Sub calibration is driving me NUTS! (1 Viewer)

[JohnnyG]

The main speakers in my home sound system are Infinity Interlude 60s. The IL60s are powered towers, each with a 12" 500 Watt subwoofer. I recently changed receivers from a Panasonic SA-HE200 to a Harman/Kardon AVR8000. I've set all the speakers to small, and the subwoofer crossover at 80Hz. So far, so good.

The subs in the IL60's feature Infinity's RABOS system and I purchased the RABOS kit and have made the adjustments.

To "dial in" the bass, I first tried using the test tones for this purpose on the RABOS CD and after playing the tones and setting the levels, I played some music. Where's the bass, I thought? To confirm the settings, I popped in AVIA and used the 2-channel analog output from the DVD player. I got a considerably different reading (~+5db) so I made the adjustments and played the same music again. Much better, but still not quite what I expected. There's a common misconception regarding the amount of bass you're 'supposed' to hear, so I just chalked it up to that and left things alone. I switched the receiver over to digital input and used AVIA to set the levels for Dolby Digital (the H/K remembers different levels for pretty much every sound mode and input).

I've watched plenty of DVDs since then and although the bass is definitely audible and sometimes even room-shaking, it still seemed like I was getting less than you would expect from TWO 500W subs.

Last night, on a whim, a decided to play the internal test tones on the H/K. To my utter amazement, those tones indicated a NINE dB increase in sub volume! We watched a couple of DVDs and now the bass is definitely there! I'm listening to music right now in the background and the bass is unmistakingly present. But is it correct, or actually too much?

Now I'm just totally confused! How can every single test I throw at it come up with different results?! What's most accurate? Did I have the right amount of bass before, or is it correct now? How the heck can I tell for sure?!

 

[Kevinkall]

Some discs will have you setup your speakers at 75db and some at 85db. I know my receivers test tones are at 75db. Have you checked to see what your receivers are set at or the setup disc you were using?

 

[Jeff Gatie]

Avia is attenuated to 85dB tones. If you were adjusting to 75dB with Avia, you wre 10dB too low.

 

[Richard Little]

I have all sorts of trouble calibrating the subwoofer with Avia.

Avia is true Dolby Digital and therefore allows the surround speaker in question to contribute significant amounts of bass to the subwoofer tone.

I have found the amount of bass each surround speaker contributes is highly dependent on the speaker design itself, and its room location. My left main is several dB higher than any other channel on the subwoofer tone. The best way to check all of the differences in the surround channels is to power down the subwoofer and run the Avia sub test tone for each channel.

With an 80 Hz crossover, it is pretty remarkable how much sound pressure the surround channel speaker can contribute to the subwoofer tone when using Avia. My left main contributes about 82 dB to the sub tone when it is calibrated to 85 dB with surround tone. This "surround speaker contribution" phenomenon can result in undercalibrating the subwoofer. For this reason most HT enthusiasts end up setting the subwoofer about 8 dB higher (93 dB) and wonder why they need to run it so hot when in reality they are not.

After calibrating the surround channels with Avia, I leave the Master Volume untouched and I actually resort to using the discrete test tones from the AVR as my initial sub calibration guide.

Here's my advice:

1) Calibrate all the surround channels to 85 dB with Avia.

2) Leave the Master Volume untouched.

3) Run all the discrete surround speaker test tones in the AVR and note the average volume. It likely won't be 85 dB, but they should all be about the same. Let's say for example, they are at 82 dB.

4) Run the discrete subwoofer test from the AVR and calibrate the sub level 3-4 dB hotter than the SPL noted in 3) above. In this example, you would calibrate the sub to 85-86 dB.

5) Now run the subwoofer test tone in Avia. I'm betting money it will show the sub running about 8 dB hot.

Regards,

Ed

This bit of info from Edward J M is very true and should help.

