The No.1 Website for Pro Audio
Correct settings on Genelecs after Sonarworks readings.
Old 15th September 2020 | Show parent
  #61
Lives for gear
 
Jason Foi's Avatar
 

Quote:
Originally Posted by Muskman View Post
AH, OK. Sorry for being so thorough, it's quite complex learning all these new things so apologies for all the questions.

To summarise, the sweep inside the room wasn't distorting my speakers(they start clipping at like 95db but I never listen that loud) and I tried to measure it's peak which seemed to be around 84db with the iphone SPL I bought.

Well hopefully those readings from earlier will help us move on to the next stage!

With a large blanket on my desk I then have issues from 800 down to the subs so I'm thinking the roof and side wall acoustic 10cm thick tiles need to be changed to broadband rockwool absorbers, if they might help combat mids-low mids?

Thankyou, again!
ok, so I had a look at your mdats and to me as a non REW ninja they appear to be valid. The monday one had healthier levels so whatever you did there as far as mic levels, stay with that. With any luck the true REW ninja @ akebrake will chime in and help decipher things.

Your frequency response needs some improvement, if you can get to +/- 5db swings from the average response that's considered really good for a home studio. You have some dips around -15db right now.. This isnt surprising, you need to get some proper bass traps in there to help iron things out a bit, and the big dip around 100hz could be SBIR related or maybe floor bounce. Can you set your crossover point to like 120hz? Your decay times weren't terrible. Perhaps sonar works can work wonders in your room, once you get rid of phase cancellations DSP can work wonders.

Was your sub on during measurements? if so, turn it up a bit, you need a little more low end.

Your desk is definitely causing issues. It shows up at about 1ms-1.5ms, and there is another spike at ~16ms (might be back wall reflections??). When you placed the blanket over your desk, the 1ms spikes vanished, which is good. Try and tilt your desk up towards you. Put a book or something under the back legs and try to angle the desktop about 12 degrees. You might be able to redirect the reflection out of your LP.

There's really only so far you're going to get in this room without adding more bass trapping, but try angling the desk and see if that helps. Hopefully you get someone a little more knowledgeable than me to go over your measurements and point out the things I've missed.
Old 15th September 2020 | Show parent
  #62
Lives for gear
 

Quote:
Originally Posted by Jason Foi View Post
ok, so I had a look at your mdats and to me as a non REW ninja they appear to be valid. The monday one had healthier levels so whatever you did there as far as mic levels, stay with that. With any luck the true REW ninja @ akebrake will chime in and help decipher things.

Your frequency response needs some improvement, if you can get to +/- 5db swings from the average response that's considered really good for a home studio. You have some dips around -15db right now.. This isnt surprising, you need to get some proper bass traps in there to help iron things out a bit, and the big dip around 100hz could be SBIR related or maybe floor bounce. Can you set your crossover point to like 120hz? Your decay times weren't terrible. Perhaps sonar works can work wonders in your room, once you get rid of phase cancellations DSP can work wonders.

Was your sub on during measurements? if so, turn it up a bit, you need a little more low end.

Your desk is definitely causing issues. It shows up at about 1ms-1.5ms, and there is another spike at ~16ms (might be back wall reflections??). When you placed the blanket over your desk, the 1ms spikes vanished, which is good. Try and tilt your desk up towards you. Put a book or something under the back legs and try to angle the desktop about 12 degrees. You might be able to redirect the reflection out of your LP.

There's really only so far you're going to get in this room without adding more bass trapping, but try angling the desk and see if that helps. Hopefully you get someone a little more knowledgeable than me to go over your measurements and point out the things I've missed.
Hey mate, thanks for your time again.

So firstly these readings are without the sub. Let’s ignore I said about the sub for now. These are carried out with the speakers flat so no 85hhz dip.
They were 40cm from the front wall but I can carry out the same measurements but with them almost all the way up to the wall to see the difference with SBIR as a first shot at changing position.

If we can split how I remedy things into frequency ranges it may be helpful for me.

The highs we can clearly see the blanket helps vastly to improve it all so I will have to do some research and maybe chose an entirely different desk setup as a compromise for sound quality.
I was wondering if people installed treatment on their desks as a weird workaround!

I notice the dips start at like 870ish downwards.

Would you think that the thick absorbers in my wall and ceiling would work to absorb 870hz?
I’m trying to work out what the frequency (generally) the decent acoustic absorbers will work with and if I should buy broadband ones with an air gap from the likes of GIK?

“ big dip around 100hz could be SBIR related or maybe floor bounce” what do you mean by floor bounce?

“ once you get rid of phase cancellations ”
How do I identify what these are, where I am getting them and how to get rid of them please?

Maybe if I run new tests with the speakers close to the front wall and with/without sub & with/without blanket we can see what happens there.

The broadband style early reflection panels, what frequency range do they cover? What’s classed as bass trapping?
I built floor to ceiling ones on the back wall.
The front wall is quite difficult because inside the room one the right corner I have 30cm width I can trap and the left corner is an old fire which I built traps around. As silly as this sounds, I can add more traps on the other side of the stud wall or is that pointless as it’s inside another room?

