SoundEasy 14 Tip - Simulating Diffraction

[robert]

Since a few of you are interested in buying or have bought SE, here is a tip.

How to import an IB measurement and simulate baffle step:

1.) Import the IB measurement file, get the level correct, enter T/S parameters if they aren't in the file. Make sure the driver diameter (Sd) is entered correctly. For how to do this, see SE manual.

2.) Go to the Diffraction / Compound tab and set up a diffraction sim of your test cabinet and run it. Be sure driver location and cabinet dimensions are correct, and that rounded/beveled edges are on if used.

3.) Go to the EasyLab Digital MLS dropdown and select "SPL/Zin" and click "add diffraction to" buffer 0. Buffer 0 IS the driver file.

4.) Save the driver file as a DIFFERENT NAME - ex. "Dayton 12S2 Diffmod 1" or something along those lines. You still want the imported IB file so you can try different cabinets and layouts.

5.) Now you have a usable model for crossover design and to see if your baffle will work with the driver's SPL.

Now: does anyone have an IB file of a HiVi F8 and/or SEAS 27TDFC???

Cheers / Robert

[dlneubec]

Hi Robert,

This is also the same basic process you use to add baffle affects to a nearfield measurement, say on a woofer, where you will be merging nearfield and farfield measurements to simulate a final, on baffle response for crossover design.

It's a little confusing because SE uses the word "diffraction" and not baffle step, but I believe it is simulating both attributes combined.

[robert]

Yep, it's done exactly like nearfield + diffraction, just a different file used! The measurement part I have well down thanks to John Kreskovsky's guide. Far Field / Near field + diffraction merge.

I was hunting for a way to take downloaded CLIO files, etc. from outside sources and sim a cabinet's effects w/o measurement. That was my favorite feature of Jeff B's freeware.

I am sure this procedure is documented somewhere in the SE manual, but thought it deserved a callout as sometimes it's a little tough to find things in there.

Cheers / Robert

[Eric.DeYoung]

I thought I would bump this thread up, because I'm having a little trouble getting my brain around a couple things.

First off, in John K's guide, it says that for far field measuring, that the driver is assumed to be mounted on the intended baffle. I'm a little confused on why later in the process, diffraction is added to the merged response of near and far field if the driver is already mounted on the intended baffle - doesn't the measurement account for the diffraction already? If its to be used in the crossover design, I would have thought that the measurement would be used without accounting for diffraction since the microphone only picks up whats "there?"

Second - if both near and far field measurements were acquired on an "infinite" baffle, would I add the diffraction/baffle step effects once I knew my placements in the intended cabinet (and cabinet dimensions), and use this curve in crossover design? Doing this would effectively bring the low frequency end of the curve down in amplitude, correct?

I guess another way of looking at it is that when doing it "Johns way," you are adding the diffraction to arrive at a simulated infinite baffle response so that you can use the measurements for other cabinet designs? However, this is a bit confusing to me because it says the measurements are already done on the intended baffle, so I'm not sure why the measurements don't already include the diffraction.

I've been thinking about this everytime I get in the car and drive somewhere, and I'm having a hard time coming to a sensible conclusion.

Thanks for any help on the subject, I just want to get measuring some drivers!

[Shonver]

You must choose one or the other. Either you have a nearfield measurement to which you add diffraction effects, or you get the in situ driver measurement that already includes the baffle step/edge diffraction in the measurement (of course, then you would not need to add diffraction effects). The only problem with SE is the basic edge shapes. No simulation of compound box shapes, such as trapezoidal or pyramid shapes with rounding of only few edges, so I prefer measured to simulated response.

[Jeff B.]

Hi Robert,

This is also the same basic process you use to add baffle affects to a nearfield measurement, say on a woofer, where you will be merging nearfield and farfield measurements to simulate a final, on baffle response for crossover design.

It's a little confusing because SE uses the word "diffraction" and not baffle step, but I believe it is simulating both attributes combined.

The reason SE uses the word "diffraction" for baffle step is because baffle step is diffraction. In fact, it is exactly the same thing as edge diffraction and it is solved using the identical math. Baffle step is simply the diffraction we see when wavelengths grow considerably larger than the baffle. This step would occur at any frequency where this relationship occurs. So whether you are using Soundeasy, or BDS, of the Edge, or my Response Modeler, the same math is calculating the whole diffraction signature across the entire spectrum - it isn't calculating baffle step AND edge diffraction as different things, because they are the same thing.

Jeff B.

[Eric.DeYoung]

I'm pretty sure I have a decent understanding of baffle diffraction/ baffle step, but I think looking over John K's measurements just confused me the more I looked at them.

So are you suggesting that near field measurements need not have diffraction effects added to them (whether measured infinite baffle or in the intended baffle) until the cabinet is designed and driver placement is defined? This would follow the image that is in the guide where low frequency data is unaffected by the baffle step addition.

And far field measurements don't need diffraction effects added to them as long as the driver is in the intended baffle.

And lastly, if I want a full driver response on an infinite baffle (which I will apply diffraction effects to once I design a cabinet), can I do only a near field measurement?

I just want to be sure I'm doing this correctly, and more importantly make sure that I understand why.

Thanks!
Eric

[augerpro]

Regarding the first post, IF you are taking an (imported) infinite baffle response and want to add the diffraction for your particular enclosure, you must plot the infinite baffle response at -6 dB. SE adds the diffraction to the reponse. The reason is that relative to the infinite baffle response the diffraction would be subtractive.

[Eric.DeYoung]

Augerpro, I read your "READ THIS FIRST" post on your page, and I think I'm starting to get the idea.

If I only do a farfield plot for woofers , then I can use this in addition to TS parameters to model the complete response, with some error in the low end (up to the gating frequency cutoff). The baffle diffraction will be added to the response when the enclosure is designed if the measurement was done on an infinite baffle.

If the measurement is done in the proper enclosure, the farfield measurement will be ready to be used for crossover design with a proper tailing.

For tweeter measurements, nothing changes from above except for the obvious exclusion of TS parameters for calculating the low end. Baffle diffraction is still added if the measurement was taken on an infinite baffle, otherwise the tweeter response is ready for crossover work.

Sound about right? I think my brain is starting to move in the right direction! (I hope )

[robert]

Correct.

Far-field measurements have diffraction, near-fields don't, so you have to sim it and add it to near-field before you merge. I never really thought about why, just, "I are a technician, so I follows the manual!" Follow John's procedure, it does work.

IB measurements do not have diffraction (no edge to diffract off of) so you have to sim it in your design. Downloaded CLIO's and such are good for starting a rough design and figuring out how the drivers "work" together, but you will find a lot of the time that the amplitudes aren't exactly right.

Cheers / Robert

[Eric.DeYoung]

I think that's just what I was missing - near fields not having any diffraction. Thanks for all the replies and explaining to a measurement noob - time to to start measuring!

Eric

[dlr]

Regarding the first post, IF you are taking an (imported) infinite baffle response and want to add the diffraction for your particular enclosure, you must plot the infinite baffle response at -6 dB. SE adds the diffraction to the reponse. The reason is that relative to the infinite baffle response the diffraction would be subtractive.

Actually, it's neither additive nor subtractive. It's a cross-product or multiplication to go from 2-pi to 4-pi using the diffraction signature in db. This is with the signature based on 0db as nominal (the high frequency 2-pi end) for the diffraction signature.

Division of the 4-pi response by the diffraction signature, the inverse, yields the 2-pi response.

dlr

[augerpro]

Thanks Dave. I didn't want to get to technical for the OP and really confuse him though