AudioCircle
Industry Circles => GR Research => Topic started by: mlundy57 on 18 Aug 2018, 07:27 pm
-
I ran REW in my new room today. There is about a 6dB dip between 60 and 200 Hz. This is so whether the crossover controls on the A370 amps are set on 80 (halfway) or 120 (wide open). The red line is the left speaker/sub only, the green line is the right speaker/sub only, and the blue line is both speakers/subs.
(https://www.audiocircle.com/image.php?id=183484)
Here are a couple pics of the room. As you can see, there is not a lot of flexibility here.
(https://www.audiocircle.com/image.php?id=183485)
(https://www.audiocircle.com/image.php?id=183486)
Any ideas?
Thanks,
Mike
-
Turn the phase knob on the crossover and see what happens.
-
Turn the phase knob on the crossover and see what happens.
I’ve tried that at 0, 90, and 180. I’ve also tried reversing the phase on the NX-165 drivers with the sub at those 3 positions. No difference
I have also moved the microphone closer to the speakers by 1’ and 2’. The dip got worse.
-
I’ve tried that at 0, 90, and 180. I’ve also tried reversing the phase on the NX-165 drivers with the sub at those 3 positions. No difference
I have also moved the microphone closer to the speakers by 1’ and 2’. The dip got worse.
What are your other settings set to on the amp?
-
What are your other settings set to on the amp?
The crossover is at max (120). There was no difference in the dip between the 12 o’clock position on the crossover and the max position. The line in filter is on AVR 12, rumble filter off, extension 14 Hz, low damping.
Volume is at 11 o’clock. Any higher and the lower bass is too much. As it is I have PEQ on with gain at 1 o’clock, bandwidth at minimum, and frequency at 12 o’clock to tame a 40Hz peak.
-
Turn off the main amp driving the top speakers and take measurements with just the subs on.
After that, turn off the subs, turn back on the main amp, and take measurements of just the top sections running.
I also recommend putting the rumble filter on, setting the damping to high and the extension to 28hz.
I suspect that you're room's too small and the low bass is overwhelming it, causing the very low end to be very boosted to the mid/upper bass. These settings should help ameliorate that.
Don't worry, your still gonna get very strong low bass. Trust me :)
-
That's a pretty big room suck out. It will have a lot to do with the size and length of the room verses how far the speakers are out into the room.
Try pulling the speakers further out into the room and see what happens.
It also helps to shoot the output of the MTM and sub separately then overlay them to see how they sum.
And stick with just looking at one speaker at a time.
-
And what Tyson just said... :thumb:
-
Turn off the main amp driving the top speakers and take measurements with just the subs on.
After that, turn off the subs, turn back on the main amp, and take measurements of just the top sections running.
I also recommend putting the rumble filter on, setting the damping to high and the extension to 28hz.
I suspect that you're room's too small and the low bass is overwhelming it, causing the very low end to be very boosted to the mid/upper bass. These settings should help ameliorate that.
Don't worry, your still gonna get very strong low bass. Trust me :)
I’ll try the changing the toggle switchs tomorrow.
I’ve already done the rest. The first thing I did was measure the left speaker with the sub off to find out where it was 6dB down. Then I turned the power amp off and the sub amp on. Then I adjusted the sub to get it 6dB down at the same point. Then I pulled down the peak at 40Hz. Next I turned the power amp back on and measured both sections together. I expected them to be additive and pull the dip up but it didn’t happen. I increased the crossover point on the sub to see if it would help but it didn’t.
Next I did the same process with the right speaker and got the same result.
Finally I measured both speakers together. The SPLs were higher but the pattern, including the dip, remained.
-
I'd still like to see the individual responses.
Not only pull them out into the room further but change the toe in and see what happens too.
-
That's a pretty big room suck out. It will have a lot to do with the size and length of the room verses how far the speakers are out into the room.
Try pulling the speakers further out into the room and see what happens.
It also helps to shoot the output of the MTM and sub separately then overlay them to see how they sum.
And stick with just looking at one speaker at a time.
