Here's my first graph. I think I need some serious help...

[RazorX]

I just finished messing around with REW for the first time tonight and I'm pretty disappointed in what I am seeing. Am I mistaken? This is pretty bad isn't it?


I'm using a RADIOSHACK 33-2055 meter.

My system consists of 2 Paradigm PW-2100 subs in an 11'x16'x8' room. I have one sub in the left corner of the front wall and the other is in the right corner of the rear wall. I have some sound absorption acoustic panels on the front and side walls. I also have two foam bass traps in each of the corners on the front wall. Here are some pics of the room if that helps.








I set it up at 48KHz/16bit and ran multiple tests. I tried it with only one of the subs on (three passes on each), with both of them on, with different crossover settings on the receiver, different speaker size settings on the receiver, and I even tried moving both subs farther away from the wall. There wasn't much difference no matter what I did.

Any thoughts or suggestions? Am I doing something wrong that you can see? I did have some challenges getting it set up and working. I had to run out and get an external sound card since my laptop only has mic in and it wouldn't work with the built-in sound card.

Then I started receiving a message saying the level was low (-64dB) when I did the level check prior to measuring. I messed around with it for a while, read tons of stuff on this sites, and reread the help files. I couldn't find anything wrong with the configs or the connections but it still read really low. I went ahead and ran the measurement a few times and then it finally started showing -21 - -26 when I ran the level check on some of the last few tests. I have no idea why it changed since i didn't change anything. Could lower battery level in the meter cause this behavior? I don't have a 9v battery around the house so I couldn't try that.

[brucek]

Any thoughts or suggestions?

Looks like a nice theater.

The RadioShack meter should output sufficient level for most any soundcard. Usually we use the meter on its 80dBSPL scale. If the battery shows good, I don't see a problem there.

Hopefully you are using a mono RCA cable from the meter and connecting that to a mono to stereo adapter at the soundcards line-in jack. The intermittent change in level you mentioned may be a result of the adapter not making great contact at the plug. You could play around with that a bit.

Generally we start measuring with sub(s) only, with the mains shut off, and the receiver in stereo mode, with all effects off. This allows you to examine and equalize the subs without the influence of any other speakers.

Two subs are difficult to equalize, but has the advantage of using their positioning as a form of equalization of the overall signal.

Anyway, you have a peak at ~36Hz. You need to equalize it out. Not a big deal. Once that peaks gone, you'll be able to turn up the wholesale level of the sub to match the level of the mains and it will sound much better.

I would start with a sub only measurement and show us that....

brucek

[Otto]

Hi Rich and welcome to the Shack!

In general, your measurement looks like you have things set up properly. Did you use the cal file for your meter, and did you get the loopback test to work (the loopback test will generate your soundcard cal). If you got that stuff working, I think you have a real measurement there!

Now, you're right about it not looking so hot. The first thing I might suggest is turning off your mains for the measurement; it looks like they are still on since the measurement is running up to and beyond 200 Hz. Let's see how those subs look without the mains in the way.

Once you get a measurement of that, I'd probably suggest moving your subs around. I would probably start by simply stacking them together, one on top of the other. I would not confine them strictly to the corner; feel free to move them out somewhat. I'd also experiment with different corners, but since your room appears to be rather rectangular, you might not notice much difference.

Anyway, there are a few suggestions that will keep you busy for a bit. Let's see what comes out of those and go from there.

Have fun.

EDIT: brucek beat me to it!

[RazorX]

Thanks for the replies and suggestions. First, I will try to answer your questions.

This graph was measured with the meter on its 80dBSPL scale and with the correct calibration file I downloaded from here. I am using cables like you mention and I think the problem may have been not getting a good contact initially. My cables are just barely long enough so there is a slight amount of tension when everything is hooked up. You guys are both correct; I did this measurement with the mains on - but you already knew that from looking at the graph didn't you?

I am at work now so I will need to wait until I get off work to play with it some more but I will do another series of tests with the mains off. I will also try moving them around, stacking, etc. and see what I get.