Thankyou mate.
Old 15th September 2020 | Show parent
  #63
Quote:
Originally Posted by Muskman View Post
Hey mate, thanks for your time again.

So firstly these readings are without the sub. Let’s ignore I said about the sub for now. These are carried out with the speakers flat so no 85hhz dip.
They were 40cm from the front wall but I can carry out the same measurements but with them almost all the way up to the wall to see the difference with SBIR as a first shot at changing position.

If we can split how I remedy things into frequency ranges it may be helpful for me.

The highs we can clearly see the blanket helps vastly to improve it all so I will have to do some research and maybe chose an entirely different desk setup as a compromise for sound quality.
I was wondering if people installed treatment on their desks as a weird workaround!

I notice the dips start at like 870ish downwards.

Would you think that the thick absorbers in my wall and ceiling would work to absorb 870hz?
I’m trying to work out what the frequency (generally) the decent acoustic absorbers will work with and if I should buy broadband ones with an air gap from the likes of GIK?

“ big dip around 100hz could be SBIR related or maybe floor bounce” what do you mean by floor bounce?

“ once you get rid of phase cancellations ”
How do I identify what these are, where I am getting them and how to get rid of them please?

Maybe if I run new tests with the speakers close to the front wall and with/without sub & with/without blanket we can see what happens there.

The broadband style early reflection panels, what frequency range do they cover? What’s classed as bass trapping?
I built floor to ceiling ones on the back wall.
The front wall is quite difficult because inside the room one the right corner I have 30cm width I can trap and the left corner is an old fire which I built traps around. As silly as this sounds, I can add more traps on the other side of the stud wall or is that pointless as it’s inside another room?

Thankyou mate.
When you filter the ETC, the energy spike is 10khz or 6300khz. Instead of trying to absorb the energy spike, you can just angle the desk as the these wavelengths are really short.

The dips around 118hz-119hz are SBIR. The last measurement right blanket on desk with the dip at 872hz is SBIR.
Old 15th September 2020 | Show parent
  #64
Lives for gear
 

Quote:
Originally Posted by thethrillfactor View Post
When you filter the ETC, the energy spike is 10khz or 6300khz. Instead of trying to absorb the energy spike, you can just angle the desk as the these wavelengths are really short.

The dips around 118hz-119hz are SBIR. The last measurement right blanket on desk with the dip at 872hz is SBIR.
Hi great I will try the simple desk angling trick to see how that changes.

I’ll move my speakers closer to the wall, is there a graph that explains which frequencies are affected by SBIR. I ask as if you are able to identify that from those readings then you must know your stuff!

Thankyou
Old 15th September 2020 | Show parent
  #65
Lives for gear
 
akebrake's Avatar
 

Quote:
Originally Posted by Muskman View Post
...
I recorded them at -45 sweep with synths on desk incase this is distorted?
Distortion is now along with Noise floor which is good. pic 1

You find it in the Distortion window together with Noise floor which is the background noise in your room (captured just before the sine sweep starts).
Noise will of course vary with cars etc. passing by.

Just wondering about the noise peak at 16kHz. A fan or hard drive? Or some other interfering gear?

I’m not familiar with Steinberg UR44 soundcard and what options you’ve got to improve gain staging through the system (like Jason suggested).

It’s possible that a later REW (5.20 beta xx version) works better with your OS/ Sound card drivers? I dont know... Y can ask on AV Nirvana/REW support.
(Just trying to help y get valid measurements )

As Thrill pointed out the table and back of synths (?) reflects strongly (0.5ms after the direct sound) which in turn ”fools” REW to catch on the second impulse as ”Direct”.
Not visible if you dont zoom in. (pic 2) and also L/R FR overlayed. pic 3

This might be important if one is picky with timing.

Maybe a reference measurement without table?
Attached Thumbnails
Correct settings on Genelecs after Sonarworks readings.-skarmavbild-2020-09-15-kl.-14.04.21.jpg   Correct settings on Genelecs after Sonarworks readings.-15-l-impulse.jpg   Correct settings on Genelecs after Sonarworks readings.-l-r-ovly-misalign.jpg  
Old 15th September 2020 | Show parent
  #66
Lives for gear
 

Quote:
Originally Posted by akebrake View Post
Distortion is now along with Noise floor which is good. pic 1

You find it in the Distortion window together with Noise floor which is the background noise in your room (captured just before the sine sweep starts).
Noise will of course vary with cars etc. passing by.

Just wondering about the noise peak at 16kHz. A fan or hard drive? Or some other interfering gear?

I’m not familiar with Steinberg UR44 soundcard and what options you’ve got to improve gain staging through the system (like Jason suggested).

It’s possible that a later REW (5.20 beta xx version) works better with your OS/ Sound card drivers? I dont know... Y can ask on AV Nirvana/REW support.
(Just trying to help y get valid measurements )

As Thrill pointed out the table and back of synths (?) reflects strongly (0.5ms after the direct sound) which in turn ”fools” REW to catch on the second impulse as ”Direct”.
Not visible if you dont zoom in. (pic 2) and also L/R FR overlayed. pic 3

This might be important if one is picky with timing.

