I need help measuring the low end response of woofers... (baffle step issues)

[fjhuerta]

Hi all,

Well, the Natalie P's are finished (I'll post some pics later)... and I'm trying to build an MTM for myself, based on what I've been trying to learn. Anyway...

I'm having serious problems measuring low end response on my woofers. The way I've been doing it is:

1) Measure woofer with a 6 ms. gate, at 1 m.
2) Take a nearfield response sample at around 84 ms.
3) Splice and join both files at around 500 Hz.

So far, so good. I have what I guess is an infinite baffle frequency response graph.

Now, here comes the troublesome part. How on earth can I get a response that takes into consideration baffle step response?

I read Vance Dickason's book. I know I should take a measurement two meters away from my speakers, on the floor boundary. That didn't work. I suppose there are too many walls around the biggest space I have for that kind of measurement, and the results are swamped with reflections.

I tried using the KHF tool on LSPCad, and I'm even more perplexed. I get horrible looking curves when using the baffle diffraction model, and I have no clue as to how should I use the simple baffle step model (which frequency should I select).

I can't continue modeling my crossover, since I believe my measurements are completely wrong... what I did was...

1) Wire the woofers (WTW configuration) in parallel.
2) Measure at 1M, 6 ms.
3) Take a nearfield response measurement.
4) Splice and join both measurements at 500 Hz.
5) Use the KHF tool to model a baffle diffraction model (considering both woofers as a single source, emanating from the center of the baffle).
6) Use the KHF tool to model a simple baffle step model.
7) Model crossovers for both situations.
8) Compare both crossovers later on.

Somehow, I feel I'm doing things wrong, and this has to be so much simpler than what I'm doing... any ideas on how to properly take into consideration BSC?

Thanks a lot!

[mazurek]

Only way I was able to measure anything in my room, was pull everything into the center of the room.

[Jed]

One way is to import your gated measurements into the FRC tools and add in the baffle response. You will need to calculate min. phase and change your Z offsets after this though- if you are using SW.

However, I don't see the need to do all this for a 2-way. Your data is more than accurate enough to handle the crossover frequency range for the intended crossover point. It might be nice to know what is happening down low but it won't really affect the chosen crossover for most 2-ways.

Are you farmiliar with claudionegro.com? Check it out because that site pretty much describes all you need to know for measurments etc.

Jed

[Marzen]

Are you farmiliar with claudionegro.com? Check it out because that site pretty much describes all you need to know for measurments etc.
Jed

Thanks for the link to Claudio's site. It helped quell some doubts I had about my measurements, & pointed out some things I had glossed over. Nice to see a step by step documentation with photos.
-Ward

[Paul H]

Somehow, I feel I'm doing things wrong, and this has to be so much simpler than what I'm doing... any ideas on how to properly take into consideration BSC?

Thanks a lot!

What size room are the measurements being taken in? What are the speaker drivers mounted to/in?

[fjhuerta]

Oops, it took a bit of time to respond to my own post.

mazurek: Thanks for the tip. I always measured so that I had a lot of space in front of the drivers. I need to try measuring in the center of the room.

Jed, thanks for the link. The measurements are accurate (I think). But, if I don't consider baffle step during the crossover design, my speakers would end up sounding lean. Am I incorrect on this one?


Paul, the room is 5 meters x 4 meters x 3 meters (about 15x12x9 feet, w,l,h). The drivers are mounted on the final boxes, about 1 meter (3 feet) over the floor. I'm measuring them with a Behringer mic pointed straight towards the tweeter center, at a 1 meter distance, with a 6ms gate.