What's new

Video Essentials and sub calibration (1 Viewer)

Mal P

Stunt Coordinator
Joined
Mar 17, 2000
Messages
127
G'day,

I've never been clear about this particular issue, but using the Radioshack SPL meter and VE, should I calibrate my sub to 75dB or 78dB, due to the roll-off the Radioshack meter exhibits?

Thanks,

Mal
 

BruceD

Screenwriter
Joined
Apr 12, 1999
Messages
1,220
Now that's a good question.

Really depends on what you are really measuring with the SPL meter. I think the point was made that shaped pink noise (20Hz-80Hz) like that used in the VE DVD would register a few dB light on the SPL meter, ergo the need for the 78dB calibration level.

But, if you happen to have a room-modal peak in the 20Hz-80Hz range that reinforces specific frequencies you may get inaccurate SPL readings anyway. In this case neither 75dB nor 78dB will yield desirable results.

To me, this means your sub/room frequency response needs to be somewhat flat (without say 10dB peaks) before you calibrate with the shaped pink noise to 78dB.
 

Steve Stogel

Supporting Actor
Joined
Jul 19, 2000
Messages
600
Bruce, if the VE pink noise for the sub is a few dB light, wouldn't that mean that 72dB would be closer to a true 75? And that if you wanted, say, a 6dB bump in the sub, you'd want to calibrate it to 78dB? This is a good question, and I hope somebody can help out with this. I'm running it at about 78-80dB, so is this 6-8dB hot?

Steve
 

Kimmo Jaskari

Screenwriter
Joined
Feb 27, 2000
Messages
1,528
I think what Bruce is saying (if I may be so bold as to interpret!) is that if your room reinforces say, 60hz, your meter will pick that up as being the loudest sound and show x amount of db. All other frequencies will be too low if you calibrate to 75 because the meter got swamped by that 60hz peak.
What one needs to do is get a hold of a bass sweep sound and measure bass output across the range, ie from 20 up to 100hz or wherever your crossover may be set. That should give you an idea about how flat your output is in your room.
Should also give you info on where to place the sub to minimize the peak problems, and if it's too bad you know you'll need a parametric equalizer for the sub channel to tame the signal! :)
At least that would be my opinion.
 

KevinHunt

Stunt Coordinator
Joined
Aug 31, 2000
Messages
121
Mal, Steve's point is correct. The Radio Shack SPL meter(analog) rolls off 2-3dB down low like you said. So if you are calibrating your sub with a 75dB reading on the RS SPL meter, your true sub calibration is really 77-78dB. If you want your sub at a true 75dB like the rest of your speakers, you would be looking for roughly a 72-73dB reading on the SPL meter. But my suggestion is to run them a little hot :D
 

Jimmy P

Stunt Coordinator
Joined
Jul 1, 2001
Messages
82
Just as some of the previous post's have stated,if you set the reference pink noise level of the mains,center and surrounds at 75dB,set the sub approximatley 3dB to 6dB less,example: mains 75dB,sub in the 69dB-72dB range when useing a Rat Shack meter.I know alot of people really like to have the sub cranked up,its great to impress your friends,but the unrealistic bass wears on you after a while.
 

ling_w

Second Unit
Joined
Sep 3, 2001
Messages
426
One shouldn't go around guessing how many dB higher or lower to set the sub. The most important aspect of setting a sub's level is to have a even transition to the mains. You won't have to worry about rolloff of your sub, room peaks or meter rolloff effecting the SPL meter's reading. The way to do that would to run a freq sweep or a narrow band pink noise/warble tone and do a level match at the crossover freq (assuming the crossover is spec'd to be -3dB @ that freq. If otherwise, compensate for it.) Then run the same frequencies with both sub and mains running so you would get a smooth even transistion from main to sub.

Of course, if your mains are set to large where it does not cross over to the sub, the above method is moot.
 