 

[Wayne A. Pflughaupt]

Johnny,

The Infinity RABOS system is a form of parametric equalization designed to take out (I think) one or two peaks. From what I’ve heard (and you seem to confirm) it does a good job of that.

Part of the problem is getting used to accurate bass. If you’re used to boomy, inaccurate bass, it’s a definite change. And, after you get accurate bass, it must be properly level-matched to get the most natural sound in your room. Typically the subs must run at a higher SPL level than the mains – an acoustic phenomenon known as a “house curve.”

Another part of the problem might be the powered subs themselves. It’s generally accepted that it’s best to have a separate sub (or two), because the speaker location that’s best for proper imaging is very seldom the best location for low frequency response.

That said, it would be best to get a test disc with 1/6-octave sine waves and an SPL meter and take some measurements. This is the only way you will ever know for sure what is happening with the bass in your room.

You might also need parametric equalization for the two subs (if that’s possible). For instance, you may need more adjustment than the RABOS system has available.

Regards,
Wayne A. Pflughaupt

 

[Steve_D]

Wayne,

I currently use an SPL meter and a test CD with 5 HZ incremental test tones. Combined with my Art, I try and smooth the low FR as much as possible, and integrate my sub with my mains as much as possible.

I am considering 'upgrading' my testing (ease of use and real time results if nothing else) and using a software based RTA and pink noise.

The above post seems to indicate the SPL meter is better?

TIA

 

[James Edward]

JohhnyG,
I posed a calibration question a few days ago and got an interesting response from a long time HTF member. Check it out below...

HERE

BTW, I also get different readings from VE versus internal test tones. Only a db or 2 difference on the mains, center, and surrounds, but also about 8-9db different for the sub. After reading the above response from Phil, I recalibrated using the receiver's tones. All seems fine, but the sub seems too hot, so I have it down about 4-5db in level from the other channels. Go figure. And I'm not one to shy away from bass...

 

[Brian L]

Avia is attenuated to 85dB tones. If you were adjusting to 75dB with Avia, you were 10dB too low.

Thats not really relevant. It just means he will need 10dB more on the volume knob to get to reference. In fact, the Avia tutorial advises setting reference to less than 85 dB.

If all channels are with a dB of each other, he is balanced.

FWIW, we all strive for a dead-nuts calibrated system, and then as soon as we hear a track thats either too bass heavy, or too low, we head straight for the sub level control.

And there is absolutely nothing wrong with that!

While a calibrated level is fine to start with, you really need to make your final adjustment by ear.

As Wayne said, you need test tones that will allow you to see (or in this case hear) what is happening in the room. You could have peaks at certain frequencies, and nulls at other frequencies.

Let me rephrase that:

You DO have peaks at certain frequencies, and you DO have nulls at other frequencies.

If a given film or piece of music has a lot of energy at those peaks, you will think that the bass is too boomy and turn down the level. Of course other material will sound weak, so back up it goes.

FWIW, I use the RABOS meter to adjust my Audio Control EQ (1/6 Octave below 80 Hz). I use the meter with a CD from Bass Mekaniks that has tones in 1/2 hertz intervals. They line up perfectly with all but one or two sliders. And they produce as steady a tone (as measured on the RABOS meter) as I have had in my room.

BGL

 

[JohnnyG]

Thank you all for the information but it still leaves my main question somewhat unanswered...how can I tell for sure that I'm running the proper amount of LFE?

I think RTA might be the only way to go, really. At least for 2 channel reproduction. I downloaded, but have not installed or run as yet, some RTA software from here:

http://www.allen-heath.co.uk/rta.asp

Does anybody have experience with this?

I guess the other point I didn't realise is that the AVR8000 is THX Ultra certified, so it's test tones are supposed to be tested accurate as part of the THX certification process, correct??

 

[JohnnyG]

To answer one of the other questions, the IL60s are powered towers, but you can 'decouple' the sub and run it as a totally independant sub/sat and this is the way I have it configured. The main part of the speaker is set to 'Small' and the subs are fed off the LFE output on the AVR.

