're trying to do.

So, what do you do next? Try to find that little reflection that goes juustt above 20dB, would be my move.

Neil
Old 19th December 2010
  #37
Lives for gear
Quote:
Originally Posted by amishsixstringe View Post
In the popular model known as "RFZ", there is a very specific order of events in time.

1. Initial impulse happens.

2. No early reflections can come back to the listening position above 20dB (SPL). Early is defined as 20ms.

3. At 20 ms (really 17-25ms) there should be a very noticeable return reflection at around -12dB below the initial impulse. This is a latter arriving reflection. It usually comes from the back wall. Often this is a diffuse sound field. The diffuse field is a much more dense 'reverb-like' tail that happens. Think of the difference between dropping 4 marbles on a metal plate at the same time verus dropping 200 marbles on a metal plate at the same time.

The later arriving signal aides with spacial imaging and is much more pleasing to listen to than a completely dead room. The 20 ms gap (ISD) is free of reflections allowing for the listener to hear the signal clearly before the latter arriving reflections occur.

There's a way better way to explain all this, I'm sure, but I suck at words.

Neil
ok so you are saying I don't have the latter arriving reflection? could that be because the back wall is made of plasterboard and cement (probably between plasterboard and cement is maybe a bit empty, but I don't know), and also there's a big window from floor to ceiling... maybe because of this sounds travels out of the room and doesn't come back in.. also there are 4 traps in the backwall and corner filled with rockwool

anyway.. not having that latter arrival reflection is a bad thing? There are ways to avoid this?

what should I do?
Old 19th December 2010
  #38
Lives for gear
 
amishsixstringe's Avatar
 

Don't worry about that yet. Really, it's not that big of a deal right now.

Honestly, your room looks pretty good on paper. Better than a lot of folks'. It's a little bit dead for my taste, but I'm sure you can use the room just fine as it is.

Neil
Old 19th December 2010
  #39
SAC
Registered User
 

hahaha - this is a bit of a mess in that i post and edit without being able to see what is simultaneously being written and posted - and I post based upon what i remember from the last glance, as i cannot see the other posts for reference as I write!!! hehhehheh

For a room ~17 feet deep, and not too wide, a 40 ms time window should get just about all we need. I don't anticipate anything too hot arriving after that.

Right now, I would just concentrate on the measurement process.
One speaker at at time driven (with no sub if present) and trying to get reproducible results.

As Amish mentioned about the gain issue, he's right, that ~10-20 dB up and down shift in level is most likely a software thing, as I am not aware of what you could be doing to cause that off hand... we just need to be sure that what we finally choose to evaluate is truly representative of the room.
Old 19th December 2010
  #40
Lives for gear
Quote:
Originally Posted by SAC View Post
Don't apologize! I think its great you're wrestling with the monster!
Thank you SAC! :D
I know I should have doing this step by step posting stuff from the beginning, but I had a limited period of time (I stretched because obviously I didn't get good results at the beginning)

Quote:
My doubts? I don't know exactly what the process is - that is my doubt. How many speakers are being drive during the measurements, type of mic (omni, etc) - all of the process issues pertaining to making proper impulse response measurements can be included here. Note, i am not questioning you, but something is happening to cause the gain flip flop mentioned earlier that should be resolved....
Ok, I try to explain what I do:

1- mic is a behringer ecm2000 (is that the right model, BTW, The Behringer :P so it is omni)

2- I've placed the mic at 38% (195 cm) of the room, exactly at half the room

3- I calibrate the mic placing a SPL level meter under/in front the mic (I mean as close to the mic as possible) and I read it oscillating around 84 db (just because I'm comfortable using 84db because I can recall it easily with the rme total mix). After that I never change the gain level or the mic position, if happen by mistake (in 3 weeks happened a couple of time) I re-position it in the same place

4- Initially I measure both speakers separately, then I measure both together, so I have an idea of what is happening in each speaker (maybe later I'll post L & R because I have to remeasure them, I haven't the recent ones), but I measure both speakers because many told me doing in this way for the lows, mids under 500hz (even Ethan if I remember correctly)

The two different measurements are different because I moved the stand a bit and toed in a bit the speakers while before where parallel.. that have probably changed a bit the level because toed in the rms is a bit higher (I suppose) but I think it will be a change of 0.5/1db max..

