SPL measurements, nitty gritty

burdette

I burned a new test tone CD yesterday... if I had more use for one, I'd be really tempted to purchase an audio/video test disc, more extensive than the "home theater setup" discs out there (such as my Sound & Vision one). I'm getting pretty severe popping at the beginnings of these tones - to the extent that I don't turn it up very loud. I've burned MP3s to CD before and played them just fine, so this isn't an equipment issue.

That said... on to specifics of the task. I read the RS manual but questions remain.

Specifically, how/where do you hold the meter to take measurements? In front of your face? Next to your face (which is what I think the manual says)? Out in front of you? At 'head' position in the main listening seat? Do you point the microphone directly at the speaker or at the screen?

What reference level do you use? I think I remember Doc saying you used 100Hz, to reference off your mains.. what level? I'm usually doing this at night with kids in bed, so I can't be TOO loud. If I need to turn it up, I'll have to make a plan to get it done during the day.

Any other details are appreciated.


The disc I used last night had 10-12 sec tones starting at 10Hz up to 106 (the CD burner software wouldn't allow me to add any more tracks, so I had to skip 107-125). The volume was NOT loud at all... in terms of accurate measurements, I think I've been defeating myself by using too low a volume.

I didn't write down numbers last night. At the low volume (80dB range) I didn't 'hear' anything until about 15 Hz.. and then I thought I was getting *extensive* port noise and that my enclosure was rattling apart. Nope.. my basement walls had become transducers.. my home-built DVD/CD shelves were buzzing. At 18-19Hz, I get a lot of wind out of the port, but no noise (On paper, the tune is 18.1). The volume *seems* to be quite a bit louder once I get to the mid-20s... but I know that loudness is harder to compare subjectively at low frequencies.

This thing definitely is NOT booming. I listened to parts of Monsters Inc and SPR last night... the low end is solid and stable, very natural sounding. I bought a twin-sized egg-crate mattress pad this past weekend and redid the entire lining - definitely a better choice than a couple of layers of batting. That did prompt a new question... one that is moot, I suppose.. can a real man have a subwoofer with a pink lining? I believe I've seen photos of the interior of some early SVS subs.. looked like pink fiberglass.

Oh, happy birthday to me.

Dr. Spec

The SPL meter should go EXACTLY where your head normally goes on the couch when you are watching DVDs or listening to music.

I use a cheap camera tripod with all the usual adjustable legs and angles, and jam it into the couch cushions and make it just right. The screw hole on the bottom of the RS meter is FOR a tripod set up and I recommend you beg steal or borrow one - it makes life much easier when you are doing multiple calibration runs. It should be facing forward at the center channel, with the head set at a 45 degree angle up from the horizontal.

First calibrate your entire system before you do your bass sweep. Set all your speaks to small, set the filter point to 80 Hz, set all the internal speak settings to 0, set the internal sub setting to -3, run the sub cable from the AVR to the PE amp, set the PE amp filter to its highest setting, set the PE amp phase control to 0.

Most AVRs default to Master Volume 00 (normal Reference Level) when the test tones are engaged. If your AVR does not - no big deal. If it does not, start the test tones and ramp up the Master Volume until the center channel reads a steady 75 dB on the meter on C-weighted Slow. Record this Master Volume setting - it is your Reference Level setting.

With the Master Volume still at the same setting, go to the other test tones for the rest of the speaks and adjust each one of them to 75 dB also.

For the sub test tone, leave the setting at -3 and adjust the PE plate amp until it also reads 75 dB. Since the RS meter reads a little low on the test tone, your sub will actually be calibrated about 2 dB "hot", which is a good starting point.

Now that your system is properly calibrated, you can move onto the bass FR sweep.

Select the highest test tone frequency in your sweep (106) and start playing it (the pops between tracks are normal). Adjust the Master Volume until you get 80 dB on the meter on C-Weighted Slow.

Then let the sweep run from 10-106. You may have to set the SPL meter scale down into the 60 range to get anything to read in the 10-15 Hz range. As the sub starts to ramp up on line, up the range to 70 and then back up to 80 as needed. You won't hear anything until the sub hits about 18 Hz, but the meter should be reading it.

Record the raw data at each tone and then input it into your spreadsheet to accomodate for the CFs and voila - your sub FR will pop out on the chart.

Unless you want to buy a new driver, DO NOT run this sweep at anything much over 80 dB. It is too long and the VC will melt if you try it at 100 dB or something insane like that.

As far as the meter being off at low volumes - it's not. Just set it to the appropriate scale at the really low stuff and then go back to the 80 dB scale when it comes fully on line (probably around 16 Hz). The CFs will take care of the inherent inaccuracy of this meter when you plot the data in Excel. I will say this in closing, though - the validity of the CFs below about 15 Hz is questionable and I would take that part of the corrected curve with a grain of salt.

Good luck and please post either the raw data for the corrected data so I can plot it and save it - I really like to keep track of this stuff to gauge how our design specs out against the actual performance.

Doc

HBombToo

I would like to add something in the form of a question.

I don't believe that a high SPL, (~85dB verse 70-75dB), for the response is necessary? I figure as long as you don't exceed or dip below the measurement sensitivity of the meter the response should be valid...

Any thoughts?

HBomb

burdette

Thanks for the information, Doc. Sounds like I was pretty much on track, but had no experience with which to gauge that.... I was in the 75-80dB range but thought I was too low.

I will try to get some 'formal' and accurate readings done soon and will post or send you the data.

We have a tripod.. got it when my daughter was born and my wife filled a roll of film and a VHS-C tape per day ... at least it seemed like it when I paid for developing!

Dr. Spec

Originally posted by HBombToo
I would like to add something in the form of a question.

I don't believe that a high SPL, (~85dB verse 70-75dB), for the response is necessary? I figure as long as you don't exceed or dip below the measurement sensitivity of the meter the response should be valid...

Any thoughts?

HBomb

That is totally true Henry. And I said as much above. I just like 80 dB because it's satisfyingly loud without causing the sub harm. You could just as well run 75 or 70 (as you stated) providing you kept the meter in the appropriate range.

Ed