I am also seriously considering the purchase of a Behringer Feedback Destroyer but I'm not sure which to get. The DSP1124P seems like it will fit the bill and is less expensive but I'm not terribly exited at the prospect of needing to get the firmware fix. I don't mind the actual task of replacing the EPROM, I just don't like the idea of waiting for the unit and then waiting for the replacement EPROM.

I have also be considering adding bass traps in one of the rear corners like I have in the two front corners. I can't do both back corners because I have a door in the way on the left rear wall. I do plan to cover the whole front wall with the same acoustic panels I already have on a portion of it to combat some reflections from the surrounds when playing at high levels. I will probably add another row (above and below) to sides walls and extend them to the front wall while I am at it. I guess waiting for the DSP firmware fix isn't that big of a deal since I will need to measure everything again after the additional wall treatment anyway.

Another thing worth mentioning: I have ordered a new sub and will be getting rid of the two PW-2100's when it comes (probably by the end of the month). The replacement sub is an Elemental Designs A5 - 350. I have read some pretty good reports on this sub so I'm hoping it will be an improvement over my current setup. If not, I will sell it and keep what I currently have (or if they all sound really good together - keep them all, Bwaaaaaa ha ha haaaaa).

I do have a question for you guys. I have two ButtKickers mounted to the floor joists of this room that are running off of a ButtKicker power amp. I didn't have it on during my measurements. What is the best way to take tune for them? Should I do an initial measurement, and appropriate corrections, and then bring them online after the "real" subs are corrected and fine tune from there or do one measurement/correction with everything on.

I'll upload a new graph, or graphs, with my findings from the activities tonight.

Thanks again.

[brucek]

I have two ButtKickers mounted to the floor joists of this room that are running off of a ButtKicker power amp. I didn't have it on during my measurements. What is the best way to take tune for them? Should I do an initial measurement, and appropriate corrections,

Huh, never thought about it before. Certainly the the voltage response is 'perfect' before it is applied to the buttkicker load (just as it is for a subwoofer load). The requirement to equalize a subwoofer is a result of that 'perfect' voltage signal being transduced into sound through the speaker into a room. This is what causes all the problems and the necessity for equalization.

Do you think the same problem exists as we transduce the 'perfect' voltage signal through a buttkicker into vibrations of a solid material? I guess that depends on whether the material has some sort of resonance in the frequency band? I really have no idea.

brucek

[RazorX]

brucek wrote:

Huh, never thought about it before. Certainly the the voltage response is 'perfect' before it is applied to the buttkicker load (just as it is for a subwoofer load). The requirement to equalize a subwoofer is a result of that 'perfect' voltage signal being transduced into sound through the speaker into a room. This is what causes all the problems and the necessity for equalization.

Do you think the same problem exists as we transduce the 'perfect' voltage signal through a buttkicker into vibrations of a solid material? I guess that depends on whether the material has some sort of resonance in the frequency band? I really have no idea.

brucek

I'm not sure but it doesn't seem to be a problem in my room. It doesn't make any difference in the graphs when they are on versus when they are off. I ran some more tests tonight and the graphs were pretty much identical regardless of the ButtKickers being on or off.

So like I mentioned...I ran some more tests, moved subs, etc. tonight. I ran at least three passes of each test to make sure there were no anomalies. I also tried each position with the polarity reversed and set to zero on each individual sub (every combination possible). Every placement test I tried sounded/looked better with the phase set to 0 on both subs so those are the graphs I'm including in this post (well, except for the first one just to show how much worse it looked than the last one)

I also tried each placement option with minor tweaks to the subs positions (distance from the walls, angled vs squared, etc.) These tweaks didn't seem to yeild much variance. The graphs in this post are all with the subs squared to the walls with the rear edge 5" from the wall and the side 2.5" from the side wall. The only exception to the distance was on the right rear stacked configuration (graph 4) only because I didn't have another cable that was long enough. This test was done with the subs 9" from the rear wall.

The first test was with everything the same as last night with the exception of the main speakers. I had them off for all of these tests. This one has the same ugly curve as the ones last night.


