My second attempt at measurements

[thewire]

I used my new external USB sound card and I want to be sure this was done correctly. I used the right CD input on my AVR from my right line output of the card. First I calibrated my subwoofer volume on the back of it to 80dB in REW. Then I raised the volume of my AVR until it said 75dB while playing the SPL calculation for testing with my mains. The output level in Windows XP was at the max setting for input and output using the Creative Surround Mixer window. It was set to stereo and there were no equalizers or effects on by default. I used a Radioshack digital SPL meter for measurements set to 80. During the subwoofer check level test prior to measurements, the output looked very good with -15 output and -15 input. During the check of the mains I got "Level OK -21.9" Before when I was using my subwoofer as the calibration I could not test my mains. I used the mic calculation file also this time. This is the result I got.


stereo - gold center red left orange right subwoofer - purple center black left blue right

The subs output looks like a little more than I'm used to seeing. During testing I did not have to change the volume on the AVR or subwoofer this way to test them both. Was I supposed to check them while calibrating the sub to test the sub and mains to test the mains as separate test? That might involve changing the output level on my AVR which I thought would change the calibration then.

Did I do this test correctly?

[brucek]

Was I supposed to check them while calibrating the sub to test the sub and mains to test the mains as separate test?

The method you use to arrive at the end result isn't that important. You generally want to start with the sub first and get it equalized. Then you can recalibrate the level in the room with the equalized sub to 75dB. Then when you check the mains with the sub you try and get them to meet nicely at the crossover at about the same level. Looks like you have a fairly large peak in your subs response that may need to be addressed...

brucek

[Wayne A. Pflughaupt]

thewire wrote:

Did I do this test correctly?

From your description and graph, everything sounds and looks right. Pretty impressive that your three subs positions all measure almost the same...

Regards,
Wayne

[thewire]

Wayne A. Pflughaupt wrote:

From your description and graph, everything sounds and looks right. Pretty impressive that your three subs positions all measure almost the same...

Regards,
Wayne

Thank you. The subwoofer currently sits centered between my stage made of plywood, filled with insulation, and also sits on top of a "sub dude". There was a 2" absorption panel behind the sub also. I calculated the absorption of my stage prior to it being built and did a demo with a very cheap DIY subwoofer that I made out of parts I had laying around.

Now for the difficult question... The last time I inquired about one these was in a proaudio/video store about six years ago. The salesman said that he wanted around 2k for one and he said it could not be setup without a pro for which they charged a price to calibrate. They offered additional calibrations at a lesser price if I remember correctly.

Will a Behringer FBQ2496 Feedback Destroyer Pro be able to lower that peak very well you think? I notice that the pre/pro I want only goes to 25Hz for the internal equalizer.

If so, where is a good place to get it? Should I get a long XLR to RCA cable or get an adapter, and use my current ones for now that I have but would like to replace anyway because I have to use a coupler? Any benefit for getting one of those now since I may want to later on?

[thewire]

This is a very nice help section. BQ2496 Feedback Destroyer Pro

I have looked some at the BFD forum area also but what are the limitations (if are) to how much a peak can be removed?

[thewire]

I lowered the measurement in REW -5 dB so I think this may give an idea of how far I need to lower it for the 75 dB target.

[brucek]

Yep, you've got about a 15dB peak to remove........

[thewire]

Does this look alright for an 80dB target? I think I will have to try a 78dB target and a 75dB one later when I can remeasure the HT. If I am turning the subwoofer volume up I should be telling the BFD to target the same level, and measure that in REW. Correct?

[brucek]

For your graph, please use a vertical scale of 45dB-105dB and a horizontal scale of 15Hz-200Hz.........

Use a 75dB target and set REW up for 75dB...........

[JohnM]

85dB would probably be a closer target level for that trace. Also when adjusting your filters check the box to "invert filter responses" and adjust the Q and gain to try and get the filter response to overlay the measurement, that will give the flattest corrected response. From the corrected trace in that plot it looks like the filter Q/BW settings have not been adjusted.

