What's new
  • Announcing New Ownership at Home Theater Forum. Learn More

Subwoofer Calibration Probelm (1 Viewer)

BraveHeart123

Second Unit
Joined
Oct 5, 2010
Messages
271
Real Name
Faisal
I am caught up in this dilemma of chasing my own tail while calibrating the subwoofer. The kind of nerd I am, I won't stop untill I tame it to perfection, but I am unable to properly calibrate it. There are huge peaks and and dips in my bedroom, and unfortunately I can't move the HT setup in livingroom.
OK let's cough out my problem and hope somebody like "Robert_J" here can help me out. I am using Klipsch SUB-12 in my set up with Onkyo 809, and BFD 1124p for correcting sub response. I used the NCH to create the test tones. I've been following the following procedure;
  • I reset the receiver to factory defaults and then run the auto calibration to adjust the sub at 80 dB using internal test tone. That is the 2nd step while running audyssey after selecting the impedence of main speakers.
  • I only set the sub level and do not go any further with Audyessy auto setup and disconnect the mic. Because Audyssey experts say, the subwoofer calibration must be done before running Audyssey auto calibration if any outboard subwoofer equalization is involed.
  • Then I setup RS analog SPL meter on a tripod at my listening position at ear height with its mic pointed upward at the ceiling, and set it c-weighted, slow response, and 80 dB.
  • I disable all the internal equalization on the reciever, disconnect all the speakers except subwoofer, dial the volume knob so that it shows 00 (Relative Scale) on the display. I set the volume at 00 coz Onkyo says the internal test tones of avr are played at reference volume automatically no matter where the actual volume is.
  • Then I run test tones through Seagate FreeAgent Theater+1080P HD Media Player, which is connected with AVR through HDMI. And check the SPL on the rat shack meter.
  • I note the in-room response at all the test frquencies, set the filters in BFD sofware on computer, add BFD 1124p in the sub chain, and run the subwoofer test tone from the reciever again to confirm it doesn't go into clipping at reference volume.
  • Then I set the filters on BFD and take the frequency response again.
My problem is that the response is no way near the 80 dB SPL set earlier through avr test tone done in step 1 (BEFORE ADDING BFD in the SOUND CHAIN) and it is all over the place.
The response curve I see after setting the filters in BFD calibration software on computer is all together different from what is actually happening in real.
The SPL goes off chart i.e. the red clipping LEDs light up constantly on BFD at certain frequencies.
Is there something wrong with how the test tones (SINE WAVES) have been created?? I mean I don't know if there is any input level mismatch of test tones created with NCH.
My entire effort goes wasted. Coz when I run audyssey after that, it sets all the speakers to 75 dB and my sub is running at 90 dB+. I am baffled.....what am I doing wrong here???
For reference, I'm attaching the response charts.
Raw Response Without BFD
38231a42_Raw_Response.jpeg

RAW SPL
a9a65ad5_Raw_SPL.jpeg

Projected Response After Adding BFD Filters
fb0f49a4_FiltersApplied.jpeg

This projected response is nowhere near the actual response. What am I doing wrong here???
 

Robert_J

Senior HTF Member
Joined
Aug 22, 2000
Messages
8,350
Location
Mississippi
Real Name
Robert
When you plot your response after setting the BFD, are you using a fresh spreadsheet with the "EQ" set to 0 on all filters?
Also, how does it sound after setting the BFD? I noticed an immediate improvement in my system.
You did everything that I did. Same software to generate the tones. Same meter. Same plotting/simulation software. Your projected response looks awesome so give us a screen shot how it really turns out so we can compare.
 

BraveHeart123

Second Unit
Joined
Oct 5, 2010
Messages
271
Real Name
Faisal
I knew you would chime in ... :) I feel like being in good hands
I don't set all the filters to 0, but disable all of them by unchecking. I guess that's the same thing.
Also, my question still remains unanswered. Why so much difference in SPL between the internal test tones and the the ones generated through NCH?? I simply find it pointless coz there exists a huge level difference to start with. I am unable to figure out starting reference point to calibrate to.
 

Robert_J

Senior HTF Member
Joined
Aug 22, 2000
Messages
8,350
Location
Mississippi
Real Name
Robert
Test tones are usually -10 to -20db from 0. But that shouldn't matter. Once the BFD filters are set, they work no matter what volume the receiver is set at.
The SPL goes off chart i.e. the red clipping LEDs light up constantly on BFD at certain frequencies.
I re-read your post and noticed this. This is odd because my BFD rarely lights the meter. The next time I go downstairs I'll check a setting on it.
 

