[klh]

I have significant room treatments in my HT. Here's a thread on that project...

http://www.audiocircle.com/index.php?topic=32017.0

Here is the latest plot of my in room response...




I only used 6 filters on my BFD parametric EQ. The amp is at full gain to try and keep the massive null at 44 Hz above the target curve. BTW, I specifically have my dialed in response above the target curve as I like to have the bass a little hot (house curve). I may cut out the boost around 20Hz, but there's not much there so I'll see how the amp and woofers take it... we'll see.

What do you think would cause the massive null? I have bass trap behind the seated position... I wonder if that's playing a role. BTW, the room is rectangular and fully isolated... 20' x 16' x 9' and the seated position right now is 40" from the rear wall.

Thanks in advance.


Edit: As you can see in the link, the screen is on the long wall and the seated position is at the midpoint left to right.

[klh]

I found Ethan's mode calculator and my room's dimensions doen't support the null at 44 Hz. It must be related to sub location and seated location. The sub is an infinte baffle located in the ceiling adjacent to the front wall at the midpoint of the room (see photo at left... it's above the screen). Obviously I cannot move the sub, so I have to adjust seating location and rear wall treatments.

Any suggestions?

[youngho]

A null due to a cancellation from a standing wave is extremely position dependent, so the effects would vary with moving the microphone from its current position. Also, looking at the construction of the bass trap behind you, it doesn't seem as though it would be tuned to a specific frequency, but it might be reasonable to repeat the measurements after removing it. The calculator is just a tool to start with. Wall construction other than concrete will often shift the primary axial mode frequencies downward somewhat. There are other kinds of modes than axial, too.

Would you mind moving the microphone (say, 2' forward from the current position the first time, 2' backwards from the current position the second time) and repeating the measurements? If the null doesn't change significantly, try removing the bass trap and repeating the measurements in the current position?

[darrenyeats]

Youngho's advice seems good to me.

When doing EQ by all means measure and then dial in the calculated amount of correction, including nulls. I have found, just as in your case, that nulls aren't fully correctable by EQ. What I do is to use the correction I calculate and not any more. Pushing it further where it doesn't want to go will probably cause more problems than it solves. For example, you can lose dynamic headroom for little or no compensating advantage.
Darren

[bpape]

Seating position could be farther away from the wall to smooth things further with less EQ but I don't think that's going to deal with the 44Hz.  My initial thought is that the 44Hz isn't a null, it's that closeness to the wall is giving you large broad peaks.  In that case, moving would smooth things.  You could also deal with it by bringing down the broad peaks to the level of the 44Hz.  As has been stated, EQing nulls isn't usually a good idea as the cancelling waves will both be increased by the same amount.

Bryan

[Ethan Winer]

I found Ethan's mode calculator and my room's dimensions doen't support the null at 44 Hz. It must be related to sub location and seated location.

The worst nulls are often due to the wall behind the listening position. This Frequency-Distance calculator is probably more appropriate for you than my ModeCalc program:

http://www.realtraps.com/sbirlbir.htm

As youngho said you can move the microphone forward and back a foot or two to see if the null frequency changes. If it does, that means you need more bass trapping on the rear wall.

--Ethan

[ctviggen]

Nulls are usually much steeper and sharper, like the one at 132.2 Hz.  They should also be easy to define, as if you move the microphone a bit, the null should improve. 

[klh]

Thanks guys. I figured it was probably from the rear wall. The crazy thing is the bass trap behind the seated position is 6' wide, 4' tall and 6" thick... solid 8# mineral wool pulled 12" away from the rear wall. That's a substantial bass trap. Ethan... I checked out your link. A null at 44Hz that is due to boundary issues requires a position 6'5" from a wall. I'm not 6'5" from any wall. My room dimensions have pretty spread out modes so I don't think it is directly because of standing waves... my guess is it has to do with the subwoofer's location and my location. I hope not... I'll try moving the SPL meter tomorrow. There's no way I'm moving the IB sub .

