A question you can't answer...baffle width

[edlafontaine]

Is there a "sweet spot" regarding the size of roundovers outside of an 8"mid bass? ...or tweeter waveguide?

I may be able to use roundovers up to a 4" radius.

The flip side I must consider is the overall width of the baffle/assembly.

4" + 8" + 4" is 16" That makes a pretty wide baffle...I think.

Is the width worth it?

[djkest]

Re: A question you can't answer...baffle width

Is there a "sweet spot" regarding the size of roundovers outside of an 8"mid bass? ...or tweeter waveguide?

I may be able to use roundovers up to a 4" radius.

The flip side I must consider is the overall width of the baffle/assembly.

4" + 8" + 4" is 16" That makes a pretty wide baffle...I think.

Is the width worth it?

I believe the roundover has diminishing returns. Zaph did some tests where there wasn't much benefit above 3/4" roundover, so anything more than 1" is probably more cosmetic than functional. The baffle width diffraction is usually more in the tweeter range than a woofer, but if you get wide enough it can get below 1000 Hz.

[Sydney]

Re: A question you can't answer...baffle width

Is the width worth it?

Since I have multiple 8" drivers in the wall, subjectively I would say yes.
Perhaps evaluating the driver in a very large baffle would help in determining?

[DanP]

Re: A question you can't answer...baffle width

I think the consensus is that anything less than 3/4" doesn't make much difference, so the larger the radius the better as far as smoothing out the diffraction signature goes. However, the effect is going to be more pronounced on the frequencies produced by the tweeter. Any chance you could do the 4" roundover near the tweeter (4" + 4" tweeter + 4" = 12") and then transition down to 2" roundovers near the woofer?

Dan

[edlafontaine]

Re: A question you can't answer...baffle width

djkest posted: ...Zaph did some tests where there wasn't much benefit above 3/4" roundover,...

I find what Zaph has on his site to mean the opposite. This and what I find through EDGE simulation tell me that larger is "better". I'm looking for help in determining "how much vs. aesthetics"

From Zaph|Audio: All of my designs call for roundovers on the baffle. 1/2" is the minimum, but 3/4" is better if you have a large shank, variable speed router than can handle it. The roundover is required to help smooth out the diffraction ripple response.

DanP, I have curved plywood sides which transition from 4" radius to ~5/8" radius over their ~40 length. The transition is not so quick as to allow what you suggest. I will use the larger radius on top.

[dthomas]

Re: A question you can't answer...baffle width

Actually the differences between a sharp edge and a radius affect frequencies as low as 500Hz. Of course this is assuming you have a wide baffle of 10" or so. They have very significant effects at frequencies starting at around 1000Hz even a bit lower. A narrower baffle has far less negative diffraction effects especially in the 1000Hz to 2000Hz range but they have more baffle diffraction loss at below 1000Hz but this tends to be smooth roll off so it can be a bit easier to deal with. And speaker directivity is going to determine how severe and at what frequencies they occur.

Also the effects that are most audible occur in the 800 to 2500Hz range so upper midrange to lower treble regions are where they are most likely to be audible. The peaking that occurs with a 5.25" driver at 1200Hz is 3db greater on an 8" baffle than on a 6" baffle regardless of the round over. Plus a .75" round over only knocks this down by 1db. If you could do about a 5" round over you could completely remove it but a 5" round over is kind of hard to do.

The issue is in order to effect the diffraction effects at 1000Hz you need a very large round over. The effects of a round over at .5" is so minimal you could easily argue it has no real audible effect. However if you could do round overs in the 2" plus range it could have quite dramatic effect. However at higher frequencies even a .25" round over can have some effect on a tweeter.

Rather than round overs I think most DIYers could benefit from large beveled edges. It is far easier to make a large bevel vs. a round over. A very large bevel also improves baffle step loss.

At the end of the day you can combat these effects in the crossover also.

[DanP]

Re: A question you can't answer...baffle width

How tall is your cabinet going to be? That is going to look really sharp with a tapered roundover from top to bottom. You're probably just going to have to pick how wide you want your baffle to be and then just choose the section that can accommodate your woofer with the largest possible radius near the tweeter. I think it would look cool if you countersink the woofer, but well into the roundover, so the top and bottom of the woofer flange are flush with the baffle, but the sides of the woofer flange are poking above the baffle a bit. This would allow you to have a larger roundover near the woofer and I don't think you'd hear any effects from this. I want to see pics. Soon.

Dan

[tombond]

Re: A question you can't answer...baffle width

Hey Ed, this curved ply wouldn't be from your shenanigans as a dumpster monkey would it? :D

[edlafontaine]

Re: A question you can't answer...baffle width

It will be awhile for my pics. Here is my inspiration:



W-frame woofers under an OB 8" under an elliptical waveguide loaded tweeter. I'll likely back mount the mid-bass and support independently from the baffle.

Tom, my resourcefulness at keeping items out of the landfill is noted...:D Yes, climbing over the top, hanging on the sides is all part of keeping in shape...along with unearthing the treasure.

[Sydney]

Re: A question you can't answer...baffle width

As Dave mentioned the width of the baffle is a consideration.
As well as the Directivity of the radiators.

[Paul W]

Re: A question you can't answer...baffle width

Download BDS (Baffle Diffraction Simulator) and play with various edges. That will provide good insight.

[Pete Schumacher]

Re: A question you can't answer...baffle width

Since your inspiration is the Orion, improved by the WG loaded tweeter, crossing in the 1000Hz range will basically take care of baffle diffraction problems with the woofer. Roundovers can be completely ignored in this case, at least for woofer consideration.

The WG will also reduce baffle edge diffraction for the tweeter, if it's large enough (8" or so).

[6thplanet]

Re: A question you can't answer...baffle width

Your inspiration, too, huh? Funny he dosent use any kind of "real" round over on it...

It will be awhile for my pics. Here is my inspiration:



W-frame woofers under an OB 8" under an elliptical waveguide loaded tweeter. I'll likely back mount the mid-bass and support independently from the baffle.

Tom, my resourcefulness at keeping items out of the landfill is noted...:D Yes, climbing over the top, hanging on the sides is all part of keeping in shape...along with unearthing the treasure.

[Pete Schumacher]

Re: A question you can't answer...baffle width

Your inspiration, too, huh? Funny he dosent use any kind of "real" round over on it...

You've just pointed out one weak link in the Orion design, the tweeter on a 180 degree waveguide with no diffraction control, and in that picture, a horrid baffle edge.

Using a guide to control the tweeter dispersion, to more closely match that of the dipole woofer, should offer complete mitigation of baffle effects on the woofer, letting it better run in true dipole mode over its range, as well as improve overall power response. But even a guide on that baffle pictured above, would still suffer from that raised edge. Smooth it flat, and even a sharp corner wouldn't be offensive if the guide is large enough.

[jmb]

Re: A question you can't answer...baffle width

A couple of quick calculations: 3 wedges with side angles of 11.25 degrees on each side with exterior widths of 5 cm (4.2 cm on the internal face) would yield about a 4 inch radius with benefits seen to better than 2700 Hz. Add an inch to the radius (6.3 cm external face and 5.5 cm internal face) and your benefits are realized to better than 3800 Hz.

Jay