How to use OmniMic and PCD to find the Relative Acoustic Offset

[Jeff B.]

(This was previously posted in TechTalk, but I added it here too - Jeff)

I have written and article on how to use OmniMic (or other measurement system) and my Passive Crossover Designer Excel tool to determine very accurately the relative acoustic offset between a woofer and tweeter's acoustic center for accurate crossover design simulation.

Finding this value is not difficult if these steps are followed, and once found will allow for very accurate crossover design simulation using my Passive Crossover Designer. Without knowing this value phase relationships will not be accurate, and as a consequence, the accuracy of the modeled response will be reduced as well.

As you will see, perfect measurements are not necessary. To make this point my example was done in my family room intentionally to show that you can do this the same way.......

To read the rest of the article please Click here to download pdf file with graphs to explain the details of this method.

Jeff B.

[arlis_1957@yahoo.com]

thanks jeff. this time it was easy to comprehend.

[dankwijoti]

Awesome, much apreesh Jeff. I have been using the PCD, but lack the mic to use it to its full potential. Hopefully that will be remedied soon.

[DanP]

Nice first post in this part of the forum, Jeff. I've been using this method of finding acoustic center offsets since you posted a thread about it in the main forum a while ago. I'm still impressed with how well it works.

Dan

[dcibel]

Just noticed there is a forum for the Omnimic!

Thanks for this Jeff, tried it out tonight, totally works and it also a simple procedure.

I am a bit confused though, as I tried this out for an econowave speaker that is in the works, crossover parts being ordered later tonight. With this method, it determined the offset between the woofer and tweeter is 0.115 meters, which is a lot more than what I think the physical dimension between the drivers is. Is there something funky that goes on with horn loaded speakers that causes a greater acoustic offset than the actual physical dimension?

[ernperkins]

I'm using Jeff's method, but not having much luck getting the waveforms to match in PCD. Because of room reflections I have to limit my gate time (I use HolmImpulse) to 3.5 msec (~ 300 Hz). Jeff's article states he uses a gate time of 8 msec (~125 Hz). I'm wondering if my extremely short gate time is the cause of my problems. Thoughts?

I've tried various HolmImpulse options: detect first peak on each measurement (no locking), detect first peak locked (woofer and woofer/tweeter measurements locked to tweeter peak) and casual impulse locked. I've also tried running the files through Response Modeler/Hilbert Transform to generate minimum phase. All to no avail.

Regards,
Ed

PS - Sorry about cross post in Omnimic forum. Perhaps I should just fork over the dough.

[DeeEyeWhy]

Now that I've entered the x,y,z data and have found the relative center how do I transfer this info into another x-over designer like SW or SE? For example SpeakerWorkshop only has input for entering the z (woofer acoustic center.) Since I can't enter the vertical off-set in SW will that throw the z info off?
Am I forced to use PCD for x-over sims if I use this method to find the relative center? What am I missing?

[Jeff B.]

Now that I've entered the x,y,z data and have found the relative center how do I transfer this info into another x-over designer like SW or SE? For example SpeakerWorkshop only has input for entering the z (woofer acoustic center.) Since I can't enter the vertical off-set in SW will that throw the z info off?
Am I forced to use PCD for x-over sims if I use this method to find the relative center? What am I missing?

Ideally, if you are planning to use Speaker Workshop or Soundeasy then you should just take your measurements using that software, you can even use the OmniMic as an USB microphone. You would just use the measured phase directly without any offset data. If, on the other hand, you want to use the OmniMic for measurements using its software, then simply use your calculated Z offset for the Z offset in Speaker Workshop. In fact, you can follow my steps directly in Speaker Workshop without using PCD at all to determine the Z offset, then you will know you have it correctly stated.

[DeeEyeWhy]

Thanks Jeff
Measuring with SW is a PITA and I can't seem to figure it out. But I can get good measurements (minimum phase) with both ARTA and OmniMic. Although measuring with SpeakerWorkshop is difficult its x-over design section w/optimizer is very nice and is very easy to learn. So, I'm trying to combine the best of all worlds, measuring with ARTA or OmniMic extracted to minimum phase, and using these minimum phase measurements in SW to take advantage of its optimizer. But as far as I can tell there is no way to add/adjust off-sets in SW beyond the Z. This is the only real problem with SW's x-over design section...it was created with measured phase in mind and not minimum.


In my dreams I imagine a easy to use all-in-one Mic/x-over designer with optimizer system. One that takes gated measurements, extracts to minimum, discovers relative centers, and then uses this data with a smart-optimizer to speed up and help you with x-over design.
If I could find this all-in-one system I'd buy it in a heart beat.

[Jeff B.]

Thanks Jeff
Measuring with SW is a PITA and I can't seem to figure it out. But I can get good measurements (minimum phase) with both ARTA and OmniMic. Although measuring with SpeakerWorkshop is difficult its x-over design section w/optimizer is very nice and is very easy to learn. So, I'm trying to combine the best of all worlds, measuring with ARTA or OmniMic extracted to minimum phase, and using these minimum phase measurements in SW to take advantage of its optimizer. But as far as I can tell there is no way to add/adjust off-sets in SW beyond the Z. This is the only real problem with SW's x-over design section...it was created with measured phase in mind and not minimum.


In my dreams I imagine a easy to use all-in-one Mic/x-over designer with optimizer system. One that takes gated measurements, extracts to minimum, discovers relative centers, and then uses this data with a smart-optimizer to speed up and help you with x-over design.
If I could find this all-in-one system I'd buy it in a heart beat.

If you're using ARTA just set your time marker and take your measurements for both drivers, then use the measured phase. You don't need to extract minimum phase and find the offset, these are additional steps that aren't necessary. That measured phase will already include the offset. Now with OmniMic it is a different matter since it doesn't have a fixed time marker that locks both drivers down to the same impulse origin.