What are you guys using for test baffles?

[Amphiprion]

I just wanted to get an idea of what people are using for test baffles when they test woofers (8" and under). I'm about to build a proper woofer and tweeter testing rig to do driver comparisons.

[JonMarsh]

What to use is a good questions- depends on what you want to know, how you want to get there.

I have a quasi-IB baffle setup for testing 10"/12" drivers.

For smaller drivers, I use a standard width box, and size, the PE 1 cu ft MTM; BS effects are easy to understand, and you can measure at 18" or so and see the driver plus box response, or at 4" or so quasi-nearfield and get a response plot that agrees pretty well with infinite baffle response, without the changes in HF response that ultra near field gives. It's trivial stocking up on front panel baffles and swapping them out. And the conditions are not far off from what you might see in an actual project. It's sort like the way Seas measures data.

Works for me, but I know others do all kinds of different things- from complete open driver near field (Mark K, for example), to IB (John Krutke).

I'd say the trick is to be careful about how you compare and draw conclusions- as long as you're measuring everything you're interested in yourself, and not comparing to other measurements, then correlation shouldn't be an issue.

[chasw98]

Jon, isn't there a standard for baffle size when doing driver measurements for drivers under a certain dimension? I believe the ISO has a published standard with the cutout offset so much by so much and the length by width specified. I always lose the link to the standard but I have seen it.

Chuck

EDIT: I meant IEC, not ISO. Thanks, Brandon.

[augerpro]

You'll have to decide if you are able to do a quasi-infinite baffle or if you have to stay small. If small, you can either use the IEC standard, or "typical" baffle. Sim it on edge so you know what the impact is on the diffraction.

Trying for a quasi-IB is better if you don't want to mentally remove the baffle diffraction all the time. How long of gate will you be able to use Mark? That and finding the midpoint between the ceiling and floor drive the design limits. My gate is 7ms so that means me entire baffle is "seen" by the mic, even though it's pretty big baffle. I think 7ms is about 2.48m, and every edge is less than that from the driver center. Once you know all that, determine your cutoff frequency. Than I just used Edge and SE to model the dimensions to be reasonably flat to about the cutoff frequency (or one octave above). Again the only dimension you can't adjust is the height of the driver center, this is the midpoint between the ceiling and floor. It is what it is.

Here is the sim of my current baffle:



This is big enough to look IB above the cutoff. The one I'm building today is 3/4" birch ply. I use brackets at the bottom that sit on a saw horse, and the top is bolted to the ceiling. I used asphalt sheets on the backside of the old, not sure what I'll do with the new one. PE only has vinyl sheets now. I'll be using some 1" x 4" oak trim glued on edge for stiffness, although this is already pretty stiff. I'll be testing bigger drivers (15") on this baffle though so I figured I should do what I can to stiffen it.

You'll need some room treatment to help with HD sweeps. I have baffles of that triangle acoustic foam above, below, and to the sides of the baffle. Some big sleeping bags are hung behind the mic stand.

For impedance I'll just be making each individual driver baffle a bit bigger than the driver itself. Then I'll through bolt those long shelf brackets at HD and Lowes on either side. 120-200 pounds of sandbags will hold the brackets still on the floor.

[JonMarsh]

You'll have to decide if you are able to do a quasi-infinite baffle or if you have to stay small. If small, you can either use the IEC standard, or "typical" baffle. Sim it on edge so you know what the impact is on the diffraction.

Trying for a quasi-IB is better if you don't want to mentally remove the baffle diffraction all the time. How long of gate will you be able to use Mark? That and finding the midpoint between the ceiling and floor drive the design limits. My gate is 7ms so that means me entire baffle is "seen" by the mic, even though it's pretty big baffle. I think 7ms is about 2.48m, and every edge is less than that from the driver center. Once you know all that, determine your cutoff frequency. Than I just used Edge and SE to model the dimensions to be reasonably flat to about the cutoff frequency (or one octave above). Again the only dimension you can't adjust is the height of the driver center, this is the midpoint between the ceiling and floor. It is what it is.

