How to interpret REW graphs

Love4sound

Well-Known Member
Joined
Nov 23, 2018
Messages
4,679
Points
113
Location
Chennai
I guess it’s safe to assume a few members are good with rew and we have a handful number of rew users. Most discussions on rew we only look or discuss about SPL graph. However other graphs like waterfall,decay and spectrogram isn’t discussed much. If any clear article available please share. I generated them but have no idea how to interpret them. If any one is aware please explain. I have shared all 3 for my front and rear subs. How bad is it? Treatable bad or untreatable bad?
 

Attachments

  • 8E575F57-6A2C-4B01-8316-291CC9311EC5.jpeg
    8E575F57-6A2C-4B01-8316-291CC9311EC5.jpeg
    113.7 KB · Views: 37
  • 401FD3B3-28B5-4E59-8C9C-8A18BBEED699.jpeg
    401FD3B3-28B5-4E59-8C9C-8A18BBEED699.jpeg
    90.8 KB · Views: 33
  • 957289BC-03BF-4EE2-BB47-0BF43B0AA716.jpeg
    957289BC-03BF-4EE2-BB47-0BF43B0AA716.jpeg
    102.5 KB · Views: 33
  • B3B19D4E-21CC-499B-8CA6-3C08D1244FBB.jpeg
    B3B19D4E-21CC-499B-8CA6-3C08D1244FBB.jpeg
    114 KB · Views: 32
  • DC9AD623-DC8C-4F3B-9555-1701621D6CF4.jpeg
    DC9AD623-DC8C-4F3B-9555-1701621D6CF4.jpeg
    94.3 KB · Views: 32
  • FE51C34E-B3C5-4060-832B-A391A67622DF.jpeg
    FE51C34E-B3C5-4060-832B-A391A67622DF.jpeg
    101.6 KB · Views: 33
Please post spectrogram of the combined response of your subs at the MLP. That's what you hear/feel. Effective passive room treatment is going to be really hard down in this range because of the large wavelengths involved. Your best bet is to manipulate delays to get rid of nulls and use EQ to tame the peaks.
 
Please post spectrogram of the combined response of your subs at the MLP. That's what you hear/feel. Effective passive room treatment is going to be really hard down in this range because of the large wavelengths involved. Your best bet is to manipulate delays to get rid of nulls and use EQ to tame the peaks.
Please find the combined graphs.This is not with EQ. I only understand we use these graphs to analyse the ringing and decay but no clue on how to interpret the graph.
 

Attachments

  • Dual waterfall.jpg
    Dual waterfall.jpg
    226 KB · Views: 38
  • Dual spectrogram.jpg
    Dual spectrogram.jpg
    211 KB · Views: 40
  • Dual delay.jpg
    Dual delay.jpg
    272.2 KB · Views: 38
These graphs are after eq and audyssey
 

Attachments

  • EQ Decay.jpg
    EQ Decay.jpg
    276.2 KB · Views: 41
  • EQ spectrogram.jpg
    EQ spectrogram.jpg
    225.9 KB · Views: 42
  • EQ waterfall.jpg
    EQ waterfall.jpg
    198.6 KB · Views: 46
You have a fairly problematic peak in the mid-late 40 Hz range. There's a couple in the mid-teens and mid-20's as well that really ought to have been EQ'd out. It also looks like you've got a fairly wide null past 70 Hz. Can you post your combined FR, with and without EQ?
 
You have a fairly problematic peak in the mid-late 40 Hz range. There's a couple in the mid-teens and mid-20's as well that really ought to have been EQ'd out. It also looks like you've got a fairly wide null past 70 Hz. Can you post your combined FR, with and without EQ?
Sure will post them later. Currently I am dealing with some other issues which I need to rectify
I am assuming the 48 Hz peak coincides with one of your room modes. Is that correct?
I am not sure how to interpret those graphs so unable to answer your question
 
Last edited:
I guess it’s safe to assume a few members are good with rew and we have a handful number of rew users. Most discussions on rew we only look or discuss about SPL graph. However other graphs like waterfall,decay and spectrogram isn’t discussed much. If any clear article available please share. I generated them but have no idea how to interpret them. If any one is aware please explain. I have shared all 3 for my front and rear subs. How bad is it? Treatable bad or untreatable bad?
From what I understand is Waterfall is a 3d representation of decay time for the frequency range measured. Using this ,one can able to find out the standing waves problem and it's decay time ,and can able to take a call on its treatment ( either passive or active) depending upon the frequency range.
Decay is 2d representation.
Spectogram is the plan view of decay time for frequency measured.
 