BraveHeart123

Second Unit
Joined
Oct 5, 2010
Messages
271
Real Name
Faisal
Test tones are usually -10 to -20db from 0
It makes sense, if it is AVR test tones you are referring to. But internal test tones are wide band pink noise, whereas the tones made through NCH are sine waves. They can never ever match in SPL at, let's say, reference volume whatever that may be.
So, that makes me wana ask......
Where should I keep the volume dial on AVR and where should the subwoofer level be on avr??
What i am really confused about is where to keep the starting point in terms of sub SPL???
Coz for sub calibration, I am running sine waves and each sine wave is different in SPL at a fixed (main volume) and (subwoofer level) on AVR. For example for an 80dB flat response, if the spl measured with rs meter is 80dB at 25hz, 95dB at 50hz, 87dB at 60Hz, etc, how do I determine if 80dB at 20 hz is the correct spl and I have to remove dips and spikes on all other frequencies to be within the required spl of +/- 3dB of 80dB.
I think it's all going wrong with the wrong selection of starting point.
Also, I am using XLR-RCA adapter to connect between AVR and BFD. Is it OK??
 

Robert_J

Senior HTF Member
Joined
Aug 22, 2000
Messages
8,350
Location
Mississippi
Real Name
Robert
Adapter doesn't matter as long as you get a signal to the BFD.
I chose to use 50.1hz as my starting frequency. I adjusted the AVR's volume knob until the meter read 85db. I then went to 100hz and worked my way backwards through the different tracks on my CD. Actually, I recorded them in reverse order. Track 1 = 100hz. Track 2 = 94.4. Track 3 etc. I think my LFE level was set to +1 or +2 but that was from the initial MCACC calibration. The attenuators on the amp are set 2 detents from wide open.
 

BraveHeart123

Second Unit
Joined
Oct 5, 2010
Messages
271
Real Name
Faisal
I chose to use 50.1hz as my starting frequency.
What does it mean?? Does it mean you take whatever the SPL at 50.1hz is as your reference SPL??
Where is the subwoofer trim level ( +, -, or 00 position??) on the AVR when you dial the volume knob to get 85dB on ratshack meter??
You say you calibrate your sub to 85dB. Are you running it 10dB hot i.e. are your speakers calibrated to 75dB spl??
Also, how does audyssey configure the sub when we run auto setup after calibrating sub like this with customised test tones being 10dB hotter than internal test tones of AVR?? Coz audyssey calibrates the speakers and sub to 75dB at the listening position at reference volume (00) no matter where the voume dial is during manual calibration of the sub using self generated test tones.
I'm sorry for questioning like this, but I need to know coz I think this level matching confusion is the culprit.
I plotted the FR after adding BFD and setting the filters on it. It looks horrendous. This graph clearly shows my dilemma of incorrect understanding of levels between internal test tones of AVR and those created by me. It does not match the projected FR curve in BFD software posted earlier.
FR After Adding BFD and Settiing Filters
e48491ec_After_BFD.jpeg
 

Robert_J

Senior HTF Member
Joined
Aug 22, 2000
Messages
8,350
Location
Mississippi
Real Name
Robert
Yes. I play the 50.1 hz tone and adjust the AVR's volume knob until the SPL meter reads 85. I don't touch the volume knob until I go through the entire disc of test tones. Sometimes, when the volume is lower than 80, I have to adjust my analog meter to the 70db setting so I can get a reading.
The sub trim (LFE setting) is also at 0 to start.
You say you calibrate your sub to 85dB. Are you running it 10dB hot i.e. are your speakers calibrated to 75dB spl??
This sub equalization is completely separate from the system calibration. Once the peaks and valleys of the sub are tamed by the EQ, then MCACC auto calibration is much more accurate. But I do boost the sub +3 after calibration because I like a little more thump.
Also, how does audyssey configure the sub when we run auto setup after calibrating sub like this with customised test tones being 10dB hotter than internal test tones of AVR??
It doesn't care what level tones were used before. It doesn't know you did anything before. MCACC or Audyssey or any other auto calibration will now work better because the sub has a flatter response than without the BFD. Equalizing the sub and calibrating the entire system are completely different processes.
I plotted the FR after adding BFD
You are correct, that is horrible. That 30hz peak is unreal. I suggest you start over but only use 3 or 4 filters. Just tame the largest peaks and see if that works. I think I use a total of 6 filters on a DIY sub that has a HUGE inductance peak at 55hz. 3 to control that one, two to boost the low end and one somewhere that I can't remember.
 

BraveHeart123

Second Unit
Joined
Oct 5, 2010
Messages
271
Real Name
Faisal
Many thanks .... I get it now. I will redo the eq tomorrow and paste the FR here.
Not related to this thread, but I've received Crown XLS1000 amp today to power up my DIY sub you helped me with in another thread. I will put that to test after I am done with this.
 

BraveHeart123

Second Unit
Joined
Oct 5, 2010
Messages
271
Real Name
Faisal
Did as per you said and things are much better than before.
FR Before Adding BFD
1b503395_BeforeBFD.x-ms-bmp

FR Before Adding BFD
8866fe4e_After_BFD.x-ms-bmp

The response is smoother than ever. :)
 

Users who are viewing this thread

Forum Sponsors

Latest Articles

Forum statistics

Threads
352,555
Messages
4,989,139
Members
143,285
Latest member
bnick8576
Recent bookmarks
0
Top