[thewire]

brucek wrote:

For your graph, please use a vertical scale of 45dB-105dB and a horizontal scale of 15Hz-200Hz.........

Use a 75dB target and set REW up for 75dB...........

Bellow 45dB and after 250Hz is where my subwoofer begins to cause reverberations problems in my room according to the RT60. I thought it may have had relevance to the filters that I am applying there at those peaks. I will not post images in the scale of 45dB-105dB and a horizontal scale of 15Hz-200Hz but I thought it may have been an issue to discuss later on.

JohnM wrote:

5dB would probably be a closer target level for that trace. Also when adjusting your filters check the box to "invert filter responses" and adjust the Q and gain to try and get the filter response to overlay the measurement, that will give the flattest corrected response. From the corrected trace in that plot it looks like the filter Q/BW settings have not been adjusted.

I forgot to adjust the Q and gain. I did the BW adjustments. Not sure about a 85 target. Sound and Vision has this to say about the subwoofer.

Bass limits for the THX-12SUB subwoofer were measured with it set to maximum bandwidth and placed in the optimal corner of a 7,500-cubic-foot room. In a smaller room users can expect 2 to 3 Hz deeper extension and up to 3 dB higher sound-pressure level (SPL). The only response difference between the VAR and THX settings is that the latter reduces levels by 9.5 dB. In either position the subwoofer has useable response up to 250 Hz, where it measured -6 dB. Dynamically, the THX-12SUB will deliver 106 dB SPL at 32 Hz or above, which is better than average.

I usually have peaks around 115dB around 60Hz currently without any equalization applied. This may be something I would try after applying filters, but certainly not something I would attempt otherwise. I will see what I can do about the target later. I have tried a 85dB before but it was overloading the subwoofer.

If I am doing a 85dB target then by the time I am turning my Polk mains up to match, they began to making pop noises. I don't want to overpower my speakers or clip my mono amps, whichever that may be.

[thewire]

This is only practice but does this look like what I should try with filters? Would this cause audible distortion most likely this way? I set five filters for this one.

[thewire]

Without 1/3 smoothing it says it is going to be worse when I boost the 60Hz area.



Placing traps on my back wall fixes that...

[Wayne A. Pflughaupt]

Steven,

When you post graphs, please give us only the (dated) response plot, corrected and target traces. The others only make things confusing.

What kind of subwoofers are you using – make and model if manufactured, or details on driver, aligment and amp if it’s a DIY?

Also, since you mentioned a stage, I assume this is a smallish dedicated room?

Regards,
Wayne

[thewire]

I see your point that could get confusing. I use a single Crystal Acoustics 12" subwoofer placed in the front of the room. It is a 200W 4ohm driver. For the new testing, I have a 20' bettercables coaxial cable coupled using the adapter to a 25' run of stereo subwoofer cable (using one side) that I used to use for my car subwoofers before I lost those. The receiver is a Pioneer VSX-1012 model.

The room is a dedicated room with an equipment closet and an entryway on the left side of the room. The closet in the front left of the room will be where equipment will be located later on. The room needs insulation in some areas and in some places in the rear of the room where not entirely finished due to budget limitations. As I am currently remodeling our other home for sale (painting, laying tile etc) we will be upgrading the HT soon. Two subs is an option but I have been advised to avoid this if possible.

Photos of the room.

http://www.flickr.com/photos/tns29/sets/

My build thread here that is very outdated at this time but will update. My father got a new 2007 Sierra, I repainted my crown molding, finished sealing my entrance door, got some 244 bass traps on the front wall, moved my speakers off the stage, and moved the sub to the front of my room onto a sub dude, and bought a new tweeter for my center channel since my last post.

build thread]


I think that the problem is not my rear wall but instead axial mode problems from my left right wall mostly relating to my ceiling. It needs better insulation and I hear problems near the steel beams up there. The pink insulation in it will maybe last another year but I hope to fix that sooner. Maybe inexpensive insulation left in the bag will help there. That may help improve the tangential and axial modes I seem to have very close around 60Hz. I'm not sure.

