rlw ???measuring reflections???

[HBombToo]

If you have the time could you please outline your test procedure and equipment list for your spl band sweep tests. I would like to put together a test settup that is not cost prohibitive but at the same time accurate within reason. Perhaps a quick study on sampling rates vrs accuracy to the real world would be helpfull.

Your insight here would be most appreciated.

Thanks in advance.

HBomb

[rlw]

I don't mean to rag on you, HB2, but since this is getting posted in the "basic" section, I don't want anybody to get the idea that they have to run out and start doing all this stuff in order to enjoy their system. I know that all the "regulars" here realize that, but there are a lot of registered users, and I don't want to chase a newbie away.

Also, I'm outlining the way I do system setup in a room - take it for what it's worth. I'll let you judge where I stand in relation to that line between a healthy hobby and an unhealthy obsession.

Measuring your system/room is not an easy task. Be prepared to be very shocked when you see the frequency response of your system for the first time. Also, you can't really measure reflections - you'll be measuring the direct sound as well

Sources: your hands, Stereophile Test CD's, and the Bob Carver/GRP "Amazing" CD.

Measuring devices: your ears.

Clap your hands. Listen to the sound of the clap. Does it echo? From where? Direct the sound of the clap towards a piece of furniture. How does it change? Direct it towards the interstection of ceiling and two walls - what happens?

Ok, now, empty your listening room of all furnishings, including your gear. Do the same thing. Walk all around the room, and get an idea of how built in fixtures resonate - listen to the way that heating vent "rings" when you clap! Listen to the slap echo as the pressure wave from your clap bounces from wall to wall.

Then, bring your furnishings back in one piece at a time. Do the same thing. Don't bring your system back in yet, though.


There's a set of test tones on one of the Stereophile CD's called MATT - the Music Articulation Tone Test. This is an "ear-ball": you listen to the tones, and by the way they change, you can tell in what frequence region you have reflection problems.

Now, take your furnishings back out and bring your system in. Walk around clapping some more.

Listen, by this time anybody who is in the house will think you're absolutely crazy. Don't let it bother you. This is where it starts to get good.

Now, play your system, and listen to some vocals. Listen to really simple music. Start getting an idea for how it sounds. Listen to some big bass. Run the MATT tones.

Now, start bringing your furnishings back in the room. What happens to all the sound now? How do vocal change? How does bass change? How do the MATT tones change? What about some warble tones?

Be sure to do this at different volumes. You'll start to realize that your room sounds different with different sound pressures hitting it. Using higher than normal SPL's during this time can reveal problems faster - everything will be exaggerated.

Once you've finished with this, you might want to measure - this will give you more data, for sure, and maybe help you out.

Meters: starting at the bottom is the RadioShack SPL meter. You can do some web-searching, and find out the compensation curve for the mic [its frequency response drops off significantly once you get out of the mid-band]. The big drawback to this meter is that you'll have to use the sweep tones, the steady-state tones, and the warble tones only - no broadband - and it will be difficult to get a good picture of your room. Likewise, the mic on the RS units are never calibrated - so you might be way off in your measurements.

Moving up: I used a third-octave analyzer with a calibrated mic. These can be pricey: on the order of a couple grand. The good news is that some higher-end shops will have them, and they might let you rent theirs. The one I used was a SoundControl SA3052 with a calibrated B&K omni mic. Drawbacks: 1/3 octave resolution might not be fine enough.

Once you get into this league, you can do broadband measuring, which is really helpful.

If I was going to do the measurement thing again, I'd look into a good quality soundcard for my PC, a software package that gave me maybe infinite resolution, and a calibrated omni-mic.

The calibrated mic is really important, as the last thing you want is a mic that is 15dB off at 80Hz! This is why moving to measurements is really expensive right away, at least IMO.

When you use your mic, I'd attach it to a tripod, and point it at the ceiling, and do a round of measuring. Then, point it left, and do the same round of measuring, then right, etc. Then I average the results.

I hope this gives you some ideas. Quite honestly, I don't think you have to spend much money at all to learn how your room behaves. The test CD's are pretty cheap - and those, your hands, and your ears are the most important pieces for this. I do think that it takes a lot of time - I'm serious about dragging all the furnishings out - when I was doing that, it took me a week or two to learn my room the first time, but it was really well worth it. I learned how differently two desks can sound, I learned how differently two wastebaskets can sound. All these little things in the room, contributing their sound, affect the final result of what you hear when you play your system.


After re-reading this post, I think that I'm just plain effin nuts.

[Edit: cleaned up some grammar. I'm still nuts.]