Home Theater Forum and Systems banner
1 - 20 of 91 Posts

·
Registered
Joined
·
101 Posts
Discussion Starter · #1 ·
As a baseline step, now that I have 2 speakers and 2 subwoofers hooked up, I ran some REW measurements in my (very) RAW shell of a home theater.

In the process of trying to do the level calibrations, I think I've concluded that one of my subwoofers must be defective as it seems to be bottoming out uncontrollably. The other one was a little more behaved and these initial measurements are with just one subwoofer.

The subwoofer in question is a passive JTR Captivator (ported) being powered by a Behringer iNuke nu6000 with a built in DSP but will all the DSP settings bypassed. The AV processor was set to a crossover of 200 Hz (the highest it would go).

Please take a look and let me know what you can glean from the squiggles....
 

Attachments

·
Premium Member
Joined
·
1,923 Posts
Is your mic calibrated for SPL? Your sweeps appear to be at an unbelievable SPL level. Are you sure your gains are set correctly through your AVR and the inuke? I would point towards the signal coming in, not the sub itself, for "uncontrollable bottoming out."
 

·
Registered
Joined
·
101 Posts
Discussion Starter · #3 ·
Great question as I wondered about the absolute numbers as well. But my assumption was that the relative levels and waterfall chart both contain some useful information.

As far as leveling, I did two leveling steps in REW:

1) On the Soundcard Properties page, I did the level check using the subwoofers. The "reference" was -12 dB. The information box suggested a measured level between -12 dB and -24 dB. I got the subs to about -20dB and it got too loud and rumbly.

2) Using the SPL meter, I did a level calibration and the display put up a certain number and I set the input dialog box to about the same number.

Then I ran my sweep. Is there something I overlooked as far as level calibration?
 

·
Registered
Joined
·
101 Posts
Discussion Starter · #5 ·
My amateur attempt at an analysis of this first measurement of my RAW room:

The frequency response of the sub is pretty even from very low (15 Hz) to about 70 Hz. There it starts a pretty significant rolloff with a pronounced null at about 110 Hz.

I have a question about running the waterfall measurements, though. In this graph, I have set range of measurements from 60-150 dB. There is a peak dB measurement of around 135 dB in the range I'm measuring. So that gives me a measured range of 135-60 = 75 dB. Is that appropriate? I would think that after that much of a reduction the audibility would be questionable. I think I read somewhere that a range of 40 dB is appropriate. If this is so, my graph may look a little bit better. I can re-run the waterfall with that range and see for sure, though.

So my course of action at this point would be to set my subwoofer crossover for 80 Hz which works out well for my main speakers. And then work on bass trapping strategies and subwoofer placement options to minimize the "ringing".

From a subjective perspective, I think the sound is terrible. There is very little visceral impact from the subs. I would have expected a punch in the gut sensation with subwoofers this hefty. It's unclear to me why the bass seems so weak despite a pretty nice frequency response measurement.

I will also re-run this tonight with both front speakers and the subwoofer and see if that changes things.

Anything else I should be looking at now to help me plan the room's acoustic tuning? Another next step would be to do some full range measurements of the main speakers. Speaking of which, with a 7 channel setup, would the recommendation be to run the 3 fronts as a set separate from the sides and rears? Or all 7 together?
 

·
Premium Member
Joined
·
6,711 Posts
2) Using the SPL meter, I did a level calibration and the display put up a certain number and I set the input dialog box to about the same number.
For SPL calibration the number you need to enter should be read from an external handheld SPL meter, not REW's meter - REW's meter doesn't show correct figures until after it has been calibrated against some external reference.

Since you are using a UMIK-1 though you need to use REW V5.01 beta 17, the UMIK cal file (as supplied by MiniDSP) has a sensitivity figure in it that REW can use to calibrate the readings directly from the mic.
 

·
Registered
Joined
·
101 Posts
Discussion Starter · #7 ·
John, thanks for the input. I was able to gather that I needed the latest beta version based on the equipment that I was using. So after downloading and installing the latest and greatest, things made a lot more sense. The ability of the software to identify the UMIK-1 and ask for its calibration file is awesome. And that was all that was needed as far as leveling and calibrating. Amazing. Thanks for the great software. I think it has really helped me already and I'm just starting this journey.
 

·
Registered
Joined
·
101 Posts
Discussion Starter · #8 ·
So after getting my software updated and letting it hook up to my UMIK-1, REW showed me some very interesting things. Including why I was having such a hard time getting any reasonable sound from my subwoofers.

I took these measurements with both the subwoofers and the full range L/R speakers active.

IMAGE 1
First was a baseline with only one subwoofer and all 3 channels being driven by my Sherbourn amplifier. I wanted to take the dedicated Behringer iNuke amplifier out of the equation in case something in the DSP module was screwing up my sound. I still had the pronounced rolloff starting around 70 Hz.

IMAGE 2
Despite changing the crossover point in the processor, I always had a big dip around the same place.

IMAGE 3
I finally found a combination of crossovers (lowering the frequency of the main L/R crossovers and raising the frequency of the subwoofer crossover) that made an improvement in that dip.
 