Maybe a reference measurement without table?
Hi, thanks for your reply.

I think the 16khz was something at the time and isn't consistent on any other readings, much have been background noise or something from across the road ha ha.

I am trying to find a happy balance of mix quality and workflow with as little drastic changes as possible.

I can try a reference without a desk but I couldn't really work without one so I'm not sure what I can do!

I think the best possible layout with angles suited to the speakers is my only option.

I must reiterate that I am far from professional and this is my third space with this being the first where I have done any proper measurements!

Thanks guys.
Old 15th September 2020 | Show parent
  #67
Quote:
Originally Posted by Muskman View Post
is there a graph that explains which frequencies are affected by SBIR. I ask as if you are able to identify that from those readings then you must know your stuff!

Thankyou
In REW select the GD tab and under controls select generate minimun phase.

Any peaks in the excess group delay measurement are outside of a minimum phase correction( with a minimum phase EQ) such as a modal peak in the low end( which would show up as a flattish line).

This is some kind of SBIR if the peak on the excess group delay measurement is in the low end below schroeder.

Above schroeder it can be SBIR or having tweeters that are wired out of phase( believe it or not this happens more often than not) or something else.
But because the peaks in the highs are so many and so narrow, just turning your head or sliding over an inch can make some of the peaks(nulls in frequency response) disappear.
Attached Thumbnails
Correct settings on Genelecs after Sonarworks readings.-screen-shot-2020-09-15-4.51.29-pm.jpg  
Old 16th September 2020 | Show parent
  #68
Lives for gear
 

Quote:
Originally Posted by thethrillfactor View Post
In REW select the GD tab and under controls select generate minimun phase.

Any peaks in the excess group delay measurement are outside of a minimum phase correction( with a minimum phase EQ) such as a modal peak in the low end( which would show up as a flattish line).

This is some kind of SBIR if the peak on the excess group delay measurement is in the low end below schroeder.

Above schroeder it can be SBIR or having tweeters that are wired out of phase( believe it or not this happens more often than not) or something else.
But because the peaks in the highs are so many and so narrow, just turning your head or sliding over an inch can make some of the peaks(nulls in frequency response) disappear.

Thankyou.

“Any peaks in the excess group delay measurement are outside of a minimum phase correction( with a minimum phase EQ) such as a modal peak in the low end( which would show up as a flattish line).“

What does this mean I need to do? I don’t know if you noticed I said I’m a complete noob! Ha ha. Sorry.

I really don’t understand what your saying to me about Schroeder and how/what I can attempt to remedy this stuff.

Sorry and thanks.
Old 16th September 2020 | Show parent
  #69
Lives for gear
 

[QUOTE=thethrillfactor;14975188]In REW select the GD tab and under controls select generate minimun phase.

Any peaks in the excess group delay measurement are outside of a minimum phase correction( with a minimum phase EQ) such as a modal peak in the low end( which would show up as a flattish line).

This is some kind of SBIR if the peak on the excess group delay measurement is in the low end below schroeder.

Above schroeder it can be SBIR or having tweeters that are wired out of phase( believe it or not this happens more often than not) or something else.
But because the peaks in the highs are so many and so narrow, just turning your head or sliding over an inch can make some of the peaks(nulls in frequency response) disappear.[/QUOTE

What would be great is if I could get some ideas of what could be causing all the dips from 800hz downwards and what sort of things I can do to remedy them.
When does the acoustic foam tiles stop being effective(generally) on the frequency spectrum.

They are 10cm thick so I’m trying to breakdown what I currently have and it’s effectiveness.

Do the readings as Jason says, tell me that I have a big issue with the low end?

Thanks
Old 16th September 2020 | Show parent
  #70
Lives for gear
 
Jason Foi's Avatar
 

[QUOTE=Muskman;14976370]
Quote:
Originally Posted by thethrillfactor View Post
In REW select the GD tab and under controls select generate minimun phase.

Any peaks in the excess group delay measurement are outside of a minimum phase correction( with a minimum phase EQ) such as a modal peak in the low end( which would show up as a flattish line).

This is some kind of SBIR if the peak on the excess group delay measurement is in the low end below schroeder.

Above schroeder it can be SBIR or having tweeters that are wired out of phase( believe it or not this happens more often than not) or something else.
But because the peaks in the highs are so many and so narrow, just turning your head or sliding over an inch can make some of the peaks(nulls in frequency response) disappear.[/QUOTE

What would be great is if I could get some ideas of what could be causing all the dips from 800hz downwards and what sort of things I can do to remedy them.
When does the acoustic foam tiles stop being effective(generally) on the frequency spectrum.

They are 10cm thick so I’m trying to breakdown what I currently have and it’s effectiveness.

Do the readings as Jason says, tell me that I have a big issue with the low end?