You have a fairly problematic peak in the mid-late 40 Hz range. There's a couple in the mid-teens and mid-20's as well that really ought to have been EQ'd out. It also looks like you've got a fairly wide null past 70 Hz. Can you post your combined FR, with and without EQ?
First graph has response for subs without eq. Red is combined, yellow is rear sub. 2nd graph is Post eq and audyssey.
From what I understand is Waterfall is a 3d representation of decay time for the frequency range measured. Using this ,one can able to find out the standing waves problem and it's decay time ,and can able to take a call on its treatment ( either passive or active) depending upon the frequency range.
Decay is 2d representation.
Spectogram is the plan view of decay time for frequency measured.
yes I understood the part where waterfall is the 3d representation for decay time, Decay graph is the 2d representation. Just not sure how to interpret it and what is considered to be normal
 

Attachments

  • 93791C9A-3850-46FF-9C54-86D69075A1BC.jpeg
    93791C9A-3850-46FF-9C54-86D69075A1BC.jpeg
    124.6 KB · Views: 26
  • B0626803-D509-46C0-B615-B9CCB733C0C9.jpeg
    B0626803-D509-46C0-B615-B9CCB733C0C9.jpeg
    110.3 KB · Views: 25
First graph has response for subs without eq. Red is combined, yellow is rear sub. 2nd graph is Post eq and audyssey.

yes I understood the part where waterfall is the 3d representation for decay time, Decay graph is the 2d representation. Just not sure how to interpret it and what is considered to be normal
What I heard is if the decay time is between 300ms to 500ms ,the bass will be superb.
Practically it's difficult to do passive correction via Treatment for frequency below 80 hz .You may need to tame it using active correction.
The important thing is planning of room , placement of speakers ,passive treatment and finally active calibration.
 
I am not sure how to interpret those graphs so unable to answer your question
USe the room mode calculator using this https://amcoustics.com/tools/amroc and check the room modes that you have. Most likely, you will find one of the dimension having this frequency. Which also means with two subs positioned properly along the axis you could cancel them.
 
USe the room mode calculator using this https://amcoustics.com/tools/amroc and check the room modes that you have. Most likely, you will find one of the dimension having this frequency. Which also means with two subs positioned properly along the axis you could cancel them.
Sure will check this out later. I am now caught with another issue with one of my sub and waiting for a resolution. Once done I will get back to alignment and integration

 
You have a fairly problematic peak in the mid-late 40 Hz range. There's a couple in the mid-teens and mid-20's as well that really ought to have been EQ'd out. It also looks like you've got a fairly wide null past 70 Hz. Can you post your combined FR, with and without EQ?
The Identical subs with different frequency response issue is resolved. Got my sub replaced and turns out one unit was faulty. Please find the requested graphs. Blue is the combined response of both subs. Black is after EQ using mini DSP. Yet to run audyssey.
 

Attachments

  • Eq Graph.jpg
    Eq Graph.jpg
    282.6 KB · Views: 43
That works too and is apparently what the OP did. AIUI, Audyssey’s approach to integrating multiple subs is questionable. So, any integration work prior to presenting Audyssey with a combined response is preferred.
Guess in my case audyssey is doing a good job with dual sub integration. The response is almost identical. I connected both subs directly to the avr removing mini dsp from the chain. In the 1st image blue graph is the sub response after AVR calibration and the red is using mini dsp and post audyssey. Now my problem is integration. For example if i set a sub distance and integrate it with mains the same value is having phase issues for center. This issue i am having with both AVR and mini dsp. In the second image the purple graph is mains crossed over at 90 and orange is center crossed over at 100 for the same sub distances set.
 

Attachments

  • AVR duals vs Mini DSP.jpg
    AVR duals vs Mini DSP.jpg
    259.6 KB · Views: 30
  • LCR.jpg
    LCR.jpg
    262.8 KB · Views: 29
The Identical subs with different frequency response issue is resolved. Got my sub replaced and turns out one unit was faulty. Please find the requested graphs. Blue is the combined response of both subs. Black is after EQ using mini DSP. Yet to run audyssey.
Hey ....are you not loosing lot of energy ?? Just for eliminating the nulls it looks like you reduced the overall SPL. May be you can target the graph 5-6 db more and still live with some minor nulls here and there ...
 
Get the Award Winning Diamond 12.3 Floorstanding Speakers on Special Offer
Back
Top