Currently the back of the subs are 3’ into the room. That puts the sub’s baffles 42” into the room and the MTM’s baffles 50”. Inside to inside the speakers are 74” apart (89” center to center) and 94” (almost 8ft) from the listening position. The LP is 19” off the back wall.
The room is pretty small, 12’x14’x8’
-
I'd still like to see the individual responses.
Not only pull them out into the room further but change the toe in and see what happens too.
More or less on the toe in?
-
More or less on the toe in?
Both. It will change the off axis cancellation pattern and how the room is loaded.
-
Both. It will change the off axis cancellation pattern and how the room is loaded.
Ok. Have to wait until tomorrow. Wife has gone to bed so I’m banned from running sweeps for the rest of the night. The bedroom is below my music room :nono:
-
Ok, I've adjusted rumble filter, extension, and damping settings. The only real difference was with rumble on the low bass was actually louder but it didn't affect the dip.
I moved the speaker (only working with the left speaker at this point) up to 6" into the room, 6"closer to the front wall, 6" closer to the center, and 6" closer to the side wall.
I adjusted toe in incrementally until the speaker almost faced the side wall and crossed in front of the LP like a horn.
I also tried removing the diffraction/absorption (GIK Alpha panel) at the first reflection point.
The only thing I really managed to do was turn the dip into a bi-modal dip at 125Hz and 175Hz.
(https://www.audiocircle.com/image.php?id=183523)
Mike
-
Just to confirm you can make sure each is playing down to or up to to the range they are suppose to cover. To do this you can turn level way down and take a near field of the upper woofer and of the lower mid. And by near field I mean from about 1/4" away from the woofer dust cap or phase plug. Then we'll know what the drivers themselves are doing.
-
Just to confirm you can make sure each is playing down to or up to to the range they are suppose to cover. To do this you can turn level way down and take a near field of the upper woofer and of the lower mid. And by near field I mean from about 1/4" away from the woofer dust cap or phase plug. Then we'll know what the drivers themselves are doing.
I’ll try that this evening when I get back from a birthday party
-
Just to confirm you can make sure each is playing down to or up to to the range they are suppose to cover. To do this you can turn level way down and take a near field of the upper woofer and of the lower mid. And by near field I mean from about 1/4" away from the woofer dust cap or phase plug. Then we'll know what the drivers themselves are doing.
The drivers are doing what they are supposed to
(https://www.audiocircle.com/image.php?id=183545)
-
The dip is the room 'talking' to you.
Why are the subs located under the mains? (Probably the worse place for them.)
Have you tried various sub locations, including raising one up?
Have you tried running one of the subs out of phase?
Have you tried the 'sub crawl' to properly locate the subs?
Why not just EQ the 6 dB dip out?
-
The dip is the room 'talking' to you.
Why are the subs located under the mains? (Probably the worse place for them.)
Have you tried various sub locations, including raising one up?
Have you tried running one of the subs out of phase?
Have you tried the 'sub crawl' to properly locate the subs?
Why not just EQ the 6 dB dip out?
It is the way they were designed. They are two part full range open baffle speakers. They could have been built as a single cabinet but they are easier to move around (especially up stairs) in two pieces.
As I understand it, raising a dip is not a good idea.
-
You can EQ some of it out with the servo amp, but it is best to try and solve it first.
-
You can EQ some of it out with the servo amp, but it is best to try and solve it first.
Solving it is going to be the thing. I’m already using the sub’s PEQ to bring down a peak at 40Hz.
Since the drivers are easily playing these frequencies it is definitely a cancellation effect in the room. But knowing what it is and figuring out how to fix it in this room are two different things.
Tomorrow I will remove the bass traps from behind the speakers and see if that helps.
-
I’d make sure the acoustical centers are on the intersections the 1/5th and 1/7th division of the room. Then, rather than try to move the speakers around, try shooting the room response +- 1 foot from your current listening position.
That’s a lot of subwoofer for that small of a room.
-
Mike,
Do you have a nearfield measurement of one servo woofer as a comparison?
-
Mike,
Do you have a nearfield measurement of one servo woofer as a comparison?
