Cancellation between main speakers

[MediumRare]

Folks, these nulls are exactly what speaker positioning is about. Here's a Room Sim grab from REW. Try it. See the dips in bass? Very specific frequencies.Move your sub or listening position around and see how the graph changes. The graphs are not real-world, only simulation, but that's why you measure your own room.

 

[DanDan]

It is a deep and wide cancellation. Fascinating in a way. I bought a second hand pair of Spendor BC1 back in the day. I checked them for phase in the studio CR at some point. Confused results. One of the woofers had been repaired and the replacement had physically reversed + and - I guess you could try flipping the wires at the terminals and see how this null reacts. Might be another clue to the puzzle. Also could you post the REW files, sometimes we can read something in the tea-leaves there.

 

[JLM1948]

Folks, these nulls are exactly what speaker positioning is about. Here's a Room Sim grab from REW. Try it. See the dips in bass? Very specific frequencies.Move your sub or listening position around and see how the graph changes. The graphs are not real-world, only simulation, but that's why you measure your own room.

Nobody doubts the existence of resonances and antiresonances. The subject is how two speakers that deliver a relativelty flat response individually can result in a significant dip when used together.

 

[mollie]

Check first one by one the right and left speaker at one fixed position wether they give moreless the same output level.
Only small variations between the two are alouwed!!! If they are not moreless the same ... check connections for fase !
If they are moreless the same: measure now the left - and later the right speaker at there normal listening position.
Make sure that the measuring mic position is now exact at the same distance to both speakers and measure now.L+R.
If the above measurements are okay, and your problem around 100 Hz is still there, than the reason will be the room which is enlarged on the left compared to the right speaker. Hopefully this will help to clarify the problem. Succes. Mollie

 

[travman83]

Lots of helpful comments here, I appreciate it. Based on some recommendations, I moved the mic forwards and backwards and did notice a shift in the null. I ended up moving my seating position back a bit, which brought the null down enough to where I was able to address the issue when adding my subs back in the mix with an 80 hz crossover. I am going to play around with the time alignment feature in REW as well to see what implementing delays gets me. I would still like to have a solid 2.0 option, but I’m not sure that is realistic in my room.

 

[MediumRare]

Nobody doubts the existence of resonances and antiresonances. The subject is how two speakers that deliver a relativelty flat response individually can result in a significant dip when used together.

What you are calling an "antiresonance" is an "antinode". That is the term for a trough caused by the incidence of two sine waves at the bottom of their cycle. That's also called a cancelation. https://en.wikipedia.org/wiki/Node_(physics)

 

[MediumRare]

Lots of helpful comments here, I appreciate it. Based on some recommendations, I moved the mic forwards and backwards and did notice a shift in the null. I ended up moving my seating position back a bit, which brought the null down enough to where I was able to address the issue when adding my subs back in the mix with an 80 hz crossover. I am going to play around with the time alignment feature in REW as well to see what implementing delays gets me. I would still like to have a solid 2.0 option, but I’m not sure that is realistic in my room.

Unlikely you have a time alignment issue. Sound travels 1.1 feet per millisecond. The speakers (and sub) would have to be distanced differently more than 10 to 15 feet before time could even possibly be a factor. Even then you could perceive it only under unrealistic lab conditions.

Play with your crossover point and phase of the sub. You might find 100 hz crossover gives good fill-in, especially with some phase shift.

 

[JLM1948]

What you are calling an "antiresonance" is an "antinode". That is the term for a trough caused by the incidence of two sine waves at the bottom of their cycle. That's also called a cancelation. https://en.wikipedia.org/wiki/Node_(physics)

In my native language, they are called "noeuds et ventres" (knots and bellies).
I guess no one had difficulies understanding what I meant.

 

[wahoospiff]

I am dealing with nearly the exact same issues with my room and my system. Getting massive bass cancellation. I also have a room opening on the left side of my room (entry foyer) as well as some other asymmetries in my large room. I tried measuring the left and right speaker phase separately in REW and exporting that data to Excel. In Excel, I computed the left minus right speaker phase and plotted it vs the REW measured SPL when both left and right speakers were playing. My "cancelation" range of frequencies had close to 180 phase angle differences between left and right speaker. It's driving me nuts. I intend to post in the near future. I also tried DSP with audiolense and that actually made the cancellation worse - it spanned an even larger frequency range.

