works for me, others may have more or better ideas
30 responses Add your response
The standard tools, like the SPl meter from radio shack are fine.
A better way IMHO is to get yourself an iPhone and buy the app for SPL meter. You can also buy a Real Time Analyzer application which will tell you frequency response from 20-20K.
Now, this method is probably 10% less accurate than above average measuring equipment, but it will put you squarely in the ball park, and it is on your phone!!!!!!
I have an Audio Control RTA which I use to measure room response for clients when needed, and I have a multitude of SPL meters, but I use my iPhone 10x more because it is SO easy.
Most of the time when taking these measurements a 10% deviation from exact is not a big deal anyway as long as you use the same device to gather all your data.
Primitive and tedious.
The optimal cost/precision/effort way is to use RoomEQ Wizard freeware with your laptop. You can get away with RS meter for this but a better mic is advised.
This will give you measurements of swept tones (without resorting to pen-and-paper or spreadsheet entries) and will also provide lots of other info.
One problem is that there is no since "usual listening position." Moving the mic or the SLM only a few inches can change the FR substantially, especially if it is handheld. Using REW, one can take several sweeps from different positions within the listening position in a matter of minutes and average them.
I guess it depends on whether you want to know something or just get a feel of what's going on.
I'm with Kal - I use Fuzzmeasure and a measurement microphone - examples are on my virtual system. Even with a tool like this it took me several days to make adjustments. A good understanding of physics helps but is not a prerequisite to using tools like Room Eq Wizard (provided you read up on the subject).
If your idea of system tuning is swapping out your $500 speaker cables for $1,000 speaker cables in order to match the brand of your $500 interconnects then simply ignore this post. (There is absolutely no point in taking measurements if you are not prepared to take appropriate action to do anything about it - in that case it may be better not to know and you'll have more fun spending your money on some cable elevators instead)
Bob I suppose, one could borrow a girl friend's laptop or just 'a' friends, over the weekend or ask them to drop by for a couple hours and bring it along..
If one goes with a mic... any suggestions there?
== = = = = = = = = = =
Do you need to afix some tube surrounding the mic.. or just how fancy does one need to get in this area. Won't a tri pod for the mic work just fine?
OK... so you get the software, mic/RS meter, Laptop, and take every measurement known to man and some as yet unknown perhaps... What then?
I can appreciate the input here and elsewhere as to attending to bass irregularities, and reflections, using traps and diffusers. What physical paraphernalia, apparatus, and so forth are used thereafter to attend to the frequency range above 200 - 300Hz??
Seems to me with the addition of physical structures, DIY stuff, etc. at best its a guessing game then, deciding which, what, where and how much of this or that.
Wouldnt using electronic means to adjust for the anomalies be the best and shortest path to a positive solution in the acoustic treatment realm? As much as others seem to preclude such added componentry to the signal If one doesnt want to clutter up the listening room with all these sonic attenuators & diffusers which are a non exacting process and far from one of simplicity, it would appear at least a more succinct and quite pertinent resolution for the majority of listeners without looking into it too deeply.
Are these electronic EQs such poorly voiced or non transparent, devices, or so immensely over priced components, most audio nuts will shy away from them and pursue the haphazard mechanical avenue instead?
Or should the entire idea of using electronic equalization be revisited, surely, at some greater length?
BTW just how musical is a flat room anyways? Not owning one, nor knowingly having not been in one, I have to ask/wonder.
The Idea of flat is that you get all frequencys at the same sound level. That doesnt mean thats the way Music is recorded. I have a Mcintosh A2 it has a Fixed output pink/white noise or specific 1/3 octive test tones 20 to 20k
I use it to find the best speaker placement in my room and check for bass issue's. I do each speaker independant at 1 meter then do mono at my listening spot
There is no simple answer to the question about what one does with the measurement results. The first thing, of course, is to see if there is any correlation with your long-term subjective assessment of system problems. If so, then those need to be dealt with. How? Physical treatments (see www.realtraps.com for ideas) and physical setup.
I reviewed and still use the XTZ system.
There will be a follow-up on the new version in November.
Behringer ECM 8000 with an XLR cable is cheap and works fine. There are better measurement mikes out there but for room evaluation this is more than good enough. You will need a microphone stand also and a USB interface to your microphone (it uses phantom power) - I'd suggest the EMU 0404 USB - again I am recommending excellent gear but at modest cost (great value).