Thanks
You can use more bass trapping to help iron out the low end, you need to turn your sub on to lift the low end a bit. Your foam is working for taming first reflections, but its not effective enough in the lower frequency range to flatten out your frequency response.

The phase cancellations from your desk and gear, we told you, angle your desk to redirect reflections away from the listening position.

SBIR, has to do with speaker placement vs room boundaries. Typically you want your monitors push right up against the wall. Moving your speakers will change the frequency of the null. For your 120hz null, like i said, you can set your cutoff for your sub high enough to fill that in. The 800hz null put absorbtion effectove to treat 800hz on the offending surface.


http://tripp.com.au/sbir.htm
Old 16th September 2020 | Show parent
  #71
Lives for gear
 

[QUOTE=Jason Foi;14976606]
Quote:
Originally Posted by Muskman View Post
You can use more bass trapping to help iron out the low end, you need to turn your sub on to lift the low end a bit. Your foam is working for taming first reflections, but its not effective enough in the lower frequency range to flatten out your frequency response.

The phase cancellations from your desk and gear, we told you, angle your desk to redirect reflections away from the listening position.

SBIR, has to do with speaker placement vs room boundaries. Typically you want your monitors push right up against the wall. Moving your speakers will change the frequency of the null. For your 120hz null, like i said, you can set your cutoff for your sub high enough to fill that in. The 800hz null put absorbtion effectove to treat 800hz on the offending surface.


http://tripp.com.au/sbir.htm

Hi mate, ok I hear you.

800hz and below, what’s the frequency range of bass trapping. If I can’t fill my front wall corners (on the floor) with bass traps but can do the ceiling corners and back wall I can add even more bass traps, would this work or is the front wall the biggest offender?

I’m gonna angle my desk and move the speakers back to the wall tomorrow and get new measurements.

What I’m trying to find out is if my readings are horrendous for a general guy who writes and makes electronic music?

I have always referenced stuff on several other systems due to never having a room even close to professional in the past! This has worked ok so far, had a Beatport top 10 last month, not that that means anything! Ha ha.

I’m actually considering some sort of sheet I pull over my desk when I do my final mix and just get used to how it is now.

Also, I will add the sub. Would you recommend rolling off the 8020s at 85hz then? I’ll go beyond that with my sub as the crossover isn’t much of an issue apparently ✌️.

Thankyou
Old 16th September 2020 | Show parent
  #72
Lives for gear
 
Jason Foi's Avatar
 

Add a sub instead of even more bass trapping to even out freq response?

At what depth does fiberglass bat insulation become the better choice over rock wool?

I dont know the gfr of foam so i cant model it for you, but i would guess at 4" thick mounted flat on the wall it stops being effective somewhere around 400hz.

Foam works, its just very expensive on a cost vs performance basis.
Old 16th September 2020 | Show parent
  #73
Lives for gear
 
Jason Foi's Avatar
 

[QUOTE=Muskman;14976633]
Quote:
Originally Posted by Jason Foi View Post


Hi mate, ok I hear you.

800hz and below, what’s the frequency range of bass trapping. If I can’t fill my front wall corners (on the floor) with bass traps but can do the ceiling corners and back wall I can add even more bass traps, would this work or is the front wall the biggest offender?

I’m gonna angle my desk and move the speakers back to the wall tomorrow and get new measurements.

What I’m trying to find out is if my readings are horrendous for a general guy who writes and makes electronic music?

I have always referenced stuff on several other systems due to never having a room even close to professional in the past! This has worked ok so far, had a Beatport top 10 last month, not that that means anything! Ha ha.

I’m actually considering some sort of sheet I pull over my desk when I do my final mix and just get used to how it is now.

Also, I will add the sub. Would you recommend rolling off the 8020s at 85hz then? I’ll go beyond that with my sub as the crossover isn’t much of an issue apparently ✌️.

Thankyou
Moving your speakers will change sbir nulls and phase cancellation nulls. Mark current locations of the speakers in case its worse, not better.

Sub crossover to ~120-130hz

Lol, bass traps are effective at bass frequencies, 200hz and below.

Angle your desk and you shouldnt need the sheet.

Sonar works is going to make a positive impact, just get as good as you can before you use it.

Top 10 on beatport is impressive to me. Congratulations. My best is top 100 in sub genre, lol
Old 16th September 2020 | Show parent
  #74
Lives for gear
 

@ Jason Foi

thx for steering the op in the right direction, first to get his measurement rig going and then to interpret data.

just one thing i'd like to add: it's mostly not needed to run sweeps/pink noise at listening levels - to illustrate: way too often, i get to shoot the pa with the audience already in so no way on earth i'd assault people with a full blast of pink noise (but i use band-limited sweeps)! results to full blast (especially in the mids to highs) usually don't look much different...

...but then the levels i'm talking about are often much higher! :-)
Old 16th September 2020 | Show parent
  #75
Lives for gear
 
Jason Foi's Avatar
 

Quote:
Originally Posted by deedeeyeah View Post
@ Jason Foi

thx for steering the op in the right direction, first to get his measurement rig going and then to interpret data.

just one thing i'd like to add: it's mostly not needed to run sweeps/pink noise at listening levels - to illustrate: way too often, i get to shoot the pa with the audience already in so no way on earth i'd assault people with a full blast of pink noise (but i use band-limited sweeps)! results to full blast (especially in the mids to highs) usually don't look much different...