 

[JLM1948]

The first thing to do is check that the system sounds better after reversing polarity of one speaker.

 

[JStewart]

I am dealing with nearly the exact same issues with my room and my system. Getting massive bass cancellation. I also have a room opening on the left side of my room (entry foyer) as well as some other asymmetries in my large room. I tried measuring the left and right speaker phase separately in REW and exporting that data to Excel. In Excel, I computed the left minus right speaker phase and plotted it vs the REW measured SPL when both left and right speakers were playing. My "cancelation" range of frequencies had close to 180 phase angle differences between left and right speaker. It's driving me nuts. I intend to post in the near future. I also tried DSP with audiolense and that actually made the cancellation worse - it spanned an even larger frequency range.

REW measurements taken with timing reference?
Could you post .mdat? Just curious.
Ive had the same problem or similar before. Woofers were not in phase. Mids and tweeters were fine. Liked the huge soundstage though.

 

[wahoospiff]

.mdat file attached as well as a couple plots. Figure smith1_SPL.jpg shows REW SPL for left only, right only, and both left and right speakers playing. Notice frequency regions where combined SPL is less than left or right only SPL indicating cancellation occurring. Figure smith_LmR_phase.pdf was creating by exporting phase measurements from REW into Excel. I calculated left minus right speaker phase angle and plotted vs frequency. When this value is near 180 degrees, phase cancellation occurs. I also displayed SPL curve but no SPL values listed - it was just much easier to plot the SPL curve that way in Excel. I definitely have some room modes near 50, 65, and 100 hz (maybe SIBO too) but it is interesting that the phase angle difference between speakers is high and

close to 180 degrees near those frequencies too.

 

[DanDan]

Interesting. Reiterating JStewart's astute comment:- REW measurements taken with timing reference?

 

[wahoospiff]

Yes REW measurements used timing reference. Thanks for any assistance.

 

[DanDan]

You are welcome. It has been a while, a very welcome heatwave here..... So assistance? Do you have a specific question?

 

[wahoospiff]

My questions would be:
-Is my measurement with REW and interpretation, i.e., that I am getting a lot of phase cancellation between left and right speakers, correct?
-Is there anything I can do about it? I tried DSP with audiolense and it actually made the cancellation worse. My room treatment options are very limited since my listening room is our living room

 

[DanDan]

The dips are caused by acoustic reflections. Acoustic treatment is the only way to address those. Peaks can be softened by Eq to some extent. Studio Control Rooms often sink the speakers into the Front Wall. No reflections. Next best thing may be placing the speakers as close as possible to the Front Wall.

 

[JStewart]

-Is my measurement with REW and interpretation, i.e., that I am getting a lot of phase cancellation between left and right speakers, correct?

Some. the speakers are not in phase around 45Hz and phase is a bit messy in the 100Hz to 200Hz range.
Here's what it looks like with the REW alignment tool calculating Left + Right, which by the way matches your L+R measurement pretty closely. (The absolute SPL is not the same leaving me to wonder if the volume was changed when any of the measurements were made.)

The output is good to 20Hz. Are these mains only or is there a subwoofer in play?

If you look at the spectrograms you can see the delays are different where the phase is. Same with Group Delay.
This is Left Speaker

This is Right Speaker:

I doubt there are any room treatments you could use for the 45ish Hz issues. Multiple subs and dsp are the likely answer here. And I'd guess you could make some progress for the rest of the spectrum, with some treatments, if you ever had a mind to.

Is there anything I can do about it? I tried DSP with audiolense and it actually made the cancellation worse. My room treatment options are very limited since my listening room is our living room

,

I'm surprised Audiolense didn't correct the phase matching between the right and left as that, and adjusting for group delay, are among its benefits.
Did you ever ask @juicehifi to comment on your Audiolense project?

 

[JStewart]

Should have mentioned above, @Mitchco has an excellent reputation if you ever consider professional help.

https://accuratesound.ca/services.html

 

[DanDan]

A fascinating display of REWs powers in the right hands. But there is an overwhelming presence of room dacay here which makes frequency response, electronic analysis, or electronic mitigations moot.
Acoustic treatment can be quite unobtrusive. e.g. the larger Studiotips SuperChunk occupy often empty corners. Here in my mostly concrete space, 3 SSCs reduced RT from 1.4S to 0.6S.