Router Facets

[Ray_D]

I have been doing some experiments with facets to smooth the tweeter FR curve. The facets were made with a router using an approach shown in the attached photos. I have seen some made here using saws; but, this seemed safer and more flexible. Fairly arbitrary size can be made with simple tooling.

The FR graph is for a DQ25. The top curve is for a square edged baffle 8" by 12" with the tweeter 2.5" down from the top. The facets start about 3" from the top edge and extend about 9.5 " down. The second curve is for 3/4" facets. The third curve is for a 1.25 facet on one side and the fourth curve is for 1.25" facets on both sides. The curves are offset to show the differences. I ran some larger facets later; the curves did not get smoother.

Ray

[Face]

Do you have larger pictures?

[Ray_D]

Yes

I have larger photos but not the FR graph.

Ray

[TacoD]

Thanks for the instructions.

[Ray_D]

Instructions

Thanks for the instructions.

That sounds like sarcasm, Taco; but, I'll respond anyway.

I did not want to spend time on the technique unless there was interest and so far there has not been much. I'm doing these experiments because all of the programs I have seen are for facets in a single direction, not tapered. Also the facets most people produce are for larger angles and I wanted to see the effect of smaller angles. I still have not tried a facet along the top.

The router technique is fairly simple. It mostly involves getting a surface parallel to the facet and running the router to carve out the unwanted material. In this case the bottom of the window board is in the plane of the facet and the parallel top is the surface for moving the router. The only downside to this approach is a lot of dust. As you can see there is a board with a hole in it that rests on another board which is clamped to the enclosure at the top angle of the baffle. In this case the other end rests at the sharp end of the facet. Because the support board at the wide end has thickness, it is clamped slightly down. The amount can be calculated or you can sight along the bottom of the window board to align it with the drawn lines or you can set it on the line and move it down the leftover amount on the last pass. When you do it this way, the router depth on the last pass must be exactly the thickness of the board with the opening and it must not go beyond the sharp end. I glued a block which limits travel of the router there. For some reason I have not figured out, a very small amount of material is left sticking up; but, a quick swipe with a sanding block removes it. I am going to next try supporting the sharp end of the window board up a little bit and aligning the bottom of the board with the drawn facet lines either by sight or sliding a small rule in to touch the lines.

I am not presenting this as a definitive approach. It's just another way to make facets and does not require computing complex angles. I expect every time I do this I'll think of a way to do it different and maybe better and so will some of the woodworkers out there.

Regards

Ray

[TacoD]

That sounds like sarcasm, Taco; but, I'll respond anyway.

I did not want to spend time on the technique unless there was interest and so far there has not been much. I'm doing these experiments because all of the programs I have seen are for facets in a single direction, not tapered. Also the facets most people produce are for larger angles and I wanted to see the effect of smaller angles. I still have not tried a facet along the top.

The router technique is fairly simple. It mostly involves getting a surface parallel to the facet and running the router to carve out the unwanted material. In this case the bottom of the window board is in the plane of the facet and the parallel top is the surface for moving the router. The only downside to this approach is a lot of dust. As you can see there is a board with a hole in it that rests on another board which is clamped to the enclosure at the top angle of the baffle. In this case the other end rests at the sharp end of the facet. Because the support board at the wide end has thickness, it is clamped slightly down. The amount can be calculated or you can sight along the bottom of the window board to align it with the drawn lines or you can set it on the line and move it down the leftover amount on the last pass. When you do it this way, the router depth on the last pass must be exactly the thickness of the board with the opening and it must not go beyond the sharp end. I glued a block which limits travel of the router there. For some reason I have not figured out, a very small amount of material is left sticking up; but, a quick swipe with a sanding block removes it. I am going to next try supporting the sharp end of the window board up a little bit and aligning the bottom of the board with the drawn facet lines either by sight or sliding a small rule in to touch the lines.

I am not presenting this as a definitive approach. It's just another way to make facets and does not require computing complex angles. I expect every time I do this I'll think of a way to do it different and maybe better and so will some of the woodworkers out there.

Regards

Ray

Sorry, that was not my intention :E You have to understand that on a public forum, also non native English speaking people try to contribute.

[mazurek]

cool technique. Did you bolt something to the base of the router to keep it from falling through the facet template hole, or was the base already big enough? Please correct me if I have a misunderstanding.

[Ray_D]

Router plate

cool technique. Did you bolt something to the base of the router to keep it from falling through the facet template hole, or was the base already big enough? Please correct me if I have a misunderstanding.

I built a router table to fit in my BT3100 and used an aluminum plate from Rockler. It was big enough. You can always make a bigger plate from MDF or Ply. The hole really in not a template; it is just clearance. You just have to establish a plane parallel to the facet and work to the depth of the facet.

Ray

[deewan]

Great idea. I've been thinking a lot about trying a cabinet like this but was unable to come up with a good way of creating the angles. This just might be the ticket. Please keep us updated if you find any improvements.

[dsrviola]

facets

I think it looks fantastic, and I admire your (and others) woodworking skills.