Quote:
And as you get everything ironed out, we will still be around to help you with the actual reflection identification process. Each step in its time...
What should I get ironed out?
Old 19th December 2010
  #41
Lives for gear
 
amishsixstringe's Avatar
 

How high is that mic off the ground? Just curious, because the one photo shows the microphone about a foot lower than the height of the tweeters on your speakers. The mic should be at the same level as those tweeters. This should also be the place you put your ears when you are listening.

Your speakers should not be parallel to the surface of your walls. They should be angled in 30 degrees each. Making an equilateral triangle that includes your head. If you put your head where it goes, and look at each of your speakers, it should be pointing directly at you. Not behind you.

Just something I may have stumbled upon.

Neil
Old 19th December 2010
  #42
Lives for gear
Quote:
Originally Posted by amishsixstringe View Post
Honestly, your room looks pretty good on paper. Better than a lot of folks'. It's a little bit dead for my taste, but I'm sure you can use the room just fine as it is.

Neil
Well, now, I've placed traps, if their placement, position, will be definitive.. I'll build a frame around them with some slats randomly spaced, in front and in the back to make them more reflective, so I hope the room will be less dead

BTW, it is dead, but not that dead :P I still have all the floor reflective, the books in the bookshelf, some "bookshelf door", part of the wall, the window.. part of the ceiling (the part over the window)

So, if the traps/rockwool stuff is now definitive, the next step will be cover most of them with spaced wood of different height and width

In fact now the room is a big mess :P but with frames, slats, I think/hope it will be better

p.s. I think probably if I do an equilateral triangle with the mic and the speakers, I'll get less comb filtering displayed, but I'll get too many highs, plus these PMC sounds better if I'm more distant to them.. so I prefer keeping them more open, parallel between them or a little toed in (is that wrong?) the PMC tweeters have a wider dispersion.. so even PMC suggest to do the triangle but 50 cm behind your head, but I can't do that keeping it equilateral because speakers are 1.5 meters tweeter to tweeter and I'm at around 160/170 from the speakers (if I remember correctly, btw more than the distance between the speakers and they are still very wide :P)
Old 19th December 2010
  #43
Lives for gear
Quote:
Originally Posted by amishsixstringe View Post
How high is that mic off the ground? Just curious, because the one photo shows the microphone about a foot lower than the height of the tweeters on your speakers. The mic should be at the same level as those tweeters. This should also be the place you put your ears when you are listening.

Neil
The mic is a little over 106 cm from the ground (3.5') while the tweeter is a little less 106 cm (I can't go lower with the stand) PMC suggest the best position is between tweeter and woofer (probably is because of the speaker design which has a better off axis response)

Quote:
Your speakers should not be parallel to the surface of your walls. They should be angled in 30 degrees each. Making an equilateral triangle that includes your head. If you put your head where it goes, and look at each of your speakers, it should be pointing directly at you. Not behind you.

Just something I may have stumbled upon.
As I said, PMC told me (giving me a manual) I should make an equilateral triangle with the "head corner" 50 cm behind my head

Dan Dan also suggest me to keep them perpendicular to the backwall so to lower the high frequencies because they are quite loud compared to the lows (it's a speaker design to improve a better off axis response) but if you want I measure even the equilateral triangle

Before I had the equilateral triangle placement and was quite fatiguing for ears, I hope in that way they will be less fatiguing.. in the meantime I'll save money to get a second amp to biamp them so I can keep the tweeter lower in volume
Old 19th December 2010
  #44
SAC
Registered User
 

Place the measurement mic at a point coincident with where the center of you head will be when working - and be aware of how you sit- bent over, sitting upright, however..

Other wise we are evaluating the response for a point where you are not working - and I am not sure what the value of that would be.

Also, as far as using the speakers off axis....lots of potential problems.
Do you know the polar response and dispersion characteristics???? By orienting them off axis toward the listening position, you are simply putting more energy into areas and onto boundaries that will more than likely simply create additional problems!

The proper selection of speakers is critical, and most have no idea what the Q and dispersion characteristics of their speaker are.
And to cause a firestorm of "but I like them", in general do not use speakers featuring the MTM/D'Appolito configuration. And if you have them, mount then VERTICALLY in order to orient the polar lobing anomalies(nulls) horizontally rather than vertically.