Attachments

·
Registered
Joined
·
101 Posts
Discussion Starter · #9 ·
Next, I switched the sub over to the Behringer iNuke amplifier and found essentially no difference. That was reassuring. The sound quality was quite a bit better even with this little tweak.

And then I started playing with the parametric EQ in the Behringer amplifier. At the end, the frequency response curve was quite a bit better.
 

Attachments

·
Registered
Joined
·
101 Posts
Discussion Starter · #10 ·
IMAGE 1
The most interesting thing happened when I added the second sub. No EQ was applied to the second sub at the start. Obviously there was a lot of cancellation going on which was especially evident around 20 Hz and between 50-90 Hz.

IMAGE 2
As a first test, I reversed the phase of the second subwoofer and things miraculously got significantly better.

IMAGE 3
A little bit of EQ tweaking and I was able to tame the new peak around 30 Hz and improve the dip around 80 Hz.

IMAGE 4
And this is what I ended up with. I raised the gain of the subwoofer amplifier a little bit and this really started giving me the visceral punch that I've come to appreciate and expect of a home theater sound system. A little exaggerated but I had a good time listening to some music sources tonight with this setup. I'm sure that when I finalize things I'll opt for a more balanced bass/treble setting, but for now the improvement in sound was dramatically improved. Plus no more bottoming out of the subs.
 

Attachments

·
Registered
Joined
·
101 Posts
Discussion Starter · #11 ·
I think I followed a logical progression of measurements, changes, re-measurements. As this is the first time I'm doing this, I'd appreciate suggestions on ways to do things better.

I haven't even looked at the waterfalls yet!
 

·
Registered
Joined
·
101 Posts
Discussion Starter · #13 ·
Taking a quick look at the waterfall graphs, they look identical. I'm not sure if I waterfalled the wrong FR graphs or if EQing has no effect on waterfall graphs?

Also, another question. The FR graphs generated seem to have useful information up to about 100 Hz. Then they just become a jumble of squiggles. Is this typical or is it a sign of some issue I need to look into?
 

·
Super Moderator
Joined
·
9,310 Posts

Sneak preview of the before and after waterfall. I have to study a little more to figure out what this is telling me...
Not exactly on your topic, but there should be enough info in this post to give you an idea of what waterfalls are about.


Taking a quick look at the waterfall graphs, they look identical. I'm not sure if I waterfalled the wrong FR graphs or if EQing has no effect on waterfall graphs?
Yes, the two waterfalls are identical. They are more meaningful with the graph floor reduced to the ambient noise level of the room, which is typically 35-40 dB. That said, it looks like your low frequency dampening is excellent.


Also, another question. The FR graphs generated seem to have useful information up to about 100 Hz. Then they just become a jumble of squiggles. Is this typical or is it a sign of some issue I need to look into?
The upper frequency raggedness is merely comb filtering caused by reflections. It shows up more in the upper frequencies than the low, which is why we recommend subwoofer graphs be presented unsmoothed. Fortunately, it looks worse than it sounds. You can use 1/3- or 1/6-octave smoothing for the graph to get a better representation of what you’re actually hearing.

Regards,
Wayne
 

·
Registered
Joined
·
101 Posts
Discussion Starter · #15 ·
Thanks for the great input. I'll check out the waterfall reference. I also have to read up on the ETC as I start to look at the main speakers. Now that I know that the subwoofers are not defective, I feel much better hooking everything up!
 

·
Registered
Joined
·
101 Posts
Discussion Starter · #16 ·
One thought looking at the waterfall graphs above. There appears to be extended decay times at 4 frequencies in the subwoofer's range: 17 Hz, 35 Hz, 70 Hz, 100 Hz. Do I even have to worry about the 17 Hz one as it is below the audible frequency range?

I'm amazed at your "well damped" comment. Currently there are no significant dampening elements in the room. I should get a picture of the room as it stands now.
 

·
Registered
Joined
·
486 Posts
17 and 35 Hz would require some very big absorbers or one tuned to those frequencies. Tuned absorbers don't always work correctly.

You may want to start looking at if the subs are time aligned with the mains and each other. That may impact your results.
 

·
Registered
Joined
·
486 Posts

·
Premium Member
Joined
·
1,562 Posts
More info would be needed for me to provide any sort of timing help.

It is surprising that inverting one the SW improved the response. I would first want to understand the reason for that. What is the measured distance of each SW to the LP (6" accuracy is good enough)? I would want to see an .mdat with a sweep of each sub individually and together (SW1, SW2, SWs) to understand the situation. [I also would need to know if the settings were are with one inverted relative to the other or if they are now the same.]

What mechanism is there to adjust delays or distance settings for the main speakers; AVR, HTPC or other? Are the current delays/distances set to measured distances for all main and SW speakers?

Overall, it looks like the setup was coming together reasonably well so backing up and revisiting delay/distance settings in more detail may not provide any practical improvement so you may want to just continue forward using the current settings.
 
1 - 20 of 91 Posts
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top