I believe it was Dave Ellis over in Audio Circles who experimented with very large round overs on baffles and said a number of people who came to listen to compare the differences found that they preferred the larger (very large) round overs better despite the fact that the measurements were non-conclusive.

Cheers

P.S. Found the thread:

http://www.audiocircle.com/index.php?topic=50882.0

See Reply #16 on the first page

[JonW]

Hi Ray,

Very nice work! :T

I’ve also been pretty interested in potential benefits from facets. We had a nice discussion about ways to cut them a while ago:

http://www.htguide.com/forum/showthread.php4?t=28981&highlight=facets

I like your idea because it provides another way to make them. And everyone who makes speakers has a router. So it should be accessible to all.

Your measurements show very nicely how the facets can smooth out a tweeter’s response. I guess that’s why some of us go to the bother to make them. I also took some measurements to see how much the facets could help- versus roundovers and a straight edge. In case you’re interested, they’re shown here, starting at post #321 and going onto the next page. The results were enough to convince me to use facets.

http://www.htguide.com/forum/showthread.php4?t=25031&page=10&pp=35

For me, at least, adding facets taxes my woodworking skills. But I think the result is worth it. My first project with large facets was a floorstanding speaker with the top facets being 20” long. They came out OK but not exactly the same side-to-side. Well, no one can probably notice but I know they weren’t perfect. I used a circular saw to cut them. More recently I’ve made some much smaller cabinets with facets. These cabinets are only 10.5” tall and it was waaaaay easier to make nice facets. I just did it all on the table saw- also a BT3100 using the very helpful sliding miter table. I’ll write up a step-by-step build thread when these speakers are farther along. With your approach I also might think it a little easier with smaller cabinets. Unless you have really large clamps, I guess.

Any details on the speakers these are meant for?

Hopefully your approach will make life easier for someone out there.

-=-=-
dsrviola-

That’s interesting. From Ray and my (and many others) measurements the frequency response looks significantly smoother with the facets. But that doesn’t mean much, I guess, when listening if it’s not all factored into the crossover.

[Ray_D]

Any details on the speakers these are meant for?[/QUOTE]

Jon

Thanks

I'm mostly playing around. I bought a pair of ZA14s and a pair of DQ25s and am playing around with 3-ways. I have one with a ZA14, a DQ25 and a pair of RS225s. I have another one with a ZA14, a RS28A and a pair of RS225s. They both sound great. The RS225s are in a separate cabinet which makes experimenting easier. I'm also going to use some RS180s. I have put them off because all the ones I have are tied up in my HT.

Ray

[cjd]

225 > 180. stick with what you've got.

I like the technique too. This could be made a bit less dusty by combining with a hand-saw and essentially doing a "flush trim" pass with the router, rather than hogging ALL of the material out. I think this is also a lot safer in the long run, especially compared to huge jigs to move stuff on a table saw (which still has a limit based on the depth of cut available)

C

[Ray_D]

Sawing does not help much

I considered sawing off part of the scrap; but, unless you cut very close to final you are not saving much as all of the mass is biased toward the base. You also have another setup. It is not worth it in my estimation. It is just a little more vacuuming.

Ray

[cjd]

You don't save much mass but you save a LOT on bite and grab risk. I don't like hogging through that much material by hand, ever. I've toasted a flush-trim bearing trying to hog out too much, and I've also had the main router collar come loose same job - too much vibration. Last time I'll do that. Like I said, a bit less dusty.

[Ray_D]

Large bite not required

You don't save much mass but you save a LOT on bite and grab risk. I don't like hogging through that much material by hand, ever. I've toasted a flush-trim bearing trying to hog out too much, and I've also had the main router collar come loose same job - too much vibration. Last time I'll do that. Like I said, a bit less dusty.

Chris

I guess that I did not provide enough explanation with the pictures. You can take as small a layer off at a time as you wish.

It is never safe router technique to take large bites and there is no reason to here. Also, the router should be moved to cut into the material rather than a "climb" cut, which means you move the router counter clockwise around the edge of the cut.

Regards

Ray

[Ray_D]

Another thought

Another feature I forgot to mention is that as you take material off the remaining surface has the same shape as the final facet. By taking small cuts, depth changes, you can see the edges approach the lines drawn on the enclosure and if they are not parallel you can make adjustments to the setup. The actual cutting goes very quickly.

I have not found any software that simulates 3D facets. I mostly use Soundeasy which has square, circular and chamfer capability. If anyone knows of any such software I would appreciate knowing. My initial experiments seem to indicate that 45 degree facets are not required. Facets of 20 to 30 degrees seem to provide most of the benefit. Obviously, doing this in software would be a lot faster.

Thanks

Ray

[JonW]

I'm mostly playing around. I bought a pair of ZA14s and a pair of DQ25s and am playing around with 3-ways. I have one with a ZA14, a DQ25 and a pair of RS225s. I have another one with a ZA14, a RS28A and a pair of RS225s. They both sound great. The RS225s are in a separate cabinet which makes experimenting easier. I'm also going to use some RS180s. I have put them off because all the ones I have are tied up in my HT.

Ray

Those look cool. :T I’d very much like to hear the results. Have fun with them.