Jump to content

Recommended Posts

Posted (edited)

I took a lot of different measurements to get an idea of my different combinations, but found a strange problem that is probably caused by my room, but not sure exactly what is wrong.

From online room acoustics videos it sounds like it might be a problem with the wall behind the speakers, but don't know enough to be sure.

 

There are several strange things about my responses:

1 - They don't change much between left and right, microphone position, speaker distance from side walls, or volume.

2 - The bass and treble are much smoother than the mid range, yet I can easily spot treble and bass differences (especially treble), but the mid range sounds fine to me.

 

Does anyone know what could be causing this, and whether this is in fact reflections from the wall behind the speakers?

 

All pictures are from REW with 1/6 smoothing, except for the one with 1/3 smoothing. Measurements were done with a MacBook Pro, Chord 2Qute DAC, Behringer UMC202HD interface, and Behringer EMC8000 mic.

 

All combinations in 1/6 then in 1/3 smoothing:

all_combinations.thumb.jpg.f803a6f2a016b538a434b5f6cc466cd8.jpgall_combinations_3_smooth.thumb.jpg.7a9387f0cd3e77cfff133356ccc44dfa.jpg

 

Left vs right vs both speakers:

cyrus_pmc_left_right.thumb.jpg.7c03d63049e18ae2b4e47e6a4bc4e46d.jpgcyrus_spendor_left_right.thumb.jpg.009b3d0a8ad9c0bcfffd32a20a0f68c5.jpgcyrus_vienna_left_right.thumb.jpg.cc2616ed249cdb31b666d2ffa8c45d39.jpg

 

SPL Difference:

vienna_spl_difference.thumb.jpg.bb9a252df1649f99999829f2fbb2d351.jpg

 

Mic positions; back of couch, front of couch, in front of couch, half way between couch and speakers, 40cm from left speaker:

dynaudio_positions.thumb.jpg.79b639c801374cb88302b1b1fcada850.jpg

 

With and without screen behind and sides of microphone, 40cm from left speaker:

dynaudio_screen.thumb.jpg.c162cbe54d8c093e891c4775eff22798.jpg

 

Short vs long sweep:

dynaudio_short_long_sweep.thumb.jpg.bed4f649f40fa3116411cd43dd09a58c.jpg

 

Edited by audio_file

Share this post


Link to post
Share on other sites

The short vs long sweep makes the biggest difference, which suggests to me that delayed reflections from lower frequencies are affecting the frequencies above them. (because the sweep goes from low to high)

Share this post


Link to post
Share on other sites

How far are the speakers from the front/side walls?

 

I believe this has something to do with an in-room reflection causing an opposing phase thereby cancelling out that specific frequency.

Share this post


Link to post
Share on other sites

I tried several distances from side walls without any difference, but the distance from the wall behind them remained constant, so it could be that one.

 

I'd have to measure the exact distances, but I think the side distance ranged from 60cm to 1m, and the back distance from the front of the speaker is around 70cm. However I might be off as I'm not home to measure right now.

Share this post


Link to post
Share on other sites
17 minutes ago, nonsynchro said:

How much freedom do you have in moving things around?

I'm using the Cardas horizontal room formula and having great results.

http://noaudiophile.com/speakercalc/

http://www.cardas.com/room_setup_main.php

Thanks, that looks interesting. I could move them further from the back wall but would lose floor space for other things.

I use something similar to cardas but closer to the wall behind them.

Share this post


Link to post
Share on other sites


48 minutes ago, audio_file said:

Thanks, that looks interesting. I could move them further from the back wall but would lose floor space for other things.

I use something similar to cardas but closer to the wall behind them.

I'm fairly sure given the highish frequency being cancelled out, it may be that they are too far from the wall in your room, I would experiment by nudging them back a few cm at a time and see if the dip lowers in frequency.

 

Share this post


Link to post
Share on other sites
15 hours ago, audio_file said:

Left vs right vs both speakers:

When you take a measurement of two speaker playing together, you will see peaks and dips in the resulting summed response which are directly caused by the interaction between the two speakers  (eg. they are not in phase alignment, or have a amplitude response).

 

What you need to do, is unless you have a very specific reason to do so (which you almost never will) is do not ever measure two speaker playing together.   Only measure the L, or the R, never both.

Share this post


Link to post
Share on other sites

500Hz is high enough to be treated by absorption panels at the back.