Here is the sim of my current baffle:



This is big enough to look IB above the cutoff. The one I'm building today is 3/4" birch ply. I use brackets at the bottom that sit on a saw horse, and the top is bolted to the ceiling. I used asphalt sheets on the backside of the old, not sure what I'll do with the new one. PE only has vinyl sheets now. I'll be using some 1" x 4" oak trim glued on edge for stiffness, although this is already pretty stiff. I'll be testing bigger drivers (15") on this baffle though so I figured I should do what I can to stiffen it.

You'll need some room treatment to help with HD sweeps. I have baffles of that triangle acoustic foam above, below, and to the sides of the baffle. Some big sleeping bags are hung behind the mic stand.

For impedance I'll just be making each individual driver baffle a bit bigger than the driver itself. Then I'll through bolt those long shelf brackets at HD and Lowes on either side. 120-200 pounds of sandbags will hold the brackets still on the floor.

I just LOVE watching other people work hard.... in this case, I won't get to, but I guess ThomasW will... :W

[JonMarsh]

Jon, isn't there a standard for baffle size when doing driver measurements for drivers under a certain dimension? I believe the ISO has a published standard with the cutout offset so much by so much and the length by width specified. I always lose the link to the standard but I have seen it.

Chuck

EDIT: I meant IEC, not ISO. Thanks, Brandon.

Yeah, the kind of thing Seas does. Then they measure at 18". I do that, too, using my PE cab 9" wide, and also measure at 4". The latter curve looks a whole lot like a typical IB measurement for each driver, and is rather easy to do... did I mention I'm lazy? Works pretty nice on 7-8" drivers and below.

[Amphiprion]

I think I am going to try quasi-IB. I might buy a 5x5 sheet of baltic birch and turn the top half into a semicircle radius = 2.5', placing a removable driver mounting jig in the center. That way the first diffraction impulse would be hugely noticeable and I could easily gate it out.

[augerpro]

Mmmm, I wouldn't do it quite like that. Assuming the ceiling height is the limiting factor in your room, what gate length can you get away with? My guess is the entire baffle edge will fall into this time window. Given that, you want to design the baffle so diffraction is minimized above the cutoff frequency (or an octave above). If you use the baffle edge to set the gate you'll end up with a really short gate. Unless of course you have a 2 story shop and are able to build one big baffle.

[Amphiprion]

Outdoor testing My office has a set of french doors that open onto a very large patio. Indoors I only have 8' ceilings.

[augerpro]

Heh, yeah that could work. Now you just have to build baffles with the edge nearest the driver to be farther out then the window length window length. For 7ms gate that would be at least 2.48 meters. How long of a gate are using?

Either that or design a baffle that minimizes diffraction. Otherwise you might as well just use a "typical" baffle width of say 9 inches. Using a circular edge may be certainly be obvious in the response, but it will be so wavy you won't make heads or tails of it. Comparing drivers would jsut be comparing one crazy wave to another.

[Amphiprion]

Maybe I could mount the drivers vertically on a horizontal sheet of plywood, and have some trim pieces cut so the baffle edges angled 45 degrees towards the ground? So the driver would be facing upward maybe 6" above the ground with the mic above it? Hope that makes sense.

[Dennis H]

My office has a set of french doors that open onto a very large patio.

Maybe cut a sheet of plywood the size of one of the doors so you can prop it in the opening and do a quasi-IB?

[Amphiprion]

Maybe cut a sheet of plywood the size of one of the doors so you can prop it in the opening and do a quasi-IB?

This is an awesome idea; unfortunately the patio is covered and also has 8' ceilings, but it would still be better than measuring in the office with all the clutter and junk I have laying around in here. But I could still put the mic on the patio and the driver facing outward and get some good results.

[augerpro]

deleted cuz I can't read sometimes ops: