[sphynx_000]

Project

I plan to update the crossovers on a pair of Zaph Audio Waveguides. I slightly modified the design with help from this forum to accommodate some different speakers that I already had on hand. A while back I measured the speakers with a radio shack mic and a one of those universal calibration files that are floating around. Not sure how accurate it was but there appeared to be some fairly significant issues with the design. I now have a better mic and plan to re-measure the speakers and update the crossovers. I am hoping that I can post some speaker measurements here and get some help again with the modifications.

Speakers
Waveguide project
Tweeter - Seas 27TDFC
Design Woofer - Seas CA18RNX
Modified Woofer - Seas P18RNX/P, H1350

Crossover (Modified Perfectionist)

Modified

- C3 => 6.2uF

- L4 => 1.75 mH

- R5 => 10 ohm

Cabinet

Design- 19.5 liters

Modified - 15 liters

Questions

- What program should I use? I have REW, and trueRTA, and could probably get access to LSP CAD if that would help?

- When I measure the speakers, should I measure both together? and where should the mic be placed?


Thanks in advance to anyone how can help me out with this project. I'm pretty excited to finally get these babies dialed in!

 

[tuxedocivic]

Get HolmImpulse.

 

[coctostan]



Quote:
Originally Posted by tuxedocivic /forum/post/20822667


Get HolmImpulse.

+1, although you can get REW to work too. You just need an impulse response which can be gated to output a FRD file.


You also need to measure impedance. You can build an impedance jig (there is a thread in this forum) or buy WT2 or WT3.


You will want to measure without the crossover in place and you will measure the drivers individually. You aim the mic on your listening axis. If you listen with the tweeter at ear height you measure there. If you listen with your ear height between the woofer and tweeter, measure there. It is important that you take measurements aiming at the same point each time.


You can use Passive Crossover Designer which is a free tool that works in MS Excel. LspCAD demo works too but you can't save designs. The full version is not needed and not cheap.

 

[sphynx_000]

If measuring the speakers independantly, this is basically redesigning the speaker from the ground up? Is there a way to just take my current measured responce (compete speaker) and and tweek the system from there?


So with mic placement, I basically place the mic exactly where I will be be listening (distance and access). When you say "aim" that means position right? Its an omnidirectional mic, so it shouldnt matter where it is pointed???

 

[tuxedocivic]

If you'd like, you can measure the system now as is and try to make changes. But that'll be shooting in the dark. For instance, we know the tweeter is an LR2 slope xo at 2000hz. What if there is a big peaker at 2500hz? Maybe it's a break up peak. Maybe a rising response. If there is a big wide dip from 300 to 700hz, maybe the .5 woofer has to big of inductor. We don't know though.


The way coctostan described it is best. For now, might as well measure the system just to see what you have, if you'd rather start that way.


To measure, get the speaker as far away from any obstructions as possible. Even raise the speaker of the ground a little. Position the mic on axis with your listening plane at 1m (3.2ft) from the listening axis (tweeter most likely). If you have a very large are you can work in, back up to 1.5 or even 2m if you can. But you want to gate the measurements so there are no reflections from the room, so likely 1m will have to suffice.


Once you've done that, post the results and hopefully an expert can offer you some tips. I recently learned how to take measurements for a speaker. It's a lot of work. Getting them good is important. I have a small room. I moved ALL my furniture out of the room to finally get it right. You could almost say I was lucky my XO worked with my first measurements

 

[sphynx_000]

OK,


I will try and get some measurements done tonight, so at lease Iwill know what i am working with.


Luckly I have a pretty large open room to work with right now. Should the speakers and mic be centered in the room then?

 

[coctostan]

Follow what Tux is saying. It is best to take measurements with the speaker and mic as far from objects as possible, but it won't matter much as we will gate them and reduce the effect of reflections near the crossover.


I'm not proposing a redesign. By measuring the speakers sans xover I can then apply the known topology and go from there. Zaph's crossover is probably pretty close and this will make it easy to tweak components. Of course it might require a completely different topology but I doubt it.


Do you have the capacity to measure impedance?

 

[sphynx_000]

Not currently. But i think I will purchase a WT3 soon.