Would it make any difference to put these behind the speaker, or is it more likely the middle part of the wall between the speakers that would cause the problem?

 

Share this post


Link to post
Share on other sites
15 hours ago, audio_file said:

probably caused by my room

You need to go back to basics, before you can be sure of the data you have.

 

Move the speaker (just one speaker) into the 'middle' of the room.... away from as many things as you can.

Measure the speaker 1m away, directly in front, at tweeter level.    Ensure that only this one speaker is playing.    Leave all other REW setting at default.

 

Describe your room size (the important information is the rough distances of walls from your speaker and microphone) ... knowing how far the surfaces are away, will tell us if there is another step needed before you can interpret you data (the step may be "gating" or "windowing" the data so that the reflections are somewhat excluded).

 

This will give you a baseline, from which you can then start looking at the question of "what your room does".

Share this post


Link to post
Share on other sites
7 minutes ago, davewantsmoore said:

When you take a measurement of two speaker playing together, you will see peaks and dips in the resulting summed response which are directly caused by the interaction between the two speakers  (eg. they are not in phase alignment, or have a amplitude response).

 

What you need to do, is unless you have a very specific reason to do so (which you almost never will) is do not ever measure two speaker playing together.   Only measure the L, or the R, never both.

I might choose one speaker next time I measure, but in this case left right and both all had the same pattern and problems.

 

The three graphs for left/right are as follows:

Graph 1:

PMC Twenty.23 left, right, both.

 

Graph 2:

Spendor SP2/3R2 left, right, both.

 

Graph 3:

Vienna Acoustics Beethoven Baby Grand SE left, right, both.

 

The left/right/both differ slightly, but are almost the same.

 

Share this post


Link to post
Share on other sites

 

17 minutes ago, davewantsmoore said:

never both

5 minutes ago, audio_file said:

I might choose one speaker next time

Please consider that never is a strong word  ;)    I chose it quite carefully.   Don't.  Ever.  Do.  It.

 

11 minutes ago, audio_file said:

in this case left right and both all had the same pattern and problems

We can't reliably determine what causes it, without narrowing down the possible causes.

 

11 minutes ago, audio_file said:

The left/right/both differ slightly, but are almost the same.

The next question is about how far surfaces are from the speaker.    Eg. side walls.   There are big cancellations (strong dips) in your measurements, which aren't normal.     Something significant is causing them.  Move one speaker like I suggested and measure with the speaker and mic as far a way from any surfaces as you can.

 

Other suspects.   Is there a mic or soundcard calibration loaded (what is it?) ..... how is you measurement system hooked up?   Do you have the output via your ChordDAC and the input via the Behringer?      Assuming the behringer has inputs and output, you should try using just the behringer (output and input) to check for issues.

Share this post


Link to post
Share on other sites


1 minute ago, davewantsmoore said:

We can't reliably determine what causes it, without narrowing down the possible causes.

Yes, just mentioned that it didn't change much. When for example I change polarity on one speaker I can see a clear difference between the two.

 

2 minutes ago, davewantsmoore said:

The next question is about how far surfaces are from the speaker.    Eg. side walls.   There are big cancellations (strong dips) in your measurements, which aren't normal.     Something significant is causing them.  Move one speaker like I suggested and measure with the speaker and mic as far a way from any surfaces as you can.

 

Other suspects.   Is there a mic or soundcard calibration loaded (what is it?) ..... how is you measurement system hooked up?   Do you have the output via your ChordDAC and the input via the Behringer?      Assuming the behringer has inputs and output, you should try using just the behringer (output and input) to check for issues.

I saw your question in your previous post, I'm just not home right now to measure all the distances. :)

I also have to find time to do another measurement, it's not simple with the time I have available.

 

I calibrated the DAC into the interface and used that in all the measurements. The calibration was quite flat, except for a slight roll off on the extreme treble and bass if I remember correctly.

 

I don't think I can use the interface both as input and output at the same time, but I'll double check.

 

The mic pointed straight up in all measurements.

Share this post


Link to post
Share on other sites
1 hour ago, audio_file said:

I'm just not home right now...with the time I have available

Sure, I wasn't expecting them right now ;) 

1 hour ago, audio_file said:

The mic pointed straight up in all measurements.

Point the ECM8000 at the speaker, as opposed to straight up (but it shouldn't make a dramatic difference).

Share this post


Link to post
Share on other sites

If you look at a frequency response with no smoothing, for say 20 to 1,000Hz range, you may find the dip is very narrow. If that is the case you may not hear its effects.