Rich,
No I don't. However, the midwoofers should be covering this range without any help from the subs.
Mike
-
I did some more moving. I pushed the speaker back until the baffle of the NX-Otica section was 3' from the front wall. At this point the dip was gone.
The green line is the NX-Otica section by itself
(https://www.audiocircle.com/image.php?id=183572)
The gold line is the sub by itself. I adjusted the crossover point until the gold line crossed the green line about 6dB down (at 60Hz).
You would think the two would couple at this point and when played together the dip at 60Hz would go away. However, this is not what happened. When I played both together, the dip at 60Hz got 2dB worse.
Mike
-
You would think the two would couple at this point and when played together the dip at 60Hz would go away. However, this is not what happened. When I played both together, the dip at 60Hz got 2dB worse.
Now start adjusting the phase control and the dipped area should go away.
-
Now start adjusting the phase control and the dipped area should go away.
Ahhh, I never really understood how/when to use the phase control to change the measurements. I'll get on that this evening
Thanks
Mike
-
Now start adjusting the phase control and the dipped area should go away.
Here is what I have after experimenting with different phase settings:
The Green line is the Otica by itself. The Purple line is the sub by itself. The Red line is both together
(https://www.audiocircle.com/image.php?id=183574)
-
Seems like the Ottica has a hump right at the bottom of it's response. You might try swapping the speakers so the wings are on the outside of the speakers instead of the inside of the speakers. I'm thinking the way they are now might be causing a "bass cavity" or "bass tunnel" with the side walls that's causing that bump in response you see at the very bottom.
-
Seems like the Ottica has a hump right at the bottom of it's response. You might try swapping the speakers so the wings are on the outside of the speakers instead of the inside of the speakers. I'm thinking the way they are now might be causing a "bass cavity" or "bass tunnel" with the side walls that's causing that bump in response you see at the very bottom.
I'm not sure they will image properly with the large wings on the outside. One thing about the graphs is that they have 2dB graduations. that makes the changes more exaggerated than they would appear with 5dB gradations.
-
Once I got the left speaker to the above point I worked on the right speaker. Once I got it as close as I could I disconnected the computer and rechecked the center image. It was now off to the right.
I recentered the image then reconnected the computer and shot new measurements. They were now off so I readjusted the speakers. I was able to get the right speaker to an in room response of plus or minus 4dB. This graph has 1dB gradations.
(https://www.audiocircle.com/image.php?id=183592)
The right wasn't quite as good. It was plus or minus 6dB. This graph has 2dB gradations.
(https://www.audiocircle.com/image.php?id=183593)
Both together came in at plus or minus 4dB. Also 2dB gradations.
(https://www.audiocircle.com/image.php?id=183594)
Mike
-
Once I got the left speaker to the above point I worked on the right speaker. Once I got it as close as I could I disconnected the computer and rechecked the center image. It was now off to the right.
I recentered the image then reconnected the computer and shot new measurements. They were now off so I readjusted the speakers. I was able to get the right speaker to an in room response of plus or minus 4dB. This graph has 1dB gradations.
(https://www.audiocircle.com/image.php?id=183592)
The right wasn't quite as good. It was plus or minus 6dB. This graph has 2dB gradations.
(https://www.audiocircle.com/image.php?id=183593)
Both together came in at plus or minus 4dB. Also 2dB gradations.
(https://www.audiocircle.com/image.php?id=183594)
Mike
That's a damn fine result! Nice work :thumb:
-
Mike,
Definitely looks like improvement.
Do you mind (for each speaker individually):
1. Removing all smoothing (or 1/48th octave).
2. Only showing 20Hz to 400 Hz (changing horizontal axis).
3. Changing vertical axis to 5dB increments.
4. And thus, changing vertical range to 50 dB.
I suspect, things are a lot flatter in your bass/midbass than they look.
Thanks,
Anand.
-
That's a damn fine result! Nice work :thumb:
Tyson,
Thanks
Mike,
Definitely looks like improvement.