I stacked both subs in the front left corner for the next test and realized the sub that was in the right rear corner had the phase reversed. I have no idea how this happened but I do have two young daughters and one is pretty ornery...I know for a fact that I never reversed it. Anyway, I fixed it and ran the test.


I stacked both subs in the front right corner for the next test. Here are the results.


I stacked both subs in the right rear corner for the next test.


I had one sub in the front right corner and one in the right rear corner for the next test. Looking better?


The last test was done with the subs in their usual positions (left front and right rear corners) but this time with the phase set to 0 on the rear sub. Better than the same sub placement as the first graph?


What are your thoughts? Is my room unkind to the bass curve or is this pretty average?

Do you think I should go ahead and pick up one of the BFD's or wait and see what the curve looks like after I do additional acoustic treatment and/or after I get the new sub? Do you think a BFD would be able to resolve the problem spots in this curve?

Should I expect the new sub to have similar problems in this room or are the individual sub designs different enough that it is impossible to tell each one is measured in the room?

Also, based on the graphs above, which one would be the best starting point for correcting with a BFD?

Thanks

[brucek]

There's not a lot of difference between the last two graphs other than level, so either of those two situations looks fairly good. It looks typical of a lot of rooms with a peak around 37Hz....

The new sub would certainly change the low end extension, but the room is determining your response at these modest SPL measurement levels. Note that the 15" sub you mention above is tuned to 20Hz and has a response down to 18Hz, while the two Paradigmns are 10" with a low end extension of about 23Hz. You will lose rather than gain by trying to mix these different subs together. When we step up the food chain, the usual reaction is to keep the lesser subs, but it isn't a good idea.

You do appear to have a room resonance about 37Hz. A BFD would help with that for sure.

brucek

[clubfoot]

Were all your sub measurements taken with the subs in corners? Did you try 1/2 or 1/4 lengths along your rear and front walls? Did you try taking measurements with the spl meter forward and aft of your usual listening position?

[RazorX]

brucek wrote:

There's not a lot of difference between the last two graphs other than level, so either of those two situations looks fairly good. It looks typical of a lot of rooms with a peak around 37Hz....

The new sub would certainly change the low end extension, but the room is determining your response at these modest SPL measurement levels. Note that the 15" sub you mention above is tuned to 20Hz and has a response down to 18Hz, while the two Paradigmns are 10" with a low end extension of about 23Hz. You will lose rather than gain by trying to mix these different subs together. When we step up the food chain, the usual reaction is to keep the lesser subs, but it isn't a good idea.

You do appear to have a room resonance about 37Hz. A BFD would help with that for sure.

brucek

Thanks for the feedback. I see your point on not keeping the two Paradigms in the mix after the new sub comes. The thought of keeping them was really just a pipe dream anyway.

I will pick up a BFD and give it a try. Now I just need to decide which one to get...

I am going to see what the curve looks like with additional bass traps in the right rear corner tomorrow. I already have enough to do two more corners but just haven't installed them since I can't do the left rear corner.

I think I will also install more acoustic panels on the front and side walls like I mentioned earlier. While I'm doing this for reasons other than bass management, I'm curious to see if that has any impact on the frequency curve of the subs. It may be a while before I get around to that project though. I have a few projects at work that are kicking my but and not leaving me as much time to play lately - but hey those are the projects that provide me with the money to do the fun projects at home.

clubfoot wrote:

Were all your sub measurements taken with the subs in corners? Did you try 1/2 or 1/4 lengths along your rear and front walls? Did you try taking measurements with the spl meter forward and aft of your usual listening position?

All measurements were taken with the subs located in the three corners I am able to place them. I tried every combination of placing the the subs in these three corners.

I did not try the subs anywhere other than in the corners but I did experiment with distances for each corner. The reason I didn't try them anywhere else was because I know I will not place them there due to the layout of the room / not functional or ascetically pleasing. Besides, I wold hate to place them there for a test and find out they sound much better there and then not be able to take advantage of that placement. That would drive me crazy.

All measurements were taken with the meter in the usual listening location. I didn't try it in any other location. Should I do this? What would the benefit be in doing this?

[brucek]

I am going to see what the curve looks like with additional bass traps in the right rear corner tomorrow.