Kimmo Jaskari

Screenwriter
Joined
Feb 27, 2000
Messages
1,528
Well, a good match with the speakers you have is essential, but that doesn't do much for room modes. Those you have to cope with either by using room treatments or with something like a behringer feedback destroyer.

I'm getting one of those myself, just haven't gotten around to it. I can't get the sub to sound good in my new apartment and it is because some areas of the low bass get excited by the room, and I need to damp those in the signal.
 

ling_w

Second Unit
Joined
Sep 3, 2001
Messages
426
If you adjust sub level based on the crossover point, the room mode adjustment (either by EQ, bass absorbers or placement of the sub) becomes a completely independent procedure. No need to think, that because the peak was reduced by 6dB, I should set the sub to 77dB instead of 78dB, then you think, my sub is down 3dB at 30hz instead of being flat to 20hz, now I might have to add 1.5dB to the reading, but then, the SPL meter is not picking up sound in the 20-32hz range, which is where the meter is most inaccurate, I will have to subtract 0.5dB....
 

BruceD

Screenwriter
Joined
Apr 12, 1999
Messages
1,220
Steve,

Kimmo hit the nail on the head and answered your question for me. Using a parametric EQ on the sub (cutting room peaks) is a quick and fast method to cleaner, deeper, calibrated bass.

ling,

The problem with your method is that basically the xovers in most HT receivers are not symmetrical. That means the actual slope of the high-pass and low-pass are not equal. You won't necessarily be able to get a good match at the xover frequency that is smooth on both sides of the xover. I know I tried this method for months. Then I bought an MLS acoustic measurement system, measured the speaker/room response and applied parametric EQ to the sub.

This is primarily because the high-pass typically has a 12dB slope and the low-pass typivcally has a 24dB slope. The high-pass is expecting a 12dB rolloff from the main speakers (which is only true with sealed speakers) and many ported speakers have different rolloff characteristics.

So, while I'd say your suggestion in theory is a good one, it is not really that good in the real world.
 

Mal P

Stunt Coordinator
Joined
Mar 17, 2000
Messages
127
Thanks for the responses everyone! So, it looks like, pretending room anomolies doesn't exist, calibrating to 75dB would mean I'm running the sub a little hot for films. That's ok, a little hotness never bothered anyone ;) I of course, turn it down -5dB when playing audio sources.
Cheers,
Mal
 

Steve Stogel

Supporting Actor
Joined
Jul 19, 2000
Messages
600
Thanks, Kimmo, for the clarification. I think that is what Bruce was saying, and that makes sense. I have never thought about that. My next question, though, is does anyone know what the VE sub test tone is Hz-wise? Is it 20-80Hz? Just 35Hz? Anybody know?

Steve
 

ling_w

Second Unit
Joined
Sep 3, 2001
Messages
426
BruceD,

I mentioned that you should compensate on gain based on dB down @ crossover freq. For L-R 4th order LP, you get it's -6dB relative to the octave or so before cutoff freq. For 2nd order HP, you get its -3dB point relative to the prior octave or two. There is some subjectivity to this method, since you have to plot your sub/main freq resp and visually eliminate the peaks and null so you could arrive at what -3dB/-6dB it is relative to. I am not assuming the speaker has a 2nd order rolloff with f3 @ 80hz. That is why I mentioned re-measurement with both main and subs running to ensure smooth transition.

Subs has been around for a long time, and that is the way it has always been measured since true sub integration was for the dedicated. Now HT has hit the masses, and they can't expect every person to go through all that trouble to set up a sub. That is why this simple pink noise method, along with the simple place the sub in the corner procedure has been devised.

You can't tell me that speaker designers that tries to integrate woofer/midrange or low woofer/woofer are using pink noise to match levels.
 

Users who are viewing this thread

Sign up for our newsletter

and receive essential news, curated deals, and much more







You will only receive emails from us. We will never sell or distribute your email address to third party companies at any time.

Forum statistics

Threads
357,037
Messages
5,129,285
Members
144,286
Latest member
acinstallation172
Recent bookmarks
0
Top