Some frequency response specs for my room, with Klipsch RF3s and a Paradigm Servo 15. (1 Viewer)

[AaronNWilson]

Room size is 5250 cubic feet.
Speaker and subwoofer positionings based upon Russ Herschlemanns home theater architect series in SGHT.
20hz - 95dB
25hz - 94dB
30hz - 98dB
35hz - 107dB
40hz - 100dB
45hz - 101dB
50hz - 105dB
60hz - 99dB
All measurements were taken from my listening positioning about 14' from the front of the room.
They were also all taken at reference volume setting on my reciever. I currently have the servo 15 set at the -42 setting which I guess means something other than decibels, just some linear measurement scale.
I could crank the sub up higher, but any higher and it no longer blends with the mains, which I find is extremely important.
I noticed at 45hz the room boundary measurement was 115dB and BOY WAS IT LOUD.
I guess what these results show me is that I need a bass equalizer and I probably also need a second servo 15.
Everyone should download a tone generator for your computer and try this out, its lots of fun .
edit: i added a graph

Aaron
[Edited last by AaronNWilson on October 28, 2001 at 07:27 PM]

 

[Chuck C]

That is quite a large room, and looks like you got some great results. Out of curiosity, what are your room dimensions?
------------------
Chuck
Chuckster's HT Site
The At-Home Home Theater (E.L.)
The Dorm Room Theater (E.L.)

 

[AaronNWilson]

Thanks, but I was a bit disappointed that I don't hit 105 dB till 35 hz.
Room dimensions are 35' x 15' x10'.
Aaron
[Edited last by AaronNWilson on October 28, 2001 at 07:30 PM]

 

[Chu Gai]

what did you use as a mic?

 

[AaronNWilson]

I used the Radioshack SPL meter.
Aaron

 

[Guy Kuo]

Aaron, go download either or both www.spectraplus.com or www.etfacoustic.com software for your PC. You'll get MUCH faster and higher resolution audio measurements for your tweaking.
------------------
Guy Kuo
Link Removed
Ovation Software, the Home of AVIA DVD

 

[AaronNWilson]

If I use those programs will I have to get a good quality microphone, or can I use the crappy 10 dollar microphone that seems to be the standard issue on PCs these days.
Aaron

 

[Kimmo Jaskari]

I believe they can interface with your RS SPL meter.
Won't be as good as forking over that $100 for the calibrated mic they sell but should work.
------------------
/Kimmo

 

[BruceD]

Here's how I first used the ETF5 software. Later I bought/built a calibrated mic and mic pre-amp.
The RS SPL meter has an line level RCA output jack (mono) that you use to feed a line level input jack on your PC soundcard or notebook. This may require an adapter (single female RCA to mono 1/4" mini-phone plug) that you can get at radio shack.
The ETF5 software uses the RS meter's output as it's listening input (the software allows a mic calibration file for the RS meter).
The line level output from the PC soundcard goes to an input port on the amplifier (you adjust a single speaker's output one at a time).
The ETF software itself generates the sound signal and then listens to itself through the RS meter.
Bruce

 

[Wayne A. Pflughaupt]

Aaron,
A good 15” sub should be sufficient for a room this size (assuming that you including all other rooms, hallways, etc. that open to the HT in the volume measurement?).
That said, it is unusual to see an overall decrease from the highest to the lowest frequencies. For instance, based on your chart and figures, if you flatten out the humps at 50 and 35Hz, you have a significantly less output at 25Hz than 60Hz. For natural-sounding bass response, there should be a so-called “house curve,” a gently rising slope from the upper frequencies to the lower. Typically much if this gain is “free,” a function of the so-called “cabin gain” that comes naturally with smallish (comparatively speaking) home theater rooms.
I’m going to guess that you have your sub somewhere other than a corner? If so, moving it to a corner will significantly increase its output and extension. Of course, you will inevitably have one or two humps, as you have already noted. However, you can eliminate them with a parametric equalizer, and then have the best of all situations: the highest SPL, the lowest extension, and the smoothest response.
Cheers,
Wayne A. Pflughaupt
------------------
My Equipment List

 

[Allan F]

Bruce,
I wasn't aware that this could be done so easily. Does the ETF software actually include the mic calibration file for the RS SPL meter? Also, will the demo version of the ETF software perform this analysis or do we need to spend the $150 on the full version.
Allan

 

[Guy Kuo]