Do you mind (for each speaker individually):
1. Removing all smoothing (or 1/48th octave).
2. Only showing 20Hz to 400 Hz (changing horizontal axis).
3. Changing vertical axis to 5dB increments.
4. And thus, changing vertical range to 50 dB.
I suspect, things are a lot flatter in your bass/midbass than they look.
Thanks,
Anand.
Anand,
I have all the measurements saved so I can do that later this week.
I have learned a number of lessons during this. One of the two biggest is that farther out into the room isn’t always better, even with OB speakers. It depends on the depth of the room. In this case I started with Otica baffles 50” into the room but had to push them back to 36” to get away from the canceling effect around 150Hz.
The other biggie was the proper use of the phase control with the measurements. MY GAWD these things sound Good!! They are better than I have ever heard them, even at LSAF.
Before Steppenwolf, Uriah Heep, CCR, etc sounded good but they didn’t’t get your toes tapping (if those don’t get your toes tapping something’s wrong). Now they have me almost dancing in my chair :D
The final movement of Mahler’s 8th, from its ethereal whispers to its layering and rising voices to its final crescendo is something else. Pipe organs too.
I’ve always known these were great speakers but they have shown me a side of themselves I haven’t experienced before.
Mike
-
Mike,
Happy for you.
But 1 thing is for sure. Without taking measurements one would never know and therefore I would encourage others to do the same.
Best,
Anand.
-
Looks pretty darn good in-room Mike :)
jay
-
The other thing you can do - now that you know where your biggest bass hump still resides, you can buy a bass trap from GIK Acoustics that's "tuned" specifically to deal with that frequency. IME its better to deal with room modes like that acoustically instead of electronically (ie, with an EQ).
-
Dont be afraid to land phase somewhere not on a 90° interval. The best phase is rarely that incrementally found.
What is the space between the speakers?
-
Mike,
Definitely looks like improvement.
Do you mind (for each speaker individually):
1. Removing all smoothing (or 1/48th octave).
2. Only showing 20Hz to 400 Hz (changing horizontal axis).
3. Changing vertical axis to 5dB increments.
4. And thus, changing vertical range to 50 dB.
I suspect, things are a lot flatter in your bass/midbass than they look.
Thanks,
Anand.
Anand,
Here are the graphs with those parameters: (in REW the graphs are showing 5dB segments with a 50dB range but when I capture the image it converts them to 10dB segments. I don't know why or how to get the JPEG to be in 5dB segments.) Figured it out, I had to change the save resolution to 1,095 in REW.
Left Speaker
(https://www.audiocircle.com/image.php?id=183738)
Right Speaker
(https://www.audiocircle.com/image.php?id=183739)
and both speakers
(https://www.audiocircle.com/image.php?id=183740)
Mike
-
Dont be afraid to land phase somewhere not on a 90° interval. The best phase is rarely that incrementally found.
What is the space between the speakers?
Jeremy,
I checked it from 0 to 180 checking at 0 degrees, 9 o'clock, 12 o'clock (90 degrees), 3 o'clock, and 180 degrees. Then I went back to where it looked bestand continued to fine tune.
It worked best around 10:30 - 11:00 o'clock.
Mike
-
Mike,
Looks fine. You’ve got a little peak just under 100 Hz (about 70-85 Hz)which is likely room related and treatments will help. The area from 100-300Hz looks quite good too especially with SBIR being an issue with all rooms and all low frequency alignments I’ve seen (ported, sealed, horn and OB). Again, placement of speakers and treatments will help here too.
4 inch absorbers, nothing less if you are going this route.
It will be interesting to see the waterfall. Set it to 450 msec and only 20Hz to 400Hz range. There may or may not be some tails where a frequency range lingers, especially below 100Hz. OB does better at this than most. Spectrogram will show the same thing, just a different way of presenting the waterfall data.
All that being said, if it sounds good, then you can leave it alone.
Best,
Anand.
-
You still didn't say what the distance is between the speakers.
IME if you have less than 6 ft between speakers you need to be within something like half the distance between them for it to almost work right. 6ft is my minimum under all circumstances. I would rather push speakers up against a wall than have them closer than that.
-
Inside to inside of the cabinets is 74”; center to center is 89”