...but then the levels i'm talking about are often much higher! :-)
Thanks deedee, i know acoustics wont change, but if he sweeps at normal playback volume it will show up any distortion and show noise floor vs signal
Old 16th September 2020 | Show parent
  #76
Lives for gear
 

Quote:
Originally Posted by Jason Foi View Post
(...) sweeps at normal playback volume it will show up any distortion and show noise floor vs signal
yep - i just wanted to point out that in terms of measurement of frequency response (from low mids to highs) and reflections (which so far imo are the op's biggest concerns), even a flawed measurement can reveal a few things and my recommendation to initially turn off the sub was pointing in the same direction: with the sub on, issues could have been further masked to some degree.

of course for some other data (distortion, s/n, decay), both the system must work properly and the sub must be turned on again.
Old 16th September 2020 | Show parent
  #77
Lives for gear
 

Quote:
Originally Posted by Jason Foi View Post
Add a sub instead of even more bass trapping to even out freq response?

At what depth does fiberglass bat insulation become the better choice over rock wool?

I dont know the gfr of foam so i cant model it for you, but i would guess at 4" thick mounted flat on the wall it stops being effective somewhere around 400hz.

Foam works, its just very expensive on a cost vs performance basis.
Interesting. So if bass traps are kinda generally good at 200 and below. What is the best treatment to cover 200hz upward then? In theory, IF my treatment was great I'd still maybe have 200-400hz untreated!

Cheers
Old 16th September 2020 | Show parent
  #78
Lives for gear
 

[QUOTE=Jason Foi;14976657]
Quote:
Originally Posted by Muskman View Post
Moving your speakers will change sbir nulls and phase cancellation nulls. Mark current locations of the speakers in case its worse, not better.

Sub crossover to ~120-130hz

Lol, bass traps are effective at bass frequencies, 200hz and below.

Angle your desk and you shouldnt need the sheet.

Sonar works is going to make a positive impact, just get as good as you can before you use it.

Top 10 on beatport is impressive to me. Congratulations. My best is top 100 in sub genre, lol
I'm gonna use all of your fantastic advice. Make measurements of where they are now and move everything as you think. Desk included, see what works out as the best option for me.

The biggest issues when I've mixed in my previous rooms (no measurements at all, just treatment in early reflection + bass traps) has been low mids. My mastering guy always said that was the problem area from my last space.
This is why I have asked quite a lot of questions regarding the mids and how I could/should remedy them. Along with the highs & lows of course!

Thanks again!!
Old 16th September 2020 | Show parent
  #79
Lives for gear
 
Jason Foi's Avatar
 

Quote:
Originally Posted by Muskman View Post
Interesting. So if bass traps are kinda generally good at 200 and below. What is the best treatment to cover 200hz upward then? In theory, IF my treatment was great I'd still maybe have 200-400hz untreated!

Cheers
Broadband traps cover, as their name says, a broad band of frequencies. Which frequencies will depend on the material, and distance the front of the panel is from the wall. In my studio my rear wall is 4m x 2.4m x 1m of solid fluffy attic insulation with a gas flow resistivity of 5000. I get some useable absorbtion down to 20hz with fantastic absorbtion from 50hz and up.

A velocity bass trap will absorb the whole spectrum unless you range limit it with srtips or wood or something.

A pressure bass trap will only absorb a very narrow bandwidth and reflect everything above its Fc.

So, if i were you i would put as much deep fiber on the back wall as you can stand, and then range limit the top and bottom thirds of the trap to return some mid to high frequency enegy

I'll attach a pic of my back wall for clarification
Attached Thumbnails
Correct settings on Genelecs after Sonarworks readings.-20190511_162250.jpg  
Old 16th September 2020 | Show parent
  #80
Lives for gear
 

Quote:
Originally Posted by Jason Foi View Post
Broadband traps cover, as their name says, a broad band of frequencies. Which frequencies will depend on the material, and distance the front of the panel is from the wall. In my studio my rear wall is 4m x 2.4m x 1m of solid fluffy attic insulation with a gas flow resistivity of 5000. I get some useable absorbtion down to 20hz with fantastic absorbtion from 50hz and up.

A velocity bass trap will absorb the whole spectrum unless you range limit it with srtips or wood or something.

A pressure bass trap will only absorb a very narrow bandwidth and reflect everything above its Fc.

So, if i were you i would put as much deep fiber on the back wall as you can stand, and then range limit the top and bottom thirds of the trap to return some mid to high frequency enegy

I'll attach a pic of my back wall for clarification
Hi, wow! Loads of useful insight! Looks really good, great job!

