diyAudio

diyAudio (http://www.diyaudio.com/forums/)
-   Multi-Way (http://www.diyaudio.com/forums/multi-way/)
-   -   Designing and measuring a speaker for a given room. (http://www.diyaudio.com/forums/multi-way/95115-designing-measuring-speaker-given-room.html)

Padel 27th January 2007 09:54 PM

Designing and measuring a speaker for a given room.
 
Hello,

Let's face the facts....... Commercial speakers are based their performance on anechoic chamber measurements, that's simply because the manufacturers want to prove that there is a least one environment (anechoic chamber) that their products (speakers) perform as they should.

But...... we, the end users we are not having such chambers in our homes, so the end result of a speaker performing real good in an anechoic chamber could be really bad in a given room because of the modes a room can add to the end result.

Now let's assume that I want a pair of speakers for my GIVEN ROOM, would'nt it be more logical to use RTA measurements rather than quasi anechoic to evaluate the performance of my DIY speakers?

Opinions?

Spasticteapot 28th January 2007 04:22 AM

Most people use a Behringer active filter unit for correcting room response. If I could afford it, I'd bi- or tri-amp and use two or three of these for an active crossover.

Cal Weldon 28th January 2007 09:04 PM

Re: Designing and measuring a speaker for a given room.
 
Quote:

Originally posted by Padel
would'nt it be more logical to use RTA measurements rather than quasi anechoic to evaluate the performance of my DIY speakers?

Opinions?

Couldn't agree more. No point in doing your fine tuning in a shop when the livingroom is where they'll be. Implementation with the environment is highly underrated IMO. To me it's the second most important cog in the wheel after the speaker itself.

454Casull 28th January 2007 09:31 PM

Re: Designing and measuring a speaker for a given room.
 
I'd say most commercial speakers are based on maximum profit, rather than performance.

Padel 29th January 2007 02:23 AM

Thanks for the replies.


Here is the procedure I have in mind...........

1. Give the units a ''Heavy Massage'' for a couple of days to stabilize the Thiel-Small parameters.
2. Based on the measured (rather than published) TS parameters, calculate the optimum box volume.
3. Build the boxes, put them in place, have an RTA measurement to make sure the low frequencies (that are more sensitive by speaker placement) are as close as possible to a flat line.
4. Make a frequency response measurement of all individual uits using RTA (Microphone at the listening position) and build the crossover around these measurements.

Well?

Spasticteapot 29th January 2007 02:28 AM

Quote:

Originally posted by Padel
Thanks for the replies.


Here is the procedure I have in mind...........

1. Give the units a ''Heavy Massage'' for a couple of days to stabilize the Thiel-Small parameters.
2. Based on the measured (rather than published) TS parameters, calculate the optimum box volume.
3. Build the boxes, put them in place, have an RTA measurement to make sure the low frequencies (that are more sensitive by speaker placement) are as close as possible to a flat line.
4. Make a frequency response measurement of all individual uits using RTA (Microphone at the listening position) and build the crossover around these measurements.

Well?

If you're really going for the best possible speakers, just get a pair of these:

http://www.activemusician.com/item--...er%2FProcessor

As said earlier, bi- or tri-amp and use one of these for the tweeters, one of these for the woofers, and one of these for the mids (if you use mids.)

You can bodge together the speakers and tweak 'em perfectly.

TerryO 29th January 2007 04:06 AM

Quote:

Originally posted by Padel
Thanks for the replies.


Here is the procedure I have in mind...........

1. Give the units a ''Heavy Massage'' for a couple of days to stabilize the Thiel-Small parameters.
2. Based on the measured (rather than published) TS parameters, calculate the optimum box volume.
3. Build the boxes, put them in place, have an RTA measurement to make sure the low frequencies (that are more sensitive by speaker placement) are as close as possible to a flat line.
4. Make a frequency response measurement of all individual uits using RTA (Microphone at the listening position) and build the crossover around these measurements.

Well?

That's exactly the best way to do it if you are going to use a passive filter.

Spaz mentions another approach that is also viable, and may even be better, as you can use the gear on any other speakers that you may build in the future.

The choice is yours, although if you want to become a Jedi Master Speaker Builder, you should probably give the passive filter a try, if for no other reason than to learn how this stuff works. I use real cheap components, with the exception of inductors which are air core of decent guage (because you can hear the difference). When you're satisfied with your prototype, then replace each component, one at a time (on one speaker, leave the other one alone, it's your control) and give a listen. Write down what differences between them you can hear at each stage, it might surprise you.

Whatever path you take on your quest, Good Luck !

Best Regards,
TerryO

Geoff H 29th January 2007 12:39 PM

I have had a couple of days doing just that. With RTA or any other software, save the screen shots with descriptive names in directories so you can back track or otherwise compare notes. And be careful not to do any tuning of the cabs just to suit the room. Moving the cabinets a few inches made more improvement than port tuning to fix a sharp rise.

Good luck with it, its worth the effort.

Geoff.

Edit. Multi tapped mains transformers make a cheap substitute for large inductors in the setting up stage. Once the freq is established, replace with the required value.


All times are GMT. The time now is 06:16 PM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright 1999-2014 diyAudio


Content Relevant URLs by vBSEO 3.3.2