 

[Wayne A. Pflughaupt]

Steve,

You DO have peaks at certain frequencies, and you DO have nulls at other frequencies. If a given film or piece of music has a lot of energy at those peaks, you will think that the bass is too boomy and turn down the level. Of course other material will sound weak, so back up it goes.

I was the same way until I correctly set up and equalized my subs and properly matched them with the mains. Once you do this and have confidence in the accuracy of your system, you know that what you’re hearing, for better or worse, is the program source. So now I don’t change the sub setting much, unless the “better or worse” is really offensive.

Regards,
Wayne A. Pflughaupt

 

[JohnnyG]

Any recommendations on inexpensive microphones to use for a software-based RTA? Your basic 'cheap' hand-held microphone seems to be in the 80Hz-12kHz range which I would assume in insufficient.

 

[Lee Bailey]

You need to be careful with sine wave testing, too long can burn your voice coils up.

 

[Wayne A. Pflughaupt]

Johnny,

I’m less familiar with RTA software programs than I am RTA hardware. The latter comes with a "calibrated" mic – a misnomer, actually, because no inexpensive mic exhibits flat response. It’s the RTA that’s calibrated to the mic.

That said, any RTA software vendor should recommend a specific mic to be used with their program. I believe with some you are able to tell it what mic you are using, and the software can be programmed to accommodate it.

Lee is correct about the potential danger of sine waves. They shouldn't be played at high volume levels for long periods of time. Fortunately, it only takes a few seconds to get an accurate reading with them.

Regards,
Wayne A. Pflughaupt

 

[Brian L]

Any recommendations on inexpensive microphones to use for a software-based RTA?

I dabbled with Spectra-Plus at one time, and it comes with calibration curves for various mics. You can also enter your own curve if you have a frequency map.

At that time, I used a Sure SM-57, which is a pretty standard mic in the recording world. Its pretty cheap too, about $100, if I recall.

Having said all that, I found doing this all with a PC to be a big PITA. I got much better, more repeatable results when I started using my RABOS meter with the Bass Mekanik disc previously mentioned.

I know that there are some users that are quite proficient with PC based measuring tools, but for me, I have decided that it is much quicker to just fire up a CD and write down the results on a spread sheet.

BGL

 

[JohnnyG]

Is this the Bass Mekanik "Power Tools" disc?

 

[Brian L]

The disc I use is called Bass Mekanik v5.0.

The test tones run from 20 to 99 Hz. They encoded it so that the whole frequncies (20 Hz, 21 Hz, etc) are in one channel, and the 1/2 frequencies (20.5 Hz, 21.5 Hz) are in the other.

Very handy, and as I said, for whatever reason, this disc produced the most stable readings that I have experienced in my room.

BGL

 

[JohnnyG]

On the RTA front, I downloaded SpectraPlus and it's occured to me that I can use my Radio Shack SPL meter as a microphone. SpectraPlus offers the ability to load a microphone compensation file and I'm sure I've seen this data posted here before for the Radio Shack meter. I did a search, but nothing showed up. Does anybody here have this info?

 

[Edward J M]

http://www.hometheaterhifi.com/utili...load-page.html

You will have to interpolate for numbers in between those values shown.

Other RTA software programs actually will measure the response of the mic and then provide a perfect corrected curve for it. That would be better than guessing per above.

If you have the time you could alter the RS meter per this link and it will then provide a WYSIWYG response down to at least 20 Hz.

Link Removed

Of you could simply buy a decent mic that provides a known unweighted flat response. Event the best mics have some response irregularities, but companies like Linear X will provide the actual correction factors for each mic and allow you to obtain a corrected response of 10Hz-40kHz +/- 1 dB.

Regards,

Ed

 

[JohnnyG]

Ed, that's fabulous information and thank you!!