Multiple Small Subs - Geddes Approach

[markus76]

Originally posted by soongsc

How do you identify the cause of difference to be room, background noise, or speaker characteristic, etc. by just looking at these?

I don't understand your question (as always). We're still talking about low frequencies, right? If you're able to do a waterfall measurement then you should be able to measure background noise or the near field frequency response of a woofer as well.
I encourage everybody to download REW and play around with it. It's free and you'll learn a lot.

Best, Markus

[soongsc]

Originally posted by john k...
Sorry if you feel I am picking on you, but you are representing yourself up as the expert and people read and listen to what you say. Thus it is more important for you to be correct, and not make brash statement like... no sound....

Also I have asked more than once in this thread (both recently and some time age) how a 3-woofer system should be set up in a hypothetical, ridged walled, sealed, rectangular room so I can post comparative simulations of the 3 woofer, 1/4 floor mounts 4 woofer, and DBA formats. I've gotten no guidance form you. My simulation attempts with 3 woofer have not been very successful.

I know, it's not real world but I still believe that such sims should define the fundamental physics of the problem. So excuse me for nit picking. It only because I know you know better.

Originally posted by gedlee



...
This is why I'm not going to get into a numbers game with people on this topic.

Well, since john k seems to have the tools and willing to do some simulations, and you seem to have an exsiting room arrangement you use as a reference, why not just use your room and sub setup as a basis for simulation and see how things come out?

[soongsc]

Originally posted by markus76


I don't understand your question (as always). We're still talking about low frequencies, right? If you're able to do a waterfall measurement then you should be able to measure background noise or the near field frequency response of a woofer as well.
I encourage everybody to download REW and play around with it. It's free and you'll learn a lot.

Best, Markus

Yes, this is still talking about low frequency. How would you subtract noise from the waterfall. Does the REW do that (I assume you have used all it's features)? Then, how would the absorption factor spectrum be generated (assuming we are still on the low frequency absorption issue)?

[markus76]

Why would you want to substract the noise floor? Either your measurement signal is far enough away from the noise floor to generate meaningful data or it's not. A waterfall even let's you see the noise floor. Just play around with the settings in REW.
Don't understand the second part of your question at all. What is a "absorption factor spectrum"?

[gedlee]

Originally posted by soongsc
Well, since john k seems to have the tools and willing to do some simulations, and you seem to have an exsiting room arrangement you use as a reference, why not just use your room and sub setup as a basis for simulation and see how things come out?

My room is not sealed and very highly damped. Johns model doesn't apply.

[cap'n todd]

Originally posted by markus76



I encourage everybody to download REW and play around with it. It's free and you'll learn a lot.

Best, Markus

Markus, I did download REW. It's like ETF on steroids. Very nice. I like that the eq seems to be geared towards just bringing peaks down, and not filling dips. I still have the same reservations about waterfalls, and of course just because a measurement program looks slick does not mean it is correcly executed. There's a lot under the hood. I know because we at Harman are developing our own measurement platform, so I ahve some familiarity with it. There's also a lot of possibilities for window choices (for example) which do affect the result, and there is often no clear choice for which window is "correct".

It would be interesting to compare it to mlssa or TEF or B&K pulse or one of the other programs that have been around a while.

Still, very promising, and free!

Todd

[soongsc]

Originally posted by gedlee


My room is not sealed and very highly damped. Johns model doesn't apply.

Absorption factors can be assigned to the room boundaries, so it's really not an issue.

[soongsc]

Originally posted by markus76
Why would you want to substract the noise floor? Either your measurement signal is far enough away from the noise floor to generate meaningful data or it's not. A waterfall even let's you see the noise floor. Just play around with the settings in REW.
Don't understand the second part of your question at all. What is a "absorption factor spectrum"?

I guess that you might not have measured the characteristics of background noise for enough cases to get a feeling how they effect measurements. Even if you may see the noise floor at a certain point in time, if it's mixed with what you really want, then there is no way you can look at what really is necessary with enought time resolution. Additionally, if you monitor noise spectrum in real time, you will find that the spectrum content is not stationary, and may vary quite significantly over 10db depending on environment, this may invalidates whatever interpretation of the waterfall data one might conclude.

[janneman]

Hi,

I am building a quartet of subs to try the approach being discussed her. I have a slightly OT question, but I'm sure you guys can help me out.

The subs are downfiring 12 inch Peerless XXL's in a cylindrical enclosure. How far should they be above the floor? The only info I have found is someone saying that the area of the opening (the height above floor * circumference of the enclosure) should be at least equal to the driver area. What do you think?

TIA,

Jan Didden

[pedroskova]

Originally posted by markus76
Todd, I don't get it. Looking at this does tell you nothing??



Instead of oohing and aahing I would know that I have to do something against all those "anomalies I have, like "holes" and non-uniform decays". And after that a second waterfall would tell me if it was worth the $64k I spent.

Where was this recorded...in an empty warehouse? 430msec? Never seen anything like that.