I have diffusers at the height of the speakers and below them I actually have a sofa which will arrive in two week which I think will do a job at asborbing also.
Old 17th September 2020 | Show parent
  #81
Lives for gear
 
akebrake's Avatar
 

Goal: Valid measurements

Quote:
Originally Posted by Muskman View Post
...I think the 16khz was something at the time and isn't consistent on any other readings, much have been background noise or something from across the road ha ha.
The 16 kHz peak is present in all the measurements. Might be A XREF 20 Mic "personality"?


But it’s generally good to know if my measurements are valid, as REW is a sensitive instrument.

1. Always make more than one sweep and see if they are repeatable. If so, keep one.
2. Checking the Distortion tab after a sweep is a good habit.
3. Just after a sweep is performed one can check the Scope tab (Captured tab in 5.20) It shows the sent sweep signal and received signals overlaid. Head room? Clipping?
4. The Sound Card calibration curve (loop back) can reveal SoundCard problems, clocking error, monitor leakage (feed back) or if wrong file is loaded.
5. Mic calibration. Is it fairly flat or looking strange? May be not needed.


I compiled some of your measured Distortion plots. Right speaker, Some with the SUB.

Fundamental and Harmonic distortion (”over tones" or "harmonics") are linked and the THD doesn’t change much between the different calibration trials as THD was relatively low. Muskman, you were correct here.
But the dB scaling (SPL”calibration”) was changing.
In the early (sept 13) measurements the noise floor was around -5dB SPL which was false of course (a lot to low). This was then adjusted to the more believable 25-30 dB.

dBr means Distortion (in dB) relative to Fundamental. It stays approximately the same between your measurements, but the Noise floor changes.
3% THD means: Total Harmonic Distortion is ≈30 dB down. dB= 20log (0.03)

8020 Loud speaker spec says: 85dB @ 1m 50- 100 Hz < 3% THD

Best

EDIT: New pics w info added
Attached Thumbnails
Correct settings on Genelecs after Sonarworks readings.-pic-1.jpg   Correct settings on Genelecs after Sonarworks readings.-pic-2.jpg   Correct settings on Genelecs after Sonarworks readings.-pic-3.jpg   Correct settings on Genelecs after Sonarworks readings.-pic-4.jpg   Correct settings on Genelecs after Sonarworks readings.-pic-5.jpg  


Last edited by akebrake; 17th September 2020 at 02:27 PM.. Reason: New pics
Old 17th September 2020 | Show parent
  #82
Lives for gear
 

Quote:
Originally Posted by akebrake View Post
The 16 kHz peak is present in all the measurements. Might be A XREF 20 Mic "personality"?


But it’s generally good to know if my measurements are valid, as REW is a sensitive instrument.

1. Always make more than one sweep and see if they are repeatable. If so, keep one.
2. Checking the Distortion tab after a sweep is a good habit.
3. Just after a sweep is performed one can check the Scope tab (Captured tab in 5.20) It shows the sent sweep signal and received signals overlaid. Head room? Clipping?
4. The Sound Card calibration curve (loop back) can reveal SoundCard problems, clocking error, monitor leakage (feed back) or if wrong file is loaded.
5. Mic calibration. Is it fairly flat or looking strange? May be not needed.


I compiled some of your measured Distortion plots. Right speaker, Some with the SUB.

Fundamental and Harmonic distortion (”over tones" or "harmonics") are linked and the THD doesn’t change much between the different calibration trials as THD was relatively low. Muskman, you were correct here.
But the dB scaling (SPL”calibration”) was changing.
In the early (sept 13) measurements the noise floor was around -5dB SPL which was false of course (a lot to low). This was then adjusted to the more believable 25-30 dB.

dBr means Distortion (in dB) relative to Fundamental. It stays approximately the same between measurements, but the Noise floor changes.
3% THD means: Total Harmonic Distortion is ≈30 dB down. dB= 20log (0.03)

8020 Loud speaker spec says: 85dB @ 1m 50- 100 Hz < 3% THD

Best

PS Hope the screen resolution is good enough to make the small letters readable...
Hi,

Thankyou for your in depth info!

I think it’s best if I record new measurements now I have followed Jason’s instructions for setting things up correctly. Just so I don’t throw off anyone into the wrong tangent!

Should I calibrate my soundcard? If so is that where you create a feedback loop from an out to an in?

My XREF calib file was completely flat until about 5k then it had a big shelf peak so when I load it in it seems to do a normal job of balancing out the highs.

Referencing your bullet points.

1. I will do more sweeps to ensure I’m getting consistency.
2. Ok, what should I be cautious of there? The spike at 16k?
3. When the REW does the sweep it tells me if I have it too loud because it cuts the measurement short and says it’s clipping. Have any of the other results shown that they are clipping?
4. , 5 answered above I guess

“Fundamental and Harmonic distortion (”over tones" or "harmonics") are linked and the THD doesn’t change much between the different calibration trials as THD was relatively low. Muskman, you were correct here.


But the dB scaling (SPL”calibration”) was changing.
In the early (sept 13) measurements the noise floor was around -5dB SPL which was false of course (a lot to low). This was then adjusted to the more believable 25-30 dB.”