I use the ETF5 system as well. I haven't spent much time with it in demo mode, but I think it limits itself to two measurements/program execution in demo mode. It will do the work, but it would be very tedious relaunching and setting up the program.
In full operation mode, the program makes the sub tuning process ridiculously simple. You just set up the mic, load the cal file, start measuring and analyzing. I set it to display the low freq response curve and automatically repeat. That yields a new, finely detailed response curve every three seconds or so. As I dial in a parametric band center frequency, I set the filter to be a deep narrow notch. You can easily see thin notch move up and down the freq response curve as you vary the center frequency to land right at the middle of a problem peak. Then you adjust width and depth of the EQ filter to bring down that peak but not kill off adjacent frequencies. Start at the lowest peak and then move on to the next peak with the next band on the EQ. It so darned simple when you can actually see the response curve plotted out in fine resolution in near real time instead of guessing or making tedious, innaccurate manual plots. Once the peaks are tamed, I go back and rebalance the sub against the mains using an SPL meter. Overall subwoofer output and headroom increases for a lot of material which was previously buried by the bad peaks.
Doing this manually might get you in the ball park with a LOT more effort and time, but when you can do it so much more precisely and quickly, it quickly dawns upon you that it was shrewd to invest a few hundred on the analysis software and mic to double the performance of that $1000, $2000 or more sub.
------------------
Guy Kuo
Link Removed
Ovation Software, the Home of AVIA DVD

 

[BruceD]

Like Guy said (great explanation Guy), and I agree in spades.
And this setup is much cheaper, more complete, and easier to use for the home enthusiast than most of the professional RTA hardware systems out there.
I've had this system for almost 2 years and it has made the single biggest difference in the quality of my system, plus I've learned a lot about acoustics.
Allan, the demo doesn't let you load a mic calibration file.
Bruce

 

[Allan F]

Guy and Bruce,
Is the extra money for the calibration mic worth it or will the Rat Shack SPL meter with the calibration file do the trick?
Allan

 

[Guy Kuo]

I went straight for the calibrated mic and preamp so I don't know how well the RS SPL Meter works as a mic. For me it was worth not having to worry abou the directionality or flatness of response of the mic. The calibrated unit is designed to be omnidirectional and is almost flat from 20 Hz to 20,000 Hz even before the individualized cal file correction is used.
------------------
Guy Kuo
Link Removed
Ovation Software, the Home of AVIA DVD

 

[BruceD]

Allan,
Like Guy, I went for a calibrated mic, but I actually bought the mic and a pre-amp kit from a guy down at San Diego State University who has since graduated and left.
I paid $98 for the mic wand (with a calibration diskete) and the preamp kit of parts.
The mic and preamp I bought were the same as described here:
Link Removed
Because the preamp is battery operated, my complete setup is really portable with my notebook PC.
This combo works better for me than the RS SPL meter. I didn't record any differences between it and the RS meter (there were some even with the RS correction factors).
Bruce
[Edited last by BruceD on October 31, 2001 at 10:55 AM]

 

[AaronNWilson]

Well I ran the Spectra Plus analysis software and I don't think it is as good as everybody is letting on. Whenever I played a 180 hz note, there was a peak showing at 400 hz, 600 hz etc. Why would I want to know this?
Plus whenever I played a 20 hz note, it showed that there was sounds right up to 200 hz being heard. Why is this happening?
There was just far too many squiggly lines that were too sensitive to room noise to be able to really see what was going on. The whole setup was too sound sensitive.
Aaron

 

[Guy Kuo]

Those are probably harmonics of the original test signal. If the acoustics of the speaker and room were ideal a sine wave reproduced by the speaker would be picked up by the microphone as just the pure, original sine wave and you would only see the fundamental frequency. In an actual room, the waveform gets altered so it isn't the original, clean sine wave that gets picked up. This is part of the sound you actually end up with in the room. The mic itself may even add some distortion. At any rate, this change in the original wave shows up as distortion and you can see it as energy at multiples of the fundamental frequency. Another reason this can arise is if the input of the audio card is gettting too high of a signal level (aka overloaded). That also will show up as distortion.
You should be able to verify that this is an acoustic problem rather than a processing problem in the program by taking the electrical signal of the test tone and feeding it directly (via an line level reduction cable) into your sound card. That should allow the system to see the original, clean signal without the acoustic distortions created by the speaker and room. The would give you energy only at the fundamental frequency. The ETF program is designed to loop part of its test signal back through the audio card to compensate for changes caused by the audio card.
This doesn't mean you should discount those energy readings. They are part of what you are hearing from your system. As you optimize your speaker position, those harmonic distortions will likely drop in severity.
I'm a lot more familiar with the ETF package. So is Bruce. Have you had a chance to try the ETF5 program?
------------------
Guy Kuo
Link Removed
Ovation Software, the Home of AVIA DVD

 

[BruceD]

This is not a problem with the ETF software package, and it is also why I chose it over SpectraPlus (it was also much cheaper).
Basically, ETF generates it's own signal to a single channel/speaker (mono), and no demo or test CD/DVD is needed. This has many benefits.
After all the connections are made, you test calibrate the signal (generated and received) by setting volume sliders on the PC software volume control, i.e. setting the SPL levels to 75dB.
Because the program gives you complete control over the time domain, you can elect to show the harmonics (like SpectraPlus) by changing the time slice of data that is graphed. This is critical in evaluating early and late reflections in the listening room.
Hope this helps. I think everyone with an interest should at least visit the ETF website, as the tutorial/demo-room on speaker/room acoustics is very informative.
Bruce