Help on figuring out what is going on.

[dustinb]

I've got my IB up and running. Tried to EQ it. Took a while but got Room EQ Wizard up and running nicely. Then took my first sweep. Nasty dip at 44hz. Thought it might be because I didn't have my riser closed in yet and I had a nice 10"x4.5'x11.5' space full of insulation vacuuming up that bass. So I closed it off and ran the sweep again. Same result. Ran one of the two subs, same response 6dB quieter. Ran the other same thing. Ran the sweep with the SPL meter up close to the sub no big dip.

Images have now been edited per Thomas' request and I've added a few more varying the EP2500 filter settings.

Anyways here's the response:



Then out of curiosity I decided to run the sweep again with the access door in my screen wall open and got this:



Both these two were with the EP2500 switches all set to the labelled off position.

Here is with the LP set to on and the 30hz position:



On and 50hz position:




Off again but in the 30hz position:


And finally with the spl meter in the manifold and equadistant from the two drivers:

[ThomasW]

You need to read the REQW pages where they tell you how to standardize the graphs. Also might add in some smoothing and lower the upper measurement point to 100Hz.

Then of course make a nearfield measurement to compare against the listening position.

Did you close off the back of the box?

[dustinb]

For standardizing are you talking about this?

www.hometheatershack.com/forums/bfd-rew-forum/36-please-read-getting-graphs-ready-post.html

Yes I've blocked off the back of the box. 5/8" ply with 1/2" osb on top. 3/4" ply cleats, it all screwed together and then siliconed tight.

What level of smoothing would you recommend be used for bass measurements?

[ThomasW]

dustinb said:

For standardizing are you talking about this?

Yes

What level of smoothing would you recommend be used for bass measurements?

Look at both no smoothing and 1/3rd octave smoothing

In addition there's something fundamentally wrong with the measurement technique or equipment, if the first plot you posted is a true nearfield measurement (taken 1" from the dust cap of one driver) .

Ran one of the two subs

Doing that creates a BIG problem. The unpowered driver one is acting as a bass trap. Both need to be operating when any measurment is being done

[dustinb]

The first plot isn't near field.

dustin.bunnyhug.net/pics/forum/theater/screen%20wall2.jpg

My screen wall has the IB opening on the right in this picture. The other opening is the access door. I have that door sealing with the same closed cell foam weather stripping as the manifold. The first plot is at the listening position with the screen wall access door closed, the second plot is at the listening position with it open. I thought I'd see the response change with some sort of cancelation happening.

When I ran one driver at a time, I saw the exact same response, just 6dB quieter. I did this just to see if there was something wrong with one driver or one drivers response in the room.

Should the clipping limiter dip switches on the EP2500 be on?

I'm going to take some more measurements now. I'll replace the above graphs with new ones in a few minutes.

[dustinb]

Looks to me from the graphs in the first post now that the labelled off position for the LP filter is wrong. I've got it set in what says the on position now.

I'm going to go play with the filters some now while I wait for a response.

[dustinb]

Well this definitely isn't what I wanted

Here's center first row.



First row right side.



Riser right side.



Riser center.

[ThomasW]

This is the second time someone has done this.....that being build an IB and fire it at a hole in the middle of the floor, and not make any mention of this in advance of the project.

Is there attic space available at the back of the room? If so consider buying 2 more drivers and put a ceiling mounted IB there. It's possible doing this will even out the in-room response.

You can test this by building a sealed test box. Remove one of your drivers and seal off the hole. Put that driver in the test box and set it at the back of the room. Now run both subs and remeasure the response. If it smooths out the null, you know what to do.

FWIW were you to remove the 1/2 walls and replace then with a railing supported by open ballasters you'd have better overall sound in the room.

I find it amazing that no one on HTT pointed out this problem months ago when you were posting there about the design.