PCD Question: Baffle Diffraction?

[Tony Nelson]

I am in process of working on baffle diffraction. I am unsure whether to model in as inverted, gain, or loss. I have the dimensions on the woofer and then when I save it to the response curve I can see the curve change. Which setting should I use? and then do I account for the step in RM or in PCD?

[joekraska]

Hmmm. I use loss. Which is what it is, as far as I know:

Frequencies under 1,000Hz will "wrap" around a small baffle and head in directions not towards the listener. The energy exists, but goes places that don't matter.

C//

[Æ]

Hmmm. I use loss. Which is what it is, as far as I know:

Frequencies under 1,000Hz will "wrap" around a small baffle and head in directions not towards the listener. The energy exists, but goes places that don't matter.

C//

Well maybe not in PCD, but it actually it does matter in your listening room, it's called "power response."

[Tony Nelson]

So does it matter which one I use? And after I do the diffraction, then I take minimum phase right? And import the final .frd into PCD?

[Mark65]

You want to model the baffle step as well. I model the diffraction (and baffle step) as loss, just seemed like the thing to do. If I'm wrong, then my new speakers won't be sounding as good as I had hoped...

I extracted phase after modeling the diffraction and baffle step, then imported into PCD. I believe (and sincerely hope) that I did it correctly.


Mark

[amplus]

I'm about to model some surround speakers and was thinking about this in PCD. If they are mounted in the wall, would the diffraction be modeled be Gain instead of Loss? Maybe this is what Jeff was thinking, since it is like an infinite baffle.

[JustinG]

I'm about to model some surround speakers and was thinking about this in PCD. If they are mounted in the wall, would the diffraction be modeled be Gain instead of Loss? Maybe this is what Jeff was thinking, since it is like an infinite baffle.

If it is an infinite baffle there should be no loss/diffraction since there are no edges or box to wrap around.

[amplus]

If it is an infinite baffle there should be no loss/diffraction since there are no edges or box to wrap around.

Ah ok... so don't use the baffle diffraction at all in this case. When would there be gain?

[Tony Nelson]

You want to model the baffle step as well. I model the diffraction (and baffle step) as loss, just seemed like the thing to do. If I'm wrong, then my new speakers won't be sounding as good as I had hoped...

I extracted phase after modeling the diffraction and baffle step, then imported into PCD. I believe (and sincerely hope) that I did it correctly.


Mark

I figured you might chime in Mark. And that is what I thought I was doing. Except in my PCD it wouldn't let me extract phase? It was like the button wasn't working or something. I am not sure. I will try again tonight and see if anything is differant.

[Jeff B.]

I'm about to model some surround speakers and was thinking about this in PCD. If they are mounted in the wall, would the diffraction be modeled be Gain instead of Loss? Maybe this is what Jeff was thinking, since it is like an infinite baffle.

Technically diffraction is a gain for a driver in 4pi space. However, most drivers are rate with half-space (2pi) sensitivity, so for them it needs to be modeled as loss. It just matters on your frequency response is derived.

[Jeff B.]

I figured you might chime in Mark. And that is what I thought I was doing. Except in my PCD it wouldn't let me extract phase? It was like the button wasn't working or something. I am not sure. I will try again tonight and see if anything is differant.

I am assuming you are referring to the Response Modeler rather than PCD, which doesn't have an Extract Phase button. This prompts me to ask this question - which version of the Response Modeler are you using? Version 2.0 will not extract phase when running within Excel 2007. You must use version 3.01 for that. It is available at the link below.

[Tony Nelson]

I am assuming you are referring to the Response Modeler rather than PCD, which doesn't have an Extract Phase button. This prompts me to ask this question - which version of the Response Modeler are you using? Version 2.0 will not extract phase when running within Excel 2007. You must use version 3.01 for that. It is available at the link below.

Ya I meant Response Modeler. Sorry for that mixup. I have used the extract phase button before, when I was just messin around with it, but now that I am serious about using it the button wouldn't work. When I was opening RM last night it was giving me a runtime error, so maybe that was the reason for it?. I am using it in excel 2007, so maybe I will download the new version tonight when I get home from work and classes.