When I did it at the start it was prior to calibrating the speakers volume to 80ish dB, the SPL meter to suit and the sweep volume so I think that is why it’s inconsistent?

Once I’d followed Jason’s info my noise floor when I left my mic setup with REW at the measure screen was around like -40dbfs (or whatever it’s measured in 🤣)
Is this plausible?

The earlier ones vs the newer ones, the only major difference was that I tested the original on a Sunday where the industrial unit across the road wasn’t operating vs the Monday when it was and had a annoying fork lift truck reversing with a siren on it! I tried to test inbetween breaks!!

Tonight I will get several new readings in the position with and without desk tilted. Save the placements of the speakers then put them up against the wall and move listening position accordingly and test with and without desk tilted.

I will carry out these test with the genelec built in 85hz roll off and my sub engaged at a cut off of around120-130hz.

My speakers comfortable output was 80db but how would I measure what volume I should have for my sub to be complimenting the speakers without it being too loud or quiet? Any suggestions so I don’t end up with too booming bass or too quiet again!

Thanks!!
Old 17th September 2020 | Show parent
  #83
Lives for gear
 
akebrake's Avatar
 

Quote:
Originally Posted by Muskman View Post
Thankyou for your in depth info!


Quote:
Should I calibrate my soundcard? If so is that where you create a feedback loop from an out to an in?
Probably not necessary now.
Only make sure Direct monitoring = OFF so you don't hear the microphone in the speaker (feed back)

Quote:
2. Ok, what should I be cautious of there? The spike at 16k?
It would be interesting to find out what it is. I don’t know. Not urgent.
What happens if you sample at 48kHz? (Mac, UR44 & REW )

Quote:
3. When the REW does the sweep it tells me if I have it too loud because it cuts the measurement short and says it’s clipping.
If one "records" too hot it clips. Some head room is needed because Room modes will make the received signal very "dynamic".

Check if gain staging through the system can be improved.

Quote:
Once I’d followed Jason’s info my noise floor when I left my mic setup with REW at the measure screen was around like -40dbfs (or whatever it’s measured in ������) Is this plausible?
If one sweeps at ≈ 80 dB SPL and measure in a room with ordinary walls and windows (not a floated room within a room style studio) the Noise floor is around that figure. Some times worse / fork lift

The IR measurement technique (REW log FM sweep)) will improve the measured result and it will be adequate for reflex hunting.

Good Luck!

Last edited by akebrake; 24th September 2020 at 05:11 PM..
Old 17th September 2020 | Show parent
  #84
Lives for gear
 

[QUOTE=akebrake;14979420]
Quote:
Originally Posted by Muskman View Post
Thankyou for your in depth info!





Probably not necessary now.
Only make sure Direct monitoring = OFF so you don't hear the microphone in the speaker (feed back)



It would be interesting to find out what it is. I don’t know. Not urgent.
What happens if you sample at 48kHz? (Mac, UR44 & REW )



If one "records" too hot it clips. Some head room is needed because Room modes will make the received signal very "dynamic".

Check if gain staging through the system can be improved.



If one sweeps at ≈ 80 dB SPL and measure in a room with ordinary walls and windows (not a floated room within a room style studio) the Noise floor is around that figure. Some times worse / fork lift

The IR measurement technique (REW log FM sweep)) will improve the measured result and it will be adequate for reflex hunting.

Good Luck!
Thanks, AGAIN!

I just recorded some more measurements but there's some other studios doing stuff around so it may have affected the noise flloor.

These new results are with sub ON, speakers rolled off at 85hz. Desk tilted and I did some meaurements at speakers 40cm from wall with listening position at 38%.

I then recorded the same setup with the speakers close to the front wall and me closer to the front wall. You can see it improves the SBIR.

I have also reduced desk furniture to just a Nord Lead2x, computer keyboard & mouse in early reflection point with some acoustic foam on the back of the synth but not sure of peoples opinions on this idea?

Thanks!
Attached Files
File Type: mdat SPEAKERS UP TO FRONT WALL .mdat (1.50 MB, 2 views) File Type: mdat 40CM + TILT + SUB.mdat (10.48 MB, 2 views) File Type: mdat SPEAKERS UP TO FRONT WALL .mdat (1.50 MB, 2 views) File Type: mdat SPEAKERS UP TO FRONT WALL .mdat (10.49 MB, 2 views)
Old 17th September 2020 | Show parent
  #85
Lives for gear
 

Quote:
Originally Posted by Jason Foi View Post
Broadband traps cover, as their name says, a broad band of frequencies. Which frequencies will depend on the material, and distance the front of the panel is from the wall. In my studio my rear wall is 4m x 2.4m x 1m of solid fluffy attic insulation with a gas flow resistivity of 5000. I get some useable absorbtion down to 20hz with fantastic absorbtion from 50hz and up.

A velocity bass trap will absorb the whole spectrum unless you range limit it with srtips or wood or something.

A pressure bass trap will only absorb a very narrow bandwidth and reflect everything above its Fc.