[Ethan Winer]

the bass trap behind the seated position is 6' wide, 4' tall and 6" thick... solid 8# mineral wool pulled 12" away from the rear wall. That's a substantial bass trap ... I'm not 6'5" from any wall.

That's a fabulous bass trap, and you have indeed proven the null is not from that location.

Next I suggest that you play a steady sine wave at 44 Hz, and with an SPL meter find where in the room that frequency is loudest. Use it like a wand and try all corners and boundaries etc. Wherever that frequency is loudest should be a good place for a bass trap. Of course, 44 Hz is very low, so prepare for even more really thick mineral wool.

--Ethan

[klh]

Thanks Ethan. I'll do just that. Since I made a disk of your test tones earlier on that'll be easy .

BTW, I took a reading with the SPL meter pulled 2' forward and the null at 44 Hz was still there (and bigger!). Not only that, but the low end support was much worse. That got me thinking... if the null is still there when I pull the meter forward but deeper and lower bass support is worse, why not put it back on the couch where my head would be and push the seat back closer to the wall. Maybe the opposite will happen (the null would be more shallow and  the lower end support would be better). Look at the result:

Here's the old:




Here's the new:



It worked like a charm . The low end bass is incredible. No boosting with parametric EQ is required despite the house curve having the low end frequencies 8dB hot (pink noise was 75dB). And, not only that, but I'm only using 5 filters. This is pretty exciting stuff!

[bpape]

All in taste I guess.  You traded a slightly smaller null for a larger peak and overall less smooth response. 

Try putting it back where it was initially and shifting 6-8" off to one side and see what it does.

Bryan

[klh]

Bryan,

I'll take the second plot over the first any day. The peak may be a couple dB higher relative to my house curve, but there is much better low end support, the null is pulled far away from my curve, I got to drop from 6 to 5 filters in the EQ and all I need to do is cut response (no boosting). That last point is really important to me because I worry about bottoming my subs. Having this response with less stress on the system give me less stress, and thus I get to enjoy the movies more . It is less than ideal, but to me it's better than what I had before. Having said all of that, I'll take your advice and try moving a bit off center. If the plot looks better (and more importantly it sounds better), then two people will benefit from the EQ and not just one (provided the measured location is 1' or more off center). That would be perfect since my wife and I could enjoy the better playback together!

Thanks for the suggestion...

Krister

[bpape]

Agreed on doing cuts only.  That's alwasy what I try to shoot for.  As for low end support, you can get that by moving next to any room boundary - that's why we generally stay away from them.  They artificially boost the bottom end impression.  If you need more bottom end boost, try getting the boundary reinforcement by sub placment nearer a boundary as opposed to your seat.  For HT viewing, being farther from the rear wall will also give you the added benefit of a better surround field.

Just a thought.  I hope the side to side works out for you. 

Bryan

[klh]

Part of the fun is the experimentation! And you're correct, there's more to the enjoyment of music and movies than an even frequency response (ambiance, etc.). Can't forget the forrest for the trees. We'll see what happens .

[klh]

BTW, my seating distance from the rear wall was greater than I thought. Previously it was 60" and now it's 42". 3.5' from the rear wall is still a fair amount... although the ambiance is probably diminished a bit, I doubt it's completely gone. I didn't get a chance to put any movies on last night as it was pretty late, so hopefully I'll be able to run through some demos tonight .

[klh]

I did Ethan's mode calc again and it correlated very well with the last plot. Peaks at 28 and 35 Hz are close enough to be confluent. Then there's a long break until confluent peaks at 56, 63 and 71 Hz. Accordingly there is a trough between the two peaks that goes down to the same level as the low end. Then there is another peak that coincides with the predicted peaks at 85 and 105 and 113 Hz. Imagine how high the peaks would be without all the bass traps!

[klh]

Obviously I would only have to deal with three relatively small peaks if I turned down the gain on the sub for a flat response. But, what fun is being flat to 15Hz (actually it's flat to 10 Hz)? I want more deep bass!