crossover simulation/measurement question

[Josephjcole]

I just recently built myself a microphone and have been reading Joseph D'Appolito's book Testing Loudspeakers. So of course I started testing my speakers right away. I built these speakers based on simulations and lots of listening. But now that I've started testing them I see that the simulations have been a little misleading.
I thought I had a second order L-R crossover at 300Hz. I've measured the impedence and it's pretty steady around 7ohms from 150Hz on with the Zobel network attached. I'm using a 7.5mH inductor with a 39uF capicitor shunted. Looking at the measurements it does not look like it reaches -6dB until almost 500Hz. For kicks I measured with just the 7.5mH inductor (and zobel), which should give a first order crossover. However this is not down a single dB at 150Hz and does not reach -3dB until 300Hz.
Can anybody give me any insights into why this would be? I'll attach the measurement pictures and impedence plots.
Thanks
Joe

[Josephjcole]

and the impedence:

[simon5]

Well your 1st order is ok, it's 3 dB down at 300 Hz like it should be.

Your 2nd order should be 6 dB down at 300 Hz.
You could look at the Impedance phase plot to see the problem.

Only thing I see it's your capacitor value is slightly too high, should be around 37-38 µF but that should not be a problem.
Maybe measure the capacitor value?

Maybe wrong wiring?

[f4ier]

It looks like you simply used textbook equations to work out those component values. You'll need to optimize those values based on both the impedance curve and raw acoustic output of your driver(s); against a target function. In this case the target transfer function is a low-pass LR curve. Remember to calibrate everything...

[soongsc]

Originally posted by f4ier
It looks like you simply used textbook equations to work out those component values. You'll need to optimize those values based on both the impedance curve and raw acoustic output of your driver(s); against a target function. In this case the target transfer function is a low-pass LR curve. Remember to calibrate everything...

That is true. I haven't seen that crosseyed doggie picture for some time now.

[soongsc]

Originally posted by Josephjcole
I just recently built myself a microphone and have been reading Joseph D'Appolito's book Testing Loudspeakers. So of course I started testing my speakers right away. I built these speakers based on simulations and lots of listening. But now that I've started testing them I see that the simulations have been a little misleading.
I thought I had a second order L-R crossover at 300Hz. I've measured the impedence and it's pretty steady around 7ohms from 150Hz on with the Zobel network attached. I'm using a 7.5mH inductor with a 39uF capicitor shunted. Looking at the measurements it does not look like it reaches -6dB until almost 500Hz. For kicks I measured with just the 7.5mH inductor (and zobel), which should give a first order crossover. However this is not down a single dB at 150Hz and does not reach -3dB until 300Hz.
Can anybody give me any insights into why this would be? I'll attach the measurement pictures and impedence plots.
Thanks
Joe

Why does the FR look like to one you posted for MLTR tuning?

[f4ier]

She's not cross-eyed! 'Cause of flash reflections I guess.

Yep. Don't have much time for this hobby now that I'm back to finishing my degree and working to support myself at the same time. After a looooong break I'm now in 2nd-year. 2 years to go

[Josephjcole]

Originally posted by f4ier
It looks like you simply used textbook equations to work out those component values. You'll need to optimize those values based on both the impedance curve and raw acoustic output of your driver(s); against a target function. In this case the target transfer function is a low-pass LR curve. Remember to calibrate everything...

This sounds good, but unfortunatly I don't know how to come up with cross-over values other than finding out the impedence at the crossover point and plugging them into the "textbook equations". What would you recomend, or do you have some suggested reading/links?

Originally posted by soongsc


Why does the FR look like to one you posted for MLTR tuning?

Because it is the same speaker. An 12" eminence crossed over to a 4" fostex, and now that I've started measuring I had a couple of questions. I didn't think crossover questions belonged on the fullrange forum.

Joe

[morbo]

What would you recomend, or do you have some suggested reading/links?

Speaker Workshop will do what you need, and IMO is your next logical step after aquiring measurement capability. Of course you could also go with something like lspcad or soundeasy, but SW is free.

[Josephjcole]

Unfortunatly I'm on a Mac so I don't have access to these programs. I've looked around a bit for software that can run these kind of functions on the Mac, but have not found anything. Is there any good way to go about it with out simulators?
thanks
Joe