I will also use the new 4.11 version and correct my graphs. I have many on my website but that are not in sets that I will need to delete later. I thought while I was away, I would practice with my older measurements while reading.

I wasn't purposefully meaning to make things confusing but rather trying to clear up my own. I apologize to anyone to that I may have done so.

Thank you.

edit: I also got another mono amplifier.

[Wayne A. Pflughaupt]

You might want to try different locations for the sub, like maybe a corner. That usually gets better readings, both smoother and better-extended, than center-of-wall placement. It also may reduce the span between the highest and lowest peaks, which will make for easier equalizing. Once you find the best location, you can add your second sub there, if it’s a match for the first.

Regards,
Wayne

[JohnM]

thewire wrote:

This is only practice but does this look like what I should try with filters? Would this cause audible distortion most likely this way? I set five filters for this one.

Those filter settings will be doing a lot more harm than good. They are very wide, conflict strongly with each other and some have very high gain which is in all cases a very bad thing.

The figure on the far right of each filter setting shows its bandwidth in Hz. The 48Hz filter, for example, has bandwidth of 55Hz so is affecting the range from around 20Hz to 75Hz. The 32.8 Hz filter is affecting from about 15Hz to about 50Hz. The 20Hz filter has high gain, 9dB, and affects from about 15 to 27Hz - much of it countered by the 32.8Hz filter. The 64.7Hz filter has 15dB of gain and affects from about 55 to 75Hz. It looks like you may be trying to use that to counter the notch in the response. That doesn't work, notches like that are caused by the direct signal being cancelled by a reflection, making the direct signal bigger also makes the reflection bigger so the situation is not improved and the gain of the filter is likely to cause clipping and distortion at several points in the chain. Is the 221.8Hz filter being used as some sort of pseudo crossover? It is well beyond the working range of most subs when there is a crossover in the path.

Avoid any filter with more than a small amount of gain (3 or 4dB). Make sure the filter bandwidths correspond to the problem you are trying to correct. Start with 1 filter on whatever is the biggest peak you are trying to address and listen to the effect of that before moving on to another, if you still hear problems in the response.

[thewire]

Wayne A. Pflughaupt wrote:

You might want to try different locations for the sub, like maybe a corner. That usually gets better readings, both smoother and better-extended, than center-of-wall placement. It also may reduce the span between the highest and lowest peaks, which will make for easier equalizing. Once you find the best location, you can add your second sub there, if it’s a match for the first.

Regards,
Wayne

The HT, equipment closet, and entrance area is partially underground with concrete on all sides including the ceiling but the ceiling has an additional support of very thick steel that is supported by the large steel beams. It was a reverberation chamber before any framing was done with the exception of an opening where one of the future doors went.

I tested with my DIY subwoofer before the room was completed at the center front wall and ended up placing it there since the bass seemed to be most even throughout the room with it being there. I also added a hookup wall plate in the right corner of the room for future use because while I was listening it was not very loud there and I thought it may be a good place for a subwoofer someday.




I bought the new subwoofer and a new driver for my DIY speaker box that I tried but it did not work having a subwoofer in my back right corner and the new one in the center. They were canceling each-other out and my new replacement driver (4 ohm) was becoming very badly distorted by my Marantz 180W @ 4ohm mono amplifier so I returned the speaker to the store.



I have moved the sub and tried different locations by ear so far. The first place I had it was the back right corner. My father complained that he could localize it from the back right seat that he usually sits in. I tried a 50Hz and 80Hz crossover and neither was working to eliminate that problem. I had placed it there by placing the sub at the center front row seat and finding where it sounded the best. I could also localize it on the right side also in the front of the room after using the receivers automatic setup during certain scenes because I had to turn up my subwoofer very loud to blend with the mains. That was before bass traps however.