And you don't adjust ('EQ') the sound, either intentionally or unintentionally by listening off axis outside the high frequency dispersion or in a polar null. If you need to do that, dump the speakers and get something that performs as desired!
Old 19th December 2010
  #45
Lives for gear
Quote:
Originally Posted by SAC View Post
Place the measurement mic at a point coincident with where the center of you head will be.
Usually I place the mic exactly at my ear level, which is 106, ear is in the centre of the head right?

Quote:
Also, as far as using the speakers off axis....lots of potential problems.
Do you know the polar response and dispersion characteristics???? By orienting them off axis toward the listening position, you are simply putting more energy into areas and onto boundaries that will more than likely simply create additional problems!
I don't know the exact polar response and dispersion characteristics because PMC doesn't give you the manual, they gave me a manual of the AML saying that what was written about positioning was the same for the TB2+ but I cannot publish part of it, they expressively told me that, so you have to trust me, what they say is you have to do an equilateral triangle with your head, then move the speakers so their axes cross 2' behind your head (50cm)

I have them mounted vertically, have you seen the picture of my studio? I've posted them some posts before

Quote:
And you don't adjust ('EQ') the sound, either intentionally or unintentionally by listening off axis outside the high frequency dispersion or in a polar null. If you need to do that, dump the speakers and get something that performs as desired!
Well, those monitors are great, perform great, have a great transient response and a wider sweet spot (due to their wide dispersion design), so I'd like to keep them

When you move your head you still have a great picture and I think that is important because you often move your head, specially when producing or when use outboard
Old 19th December 2010
  #46
Lives for gear
I hope DanDan come in because he probably can explain better then me that PMC thing, he is the one who told me this, then I asked to PMC just for a confirmation and he was right.. :D (tnx Dan)

BTW Thanks to you all!

Well, I can do other measurements, so, SAC, if you want, tell me exactly what measurement do you need and what I should care of, so I can give you proper measurements to evaluate

p.s. do you have a link with a sort of tutorial to find resonances using ETC? Tomorrow I may try doing this too

THANK YOU SO MUCH SAC!!!


(I just want to keep things positive because I've spent so many time on it that's easy for me to lose motivation :P and also I really care about acoustic, sort of perfectionist.. :P)
Old 19th December 2010
  #47
SAC
Registered User
 

My comments were/are about speakers in general and apply to those speakers as well. There is no need to get into analyzing those speakers in particular.

There is a reason such speakers as the Kino****as (used in the Hidley rooms and featuring TAD horns), Westlakes and many more have used high Q controlled dispersion extended frequency horns - both for pattern control and keeping energy off the boundaries (and hence eliminating reflections through pattern control) while also eliminating a crossover in the critical 800Hz- 2500Hz/4000Hz bandpass so as not to introduce phase error/group delay errors in the middle of the our ears critical bandpass.!

Not only does orienting them toward you, on axis help insure that you are within their optimal coverage area (and out of the polar lobing nulls that 'comes with' the MTM configuration), but you minimize the dispersion of energy specifically onto the boundaries we are trying hard to treat to remove said energy!

These aspects of speaker selection have been fundamental to almost all of the modern room designs, and yet few consumer grade monitors provide any such documentation adequate to make informed product choices. I wonder why???? perhaps for the same reason they too often only provide a heavily smoothed free field anechoic frequency response that provides no anticipated spatial loading characteristics!

Use the speakers you have. But use them on axis and minimize the energy spill onto the side boundaries. If you need to lower the highs, here is one legitimate use for EQ for the adjustment of level of the direct signal.
Old 19th December 2010
  #48
Lives for gear
Back reading what SAC told me, I'm trying to understand :D

//Also, as indicated in the oval, the near clutter will be due to the speaker driver alignment itself and/or cabinet/spkr mounting diffraction and reflections from the spkr shelf mount. You want to clean this up as these are deadly with such a short delay.


The speakers are mounted over a very heavy stand (I think is around 100kg, the stand is composed by "two legs" filled with sand, 25kg of sand each, and a "bridge" 4" thick, all the stand is made of MDF 1.2" thick, the only doubt is maybe the bridge, even considering is 3 layers of MDF in total 3.6" thick, maybe vibrates a little and the part under it reflect with the floor, so I'm considering to place some 2" thick rockwool under the stand "bridge"

I find a bit difficult to place the speakers on the stand at exactly the same distance, I mean I can, but the speakers has round border and it's hard to be millimetrically precise)

What could be that deadly diffraction? I haven't used Auralex Mopad, before I had used them, but then I've take them off because the response was better :-/

Anyway, thank you so much! :D
Old 19th December 2010
  #49
Lives for gear
Quote:
Originally Posted by SAC View Post
My comments were/are about speakers in general and apply to those speakers as well. There is no need to get into analyzing those speakers in particular.