You'll find the traps may not be effective at the lower frequencies.

All measurements were taken with the meter in the usual listening location. I didn't try it in any other location. Should I do this? What would the benefit be in doing this?

The benefit is that sometimes when we move the listening position just a few feet it makes a huge difference in peaks and dips that are resistant to change when the subs are moved around. It's always worth a try.

brucek

[Wayne A. Pflughaupt]

RazorX wrote:

I will pick up a BFD and give it a try. Now I just need to decide which one to get...

The cheap 1124 works just fine.

Regards,
Wayne

[RazorX]

brucek wrote:

You'll find the traps may not be effective at the lower frequencies.

Ah, I didn't realize that. So the traps primarily remove peaks in the higher portion of the subs range? Something like in the ~50Hz - ~80Hz range perhaps. Could the traps I already have installed be causing the dips I have in ~50Hz - ~60Hz range and/or the ~70Hz - ~80Hz range. How much should I be concerned with these two dips? They look bad to me but I know I don't really know what I'm doing here yet. I don't really want to remove the ones I have installed because of how they are installed and the walls would require repainting if I removed them but I would do so if I had a high confidence in gaining a substantial improvement by doing so.

brucek wrote:

The benefit is that sometimes when we move the listening position just a few feet it makes a huge difference in peaks and dips that are resistant to change when the subs are moved around. It's always worth a try.

But wouldn't that be a misleading graph since it wasn't measured from the location where I listen to the system?

[RazorX]

Wayne A. Pflughaupt wrote:

The cheap 1124 works just fine.

Regards,
Wayne

Thanks Wayne. I am going to see if I can find either the DSP1124 or the FBQ2496 locally tomorrow so I can start playing with it this weekend. I'm so geeked on this now.

[brucek]

But wouldn't that be a misleading graph since it wasn't measured from the location where I listen to the system?

The inference was that you might want to move your listening position. If that's not possible, you're correct.

So the traps primarily remove peaks in the higher portion of the subs range?

I think traps are normally used for frequencies above ~80Hz to smooth the response. Below that a trap would become impossibly large. That's where EQ is normally used.

brucek

[clubfoot]

As brucek said you may have the option of moving your usual listening position fwd, aft or side to side and still choose a sub location that works well. If you can temporarily remove the existing traps, go ahead, give it a shot, this is the fun part of what we do and we'll all learn something by your feedback.

[RazorX]

Ah, thanks for the clarification.

Unfortunately, I am pretty much locked into the current seating position because of the location of the windows in the room.

I just checked the specs on the bass traps I'm using and they are indeed designed to a tackle frequencies much higher than my peak. Here is some details on of the frequencies they are designed to address.

Freq. - NRC
125 Hz - 1.24
250 Hz - 1.28
500 Hz - 1.45
1 KHz - 1.39
2 KHz - 1.27
4 KHz - 1.31
Overall - 1.35

I don't think it is worth the time or trouble to remove any traps or add additional bass traps after seeing the frequencies they tackle.

[clubfoot]

Yea, the traps would have to be "BIG" to do much good at low frequencies unless you want to build a stepped/raised platform and fill it with insulation and vents?!
The BFD should work nicely,...

[RazorX]

I decided to play around with some filters since I have a FBQ2496 on the way now.

Here is my first attempt at applying filters and a house curve to the following graphs.

This one is with the subs in the positions I normally run them in FL and RR corners.


This one is with both subs stacked in the RR corner. I did this one because this is the location the new sub will go in. It is really the only place I can put it because of the size of the enclosure.


The house curve file I am using contains the same settings as the example given in REW. The only reason I used this one is because it was the example given in REW and I haven't had a chance to experiment with house curves yet. Those settings are:
20 6.0
80 0.0

Feel free to let me know your thoughts.

[brucek]

Feel free to let me know your thoughts

Yeah, removing that single note at 37Hz will allow you to set the subwoofer level higher and get a much smoother sound overall. Right now all you're hearing at that listening position is 37Hz........

brucek

[clubfoot]

I prefer the first corrected graph because it has more usable extension low down.