So, if i were you i would put as much deep fiber on the back wall as you can stand, and then range limit the top and bottom thirds of the trap to return some mid to high frequency enegy

I'll attach a pic of my back wall for clarification
Hey mate.

So I tilted my desk and tested at the 38% position then moved the speakers to the back wall which you can see SBIR is improved in the files on the post above.

Just averaged the readings out Green is front wall + sub. Orange is 40cm away from wall + sub.

Green sounds a lot better in the room. And seems the best overall option right now?
You think I need to now remedy the bass in the room?
I got a 10db boost at like 200hz down to 85hz now but I cant work out if thats my sub taking over somehow?

Cheers
Attached Thumbnails
Correct settings on Genelecs after Sonarworks readings.-screenshot-2020-09-17-19.42.11.jpg  
Old 17th September 2020 | Show parent
  #86
Lives for gear
 
akebrake's Avatar
 

Quote:
Originally Posted by Muskman View Post
... I just recorded some more measurements but there's some other studios doing stuff around so it may have affected the noise flloor.
Hi,

It's bed time for me now but I observed you messed up the calibration on a couple of measurements. Some have no mic calibration.
Some have a faulty Sound Card Calibration.

Clicking "Change Cal" (below a measurement description) will open a new window with option to Browse or Clear a calibration.

Best
Attached Thumbnails
Correct settings on Genelecs after Sonarworks readings.-calib-err.jpg  
Old 17th September 2020 | Show parent
  #87
Lives for gear
 
Jason Foi's Avatar
 

Quote:
Originally Posted by Muskman View Post
Hey mate.

So I tilted my desk and tested at the 38% position then moved the speakers to the back wall which you can see SBIR is improved in the files on the post above.

Just averaged the readings out Green is front wall + sub. Orange is 40cm away from wall + sub.

Green sounds a lot better in the room. And seems the best overall option right now?
You think I need to now remedy the bass in the room?
I got a 10db boost at like 200hz down to 85hz now but I cant work out if thats my sub taking over somehow?

Cheers
Try flipping the dip switches on the back of the monitors to 1/2 space and 1/4 space and see if either setting helps with the 10db bump flatten out. I'll take a look at your measurements later and get back to you
Old 18th September 2020 | Show parent
  #88
Lives for gear
 

Quote:
Originally Posted by Jason Foi View Post
Try flipping the dip switches on the back of the monitors to 1/2 space and 1/4 space and see if either setting helps with the 10db bump flatten out. I'll take a look at your measurements later and get back to you
Hey mate,

My model doesn't have that function. They are the older ones. I have a treble tilt, bass roll off at 85hz and then two additional bass cuts like 3 +6db.

I am trying to work out a way to adjust the sub volume to suit the 80db I calibrated for the monitor volumes. If I push it too much it sounds great but then it wont be flat! Any suggestions for this maybe?
Old 18th September 2020 | Show parent
  #89
Lives for gear
 

Quote:
Originally Posted by akebrake View Post
Hi,

It's bed time for me now but I observed you messed up the calibration on a couple of measurements. Some have no mic calibration.
Some have a faulty Sound Card Calibration.

Clicking "Change Cal" (below a measurement description) will open a new window with option to Browse or Clear a calibration.

Best
Hi, yeah I was adding in the calib files there. I must have got distracted. I have amended them so they have the XREF calib file for the mic but NO calib file on the soundcard. Need to investigate how I calibrated it wrong. Still no idea where the 16khz dist comes from as there's no consistent high pitched sounds I can hear anywhere! Unless it's the speakers distorting or something?

Attached you now have readings with mic calib files only.

Same as before 40cm away + up to front wall. Desk tilted, sub on and some acoustic foam resting on back of synth in desk early reflection point for a test.
Old 18th September 2020 | Show parent
  #90
Lives for gear
 

Quote:
Originally Posted by akebrake View Post
Hi,

It's bed time for me now but I observed you messed up the calibration on a couple of measurements. Some have no mic calibration.
Some have a faulty Sound Card Calibration.

Clicking "Change Cal" (below a measurement description) will open a new window with option to Browse or Clear a calibration.

Best
Also,

As I was asking Jason below.

These readings are different from earlier in the week because I have now introduced my sub to cut off at about 130hz(guessing based on the front dial of my sub) and rolled off the Genelec 8020 at 85hz. I notice the bottom end is considerably louder on the readings now and I am probably giving a false impression as I need some guidance on how to calibrate the sub volume. I set up the sweep based on comfortable listening at 80ish DB but does this method need applying to the sub somehow?

Thanks again.
📝 Reply

Similar Threads

Thread / Thread Starter Replies / Views Last Post
replies: 4542 views: 1199963
Avatar for throbert
throbert 1 day ago
replies: 87 views: 45417
Avatar for robearth69
robearth69 1st May 2015
replies: 935 views: 107999
Avatar for danger
danger 13 hours ago
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…
🖨️ Show Printable Version
✉️ Email this Page
🔍 Search thread
🎙️ View mentioned gear
Forum Jump
Forum Jump