There is a reason such speakers as the Kino****as (used in the Hidley rooms and featuring TAD horns), Westlakes and many more have used high Q controlled dispersion extended frequency horns - both for pattern control and keeping energy off the boundaries (and hence eliminating reflections through pattern control) while also eliminating a crossover in the critical 800Hz- 2500Hz/4000Hz bandpass so as not to introduce phase error/group delay errors in the middle of the our ears critical bandpass.!

Not only does orienting them toward you, on axis help insure that you are within their optimal coverage area (and out of the polar lobing nulls that 'comes with' the MTM configuration), but you minimize the dispersion of energy specifically onto the boundaries we are trying hard to treat to remove said energy!

These aspects of speaker selection have been fundamental to almost all of the modern room designs, and yet few consumer grade monitors provide any such documentation adequate to make informed product choices. I wonder why???? perhaps for the same reason they too often only provide a heavily smoothed free field anechoic frequency response that provides no anticipated spatial loading characteristics!

Use the speakers you have. But use them on axis and minimize the energy spill onto the side boundaries. If you need to lower the highs, here is one legitimate use for EQ for the adjustment of level of the direct signal.
Ok thank you, so I'll biamp them (an eq will cost like an amp, so I think lowering an amp volume is more natural)

So, tomorrow homework:

- Do a left - right separate measurement (Frequency response, Waterfall, Impulse) and one with both

- Place the monitors using the equilateral triangle placement and do the same (L-R separate measurements FR, Waterfall, Impulse)

Try to figure out the speaker diffraction/mounting clutter.. even if I don't have an idea of what can cause this.. I'll try Mopads again. Don't you think maybe is caused by the fact they are not in an equilateral triangle placement? Or maybe they are angled a little differently (we are talking of eventually few millimeters.. because I pay a lot of attention when I move them)
Old 19th December 2010
  #50
SAC
Registered User
 

You do not want any 'step' transitions of the speaker baffle and/or mount.

Abrupt changes in the slope of adjacent surfaces result in the transition acting as a secondary point of transmission - almost like a small speaker.

You set them flush on stands and all surface transitions should be gently curved.

The base of the stands should not protrude in front of any part of the speaker (as is SO typical in bridge mounted speakers).

If the Mopads are like the mounts so many use that protrude in the front, do not use them.

The topic is complex, but not difficult to understand in general.


Here area few sources of varying complexity.

Frequently Asked Questions

Diffraction from baffle edges

True Audio TechTopics: Diffraction Loss
Old 19th December 2010
  #51
Lives for gear
so you just mean "place the speaker a bit out of the stand" now they are exactly on the border of the stand because it helps me keeping the symmetry when I do measurements, but I already planned to place them a bit ahead because usually I like more how they sound and the measurements are a little better :D
Old 19th December 2010
  #52
Lives for gear
 
amishsixstringe's Avatar
 

Those articles you like are great, SAC! Thanks!

Neil
Old 19th December 2010
  #53
Lives for gear
+1 yes! tnx too!
Old 19th December 2010
  #54
Lives for gear
 
amishsixstringe's Avatar
 

Also, is it logical to assume that tweeters with a wider dispersion would be more prone to worse diffraction based on the amount of energy actually even reaching the edge/baffle? If a tweeter had a perfect dispersion of say <90 degrees, this effect would not even be possible, correct?

Neil
Old 19th December 2010
  #55
Lives for gear
ok so tomorrow I will place the speakers a little out of the stand.. 2cm (1" or so)
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
Similar Threads
Thread
Thread Starter / Forum
Replies
soundgeezer / Studio building / acoustics
26
kas_lar / Studio building / acoustics
2
nixoblivion / Bass traps, acoustic panels, foam etc
13
Audiop / Gear Shoot-Outs / Sound File Comparisons / Audio Tests
11
iquinn / Mastering forum
9

Forum Jump
Forum Jump