Big null when measuring both speakers but not in single speaker measurement

[paullambert100]

Hi everyone,

I noticed something odd when doing measurements of my newly treated room. When I measure the speakers by themselves (L and R) and when I measure them both at the same time (L + R) a huge dip in the 6k-9k occurs in the stereo version of the measurement vs the individual L + R speakers. I'm kind of stumped.

A bit of info on the room. It is 11x12x9ft high. I have put wall to ceiling soffit's Bass Traps on all four corner, three 6" monster traps on the front wall, three 4" panels on the side walls, three 4" panels on rear wall, and a 9" 4x8 ceiling cloud. I did alot of research and though I had enough thick panels to avoid a completely dead high end but I'm not sure what might be causing this occurrence. Any advise would be greatly appreciated!

I have included 3 screen captures of the R (in red), L (in blue) and both speakers (in turquoise) and the REQ file aswell.

Secondly, what would be considered normal deviation between speaker responses? Is there a maximum limit in db before things get to chaotic? There a few spots in the measurements where the curves between L + R speaker deviates by around 5-8db. Thank you so much!

 

[sm52]

So it happens when the microphone is not exactly at the same distance from the two speakers. I summed measurements R and L. So should look both together.
However, REW shows a difference of these distances as 0,1 mm. If you did not shift a microphone, then the reason maybe unequal time of passage of a test signal from the right and left channel when they work together.

 

[martins5]

Hi everyone,

I noticed something odd when doing measurements of my newly treated room. When I measure the speakers by themselves (L and R) and when I measure them both at the same time (L + R) a huge dip in the 6k-9k occurs in the stereo version of the measurement vs the individual L + R speakers. I'm kind of stumped.

A bit of info on the room. It is 11x12x9ft high. I have put wall to ceiling soffit's Bass Traps on all four corner, three 6" monster traps on the front wall, three 4" panels on the side walls, three 4" panels on rear wall, and a 9" 4x8 ceiling cloud. I did alot of research and though I had enough thick panels to avoid a completely dead high end but I'm not sure what might be causing this occurrence. Any advise would be greatly appreciated!

I have included 3 screen captures of the R (in red), L (in blue) and both speakers (in turquoise) and the REQ file aswell.

Secondly, what would be considered normal deviation between speaker responses? Is there a maximum limit in db before things get to chaotic? There a few spots in the measurements where the curves between L + R speaker deviates by around 5-8db. Thank you so much!

Activate REW acoustic reference for your measurements to see the MLP delay for your right and left speakers. I'm betting you will find they aren't the same. Adjust AVR speaker distance so they both match.

 

[paullambert100]

Hi thank you for the reply. When you say adjust AVR speaker distance do you mean my monitors? I am using a universal audio x6 mic pre and not sure if this was pointed toward someone with a home theatre set up. I measured the distance of the speakers and put the mic in what I thought was exactly the middle. Thanks!

 

[DanDan]

It is very difficult to place the mic exactly in the middle.

 

[paullambert100]

Hi guys, thanks your assistance on this. I went back and tried again and what I think is a more balanced in getting the mic to center but when I even move slightly left or right (as in 0.5") I get a huge deviation in the nulls etc.. Does this mean that if even move my head slightly left or right when mixing everything is going to be out of whack ? Here is the screen shot, the center position in red, slightly off to left in blue and slightly off to right (brown). I'm basically trying to understand if this is normal behaviour or if I'm doing something wrong. Thanks again for all your help.

 

[sm52]

These are questions of measurement. When the head moves, the pits, as in the graph, will not be heard. The difference as in the graph in post #2 appears with a difference in the distance to the speakers of 27 mm.

 

[paullambert100]

Okay thanks for the clarification! I was freaking out a bit there. So would you suggest forgetting about measuring both speakers at the same time and just focus on individual measurements of L + R or is there some unseen benefit (maybe low end) that is useful when measuring both speakers at the same time? I was just following along the GIK measurement starter video and noticed they took a measurement of both speakers. Thanks!

 

[DanDan]

Measure individual L and R for obvious reasons. Include L+R to analyse the interactions of the LF drivers and the Room, at LF only, say under the Schroeder threshold or say 300Hz. We have two ears. Due to distance it is very unlikely that both will experience that same comb filtering, nulling. Ideally measurement should be made with a Jeklin Disc or such, Binaural with a blocking disc to represent the blocking caused by our heads. The two ear position measurements can be combined in various ways in REW without the cancellation effects that a single point source mic will suffer.