Kschiu, I did borrow a friend's RTA to verify my general tweeter level setting. That gave a reasonable overall response curve but was lacking in detail (the measure bars move slightly with the pink noise).
For a more detailed adjustment of speaker placement for smoothest bass response, the RS SPL meter (which I calibrated at work against a B&K SPL) and Stereophile Test CD (any of 1-3 will work) did a reasonably good job.
I placed the SPL meter on a photo tri-pod, positioned where my head would be in the prime listening location. Then with a volume level of about 85 dB I ran and recorded the reading for 1K (reference sound level), then each of the tones from 200 Hz on down to 31 Hz (lower frequencies are notably incorrect on the RS meter but we are looking at relative measurements here so constant errors should be acceptable) on the test disk. I then added the total deviation from the 1K reference for all bass frequencies (ignoring + or -) for a given speaker location. This supplied a rough picture of the smoothness of the bass response. I continued making such readings as I moved the speakers out from the front wall. The response (deviation) became flatter as I moved the speakers out to a given point, then they began increasing as I moved them further. This was time consuming since I moved the speakers 2" at a time until I found the range with less variation, then I moved them 1" at a time. I was quite satisfied with the results.
Pryso- That will, indeed, give you a very rough estimate unless you sample and average several locations in/near the main listening position. The reason is that the FR, especially in the bass, can vary substantially over a span of several inches (similar to the distance between your ears) and no one sample is entirely representative of the sound at the main seat.
Has anyone tried this item?
>Phonic PAA3 Handheld Audio Analyzer w/USB Interface
I had a dealer use this in conjunction with his laptop to take measurements.
Wondering how this compares to the XTZ analyzer?
Kal, thanks for your comments. I attempted to be as "rigorous" as possible with this somewhat crude method.
First I placed the SPL meter at the location where my head would be with normal listening. Obviously this required moving the listening chair to position the tri pod. Also I measured my ear hight from the floor when seated and set the meter at that same elevation. BTW, my RS instructions say to position the analog meter at 90 degrees from the sound source. A friend with the same model meter (but probably newer by several years) has instructions to point the meter at the source!?!
Then as I moved my speakers in 2" increments I found one distance where the aggregate deviation from the 1K reference was the lowest. So I then moved the speakers forward and backward by 1" from that point, measuring each. This was like focusing a telephoto lens and allowed me to position my speakers within an inch for smoothest response. I found this more satisfactory than I had been able to position by ear previously (although admittedly I had not tired so many positions).
Given your point on the small distance variations, I don't see how any other process could provide a better approximation.
K, you have received good advice so far.
I would like to add that the most musical in-room response is not a straight line in the treble range. Most listeners perceive a flat treble in-room response as bright-sounding.
A gradual roll-off starting around 12 KHz is correct for off-axis, far field listening in a room with standard furniture / carpets / curtains.
By off-axis I mean slight toe-in, as recommended by most manufacturers.
Adjust speaker tilt and toe-in to achieve a gradual slope, with no major peaks or valleys.
I hope this helps
To do this properly you will need software (RoomEQWizard works fine for "snapshopts", but I prefer TrueRTA as I can see the changes visually as I move things around), and a properly calibrated microphone along with a mic preamp.
If you are taking this project seriously, don't use the RS SPL meter, as it is well known that it is not accurate.
I use a portable phonic PPA3 RTA and achieve consistently good results.
Its built-in microphone is calibrated, you can store the data in your computer through an USB port and calculate the average of several room response graphs (just like John Atkinson does in Stereophile room response meausurements).
It comes with 31 band spectrum analyser, SPL reading, EQ setting calculations/suggestions, RT-60 calculator, phase checker, a built-in test tone generator and a test CD.
Last time I saw it cost U$ 500, a great investment if you really want your system to shine.
The learning curve is quick, in 30 minutes you are set to go.
In most systems I can find a much better loudspeaker positioning after 3-4 readings.
I tried the PAA3 and I currently use Fuzzmeasure. I find the sweep of PAA3 takes too long and very susceptable to ambient noise. Fuzzmeasure allows 1/3 octave to 1/48 octave smoothing with RT60 calcutalions at 1/3 octave. The sweep time and freq range is also adjustable. The curves from each measurement can be superimposed on the same grap so you can see the result of each change you do to the room/speaker.
PAA3 is useful in other ways and very portable. But for frequency response curve and reverb measurements, Fuzzmeasure is more precise and informative