I ended up moving the subwoofer to the center of the back wall where according to my folks, they were unable to localize the sound anymore and it did not sound "boomy" to them there. I left it there sitting on of my equipment rack because I was unable to place it behind or bellow it. That is where I tested my first time with REW and began placing room treatments. The measurement looks very similar to what I see now with it placed in the front of the room but I will have to try placing it there on the riser (which I am not sure will work) after I move the equipment to retest that location more accurately.

I tried the right side of the room and also the left side like shown bellow after installing bass traps - the new 244 traps and prior to doing any new testing with REW more recently. It sounded excellent on the right side of the room but I was still able to localize it. I could not place it on the floor which I think had to do with the problem. I tried facing it different directions but it seemed to make things worse. The left side sounded very quiet in comparison to the right.



The closer I put it to a wall, the more I am able to localize bass there with the exception of placing it between my left tower and center channel as seen bellow. It sounded good there but I was able to localize higher frequency sounds there around the crossover. It would probably be the 80Hz peak that I am seeing during my testing.



I would mostly like a second subwoofer to extend the low frequency which I understand is possible. I think that I would place it in the back of the room if I could and set the phase on the subwoofers accordingly. I have the phase set to normal in the front of the room now and in the back of the room, reverse was working best. I found that in the back it liked to be slightly closer to the left wall, which has rooms on the other side and is not drywall/insulation/bedrock such as the right side of the room. It might look something like bellow except that I would try tweaking them left and right slightly which I am able to do with good results. The speakers in the SketchUp model are not to scale but the rest of the room is. I have planned to first try placing the dipoles on the right and left sides of the front row and move the rear channels away from the center of the back wall and more in a Dolby type configuration. I'm not sure about the bookshelves yet. I have since moved my panels on the front wall together, and placed standard sized GIK 244 bass traps on the front stage area covering the front wall and in front of the screen, which I need to complete placing more under the screen area.

There is conduit behind my screen for running wiring but I have not yet decided if I will use that or hide it behind baseboard on the stage.

It looks like this currently.




Moving it forward and back did not make any difference but if moving it to the left or right between the stage a few inches helps, I'm not sure yet. It seemed to help in the back of the room but it could have easily have been my imagination. I will test that with REW and see what happens.

JohnM,

Thank you that explanation helps allot and I think it is beginning to making better since to me now. The higher filter I tried applying was trying to lower the peak around 250 because according to the RT60 on my measurements with the sub, it went off the chart for one of my seats, and the other seat was slightly above .08 seconds at that frequency to about 300Hz. It did a similar thing for my sub higher above that also in the testing. I had the subwoofer speaker target in my settings with the crossover at 80Hz and I had thought the sound would not be going so high with these settings enabled unless there was some kind of adjustment I was supposed to be tweaking. It is my mistake, and I think this is only a result of sound after the speaker has began loosing signal. If that is not what I am seeing, please let me know.

[thewire]

The RT60 measurement of the subwoofer looks fine with a one octave filter. It goes until 200Hz and then falls. I think this must have been where I made an error in looking at the measurement. It does much less with mains+subwoofer for one third octive bands.

[clubfoot]

The easy way to find the proper location for a sub is to place the sub in your usual seating position and move your mic around the room. Disregard what you hear,...use REW to take multiple measurements and try to find a response that tracks the "ideal" 75db target curve. The closer you get it to track the curve the lower the number of filters you will need to employ,...the better it should sound. If you place the sub behind you, it will probable need a phase adjustment to blend with the mains. Small changes in location can make a big difference in measurement!

Hope this helps.

If you take a look at my two plots in my sig those are without smoothing and no filters, BFD in bypass, only phase for smoothing with mains.

[thewire]

I unchecked the smoothing and tried the "find peaks" with an older measurement I did with version 4.10 using a 75dB target the other night. This had been before I realized however not to be using the left channel for sound card calibration. The response was slightly shifted to the right where 25Hz was at 75dB instead of 20Hz and bellow 15Hz it was climbing. REW recommended one filter at near 33Hz where the subwoofer is at right now and it was not more than -15. I think I may be alright, but I will at least experiment with