Share this post


Link to post
Share on other sites
1 minute ago, Hipper said:

If you look at a frequency response with no smoothing, for say 20 to 1,000Hz range, you may find the dip is very narrow. If that is the case you may not hear its effects.

The dips are all relatively narrow, and will be less audible than they look.... but still quite audible in general (they won't sound like you might expect a "dip" to sound like).

 

The real question is what are they? .... as depending on what they are, they can likely be addressed.    They are all ~too high in frequency to be due to anything big/far, assuming that only one speaker is playing, and the room is not comically small.

 

I would be concerned about the cause.... and wouldn't be surprised if there was some sort of measurement issue, or strange setup issue/problem which has gone unnoticed, or hasn't been explained  (hence all the questions).

Share this post


Link to post
Share on other sites
11 minutes ago, davewantsmoore said:

the room is not comically small.

It's not. :)

Share this post


Link to post
Share on other sites

need more info on your speakers, but if a single speaker has the same dip regardless of position (mike or speaker), then I'd suspect a crossover issue (especially if both speakers measured individually have the same dip) - eg crossover is set at 500Hz, but 1 driver has the wrong polarity.

For example 2nd order Butterworth crossover filters require 1 driver connected out of phase otherwise you get a big null.

 

cheers

mike

Share this post


Link to post
Share on other sites


try a close miked measurement inline with the tweeter, or if you can, drag a speaker outside and measure at 1m

 

mike

Share this post


Link to post
Share on other sites
2 minutes ago, almikel said:

need more info on your speakers, but if a single speaker has the same dip regardless of position (mike or speaker), then I'd suspect a crossover issue (especially if both speakers measured individually have the same dip) - eg crossover is set at 500Hz, but 1 driver has the wrong polarity.

For example 2nd order Butterworth crossover filters require 1 driver connected out of phase otherwise you get a big null.

 

cheers

mike

PMC Twenty.23:

Two way

8 ohm

87dB at 1 watt

1.8 kHz crossover point

 

ATC SCM 19:

Two way

8 ohm

85 dB at 1 watt

2.5 kHz crossover point

 

Dynaudio Contour S1.4 LE:

Two way

4 ohm

85 dB at 1 watt

1.9kHz crossover point

 

Vienna Acoustics Beethoven Baby Grand SE:

Three way

4 ohm

90 dB at 1 watt

Unknown crossover points

 

Spendor SP2/3R2:

Two way

8 ohm

88 dB at 1 watt

3.2kHz crossover point

 

Share this post


Link to post
Share on other sites
8 minutes ago, almikel said:

try a close miked measurement inline with the tweeter, or if you can, drag a speaker outside and measure at 1m

 

mike

I did at 40cm.

Share this post


Link to post
Share on other sites
1 hour ago, audio_file said:

I did at 40cm.

so all those different speakers had the same dip at 500Hz? with the mike in the same position?...would have to be SBIR - too high for modal - but I would have expected the frequency to change with mike/speaker position if SBIR???

Too tricky for me...

Share this post


Link to post
Share on other sites
53 minutes ago, almikel said:

so all those different speakers had the same dip at 500Hz? with the mike in the same position?...would have to be SBIR - too high for modal - but I would have expected the frequency to change with mike/speaker position if SBIR???

Too tricky for me...

The 500Hz dip can be seen in all my graphs.

I showed the graph with different mic positions.

The speakers were at different distances from the side wall, but all about the same from the wall behind them.

That's why I thought it's likely the wall behind them.

Share this post


Link to post
Share on other sites
Posted (edited)

I tried to make a diagram of my lounge as best I can.

The racks are on the side and the speaker positions can be seen in red.

The distances from the side wall (from the tweeter) were: 61 mm, 90 mm, and 112 mm.

The distances from the back wall (from the tweeter) were: 70 mm, 65 mm, 64 mm.

Note that the one speaker is not blocked by the couch, it is on a stand that is higher than the arm rest.

 

lounge.png.f46d42ad9afe83b0e957aedb5886b563.png

Edited by audio_file

Share this post


Link to post
Share on other sites


They are fairly close already.

I'm fairly sure given the highish frequency being cancelled out, it may be that they are too far from the wall in your room, I would experiment by nudging them back a few cm at a time and see if the dip lowers in frequency.
 

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

  • Recently Browsing   0 members

    No registered users viewing this page.

×