Strategies for non-90 degree round overs

[John Whitley]

I’m currently working on a design that would include a beam with acute and obtuse roundovers, such as the cross-section below. The suspect corners are marked. Are there any standard(-ish) approaches to these cuts, or am I solidly into the realm of custom router bits?

[Chris Hall]

I take it that you do not have a shaper?

How clean and precise a cross section do you need to produce?

How many parts are you making? If it is more than one or two, and you don't have a shaper, I would consider having corrugated shaper knives made and then having a shop with a milling head that accepts corrugated knives to run x number of linear feet for you. One shaper knife profile could do both the acute and obtuse round overs, and the result would be perfect. Plus you would own the knives and therefore retain some small measure of design control. A pair of knives would cost no more than a custom router bit, give a cleaner result and last far longer, so it is a lot more economical in the long run.

If the result need not be so perfect, then you could work your way there with a combination of partial cuts with round over bits (for the acute angle) and hand plane work, and for the obtuse angle, a series of passes with a tablesaw at 3 or 4 angle settings could rough out the curve, and then you finish with plane or sanding block shaped to the curve.

[Chris Hall]

Of course, if you wished to stick with hand tools, and there was enough of this section to produce so as to justify the expense of the tools and dedicated sharpening media, then you could certainly process both those shapes with a pair of planes.

[Brian]

The 1/4 R might be worth having a bit made, for the larger 1/2 R I would apply layout lines and plane for a smooth radius. I have various rounder planes and find them useful enough to have on hand and to keep sharpening media, they're tedious to setup because you need to fit the chipbreaker well to the blade and fit the blade well to the plane sole in order for the whole thing to work out well. They're worth the aggravation as they do produce a very fine finish.

Make this part in straight grained material to make your life easy, the radius planes are going to tearout on reversals since they're so tedious WRT chip breaker setup that I don't find them capable of extremely tight chip breaker settings usually required of fine finish planing.

With only two to make and without a special plane on hand, I would just lay them out and cut them with a standard plane, back the blade out to a Kez level of fine-ness for the last round and you'll have something without noticeable facets.

[John Whitley]

I take it that you do not have a shaper?

Not yet. I’m broadly aware that a shaper is the obvious route here, but don’t have any on the ground experience with them. All the earlier prototype designs have had deliberately simplified profiles, easy to handle with standard router bits.

How clean and precise a cross section do you need to produce?

Pretty accurate. I plan to make the support bar in the second image via a laser-cut router template. The rails should nestle into the matching part of the support bar’s profile, ideally with no visible gapping.

How many parts are you making? If it is more than one or two, and you don't have a shaper, I would consider having corrugated shaper knives made and then having a shop with a milling head that accepts corrugated knives to run x number of linear feet for you. [ ... ]

For this part, likely a dozen at a go at the outset, so getting shaper knives made and finding a shop to handle the milling seems wise. Really, I need to do my homework on shapers, since I have another major part that could use a shaper for both the profile and tenoning, at three dozen per run.

[John Whitley]

Of course, if you wished to stick with hand tools, and there was enough of this section to produce so as to justify the expense of the tools and dedicated sharpening media, then you could certainly process both those shapes with a pair of planes.

Thanks Chris. I’m actively contemplating this kind of route for some later work. I’ve been building a sketchbook of design ideas for pieces that aren’t production models, more likely one-off or limited runs, which allow me to play with design concepts and build approaches which aren’t amenable to (tiny) mass production.

One nice example of this approach I’m considering in the fiber arts world are Golding spinning wheels. Scroll down a ways... There are a variety of Golding products that are vastly more production-centric, but the wheels are something else entirely (and priced accordingly).

The 1/4 R might be worth having a bit made, for the larger 1/2 R I would apply layout lines and plane for a smooth radius. I have various rounder planes and find them useful enough to have on hand and to keep sharpening media, they're tedious to setup because you need to fit the chipbreaker well to the blade and fit the blade well to the plane sole in order for the whole thing to work out well. They're worth the aggravation as they do produce a very fine finish.
[...]

Thanks Brian. I’ve been keeping the idea of standard and custom moulding planes in mind for a while now, learning a bit about how moulding planes are made, and their care and feeding. But I haven’t taken the plunge there yet. I have a few ideas in mind for which custom plane(s) seem like a good idea or perhaps even the best approach.

[Chris Hall]

If you get a shaper, the thing to look for, more than almost any other feature, is that it have an Aigner 'Integral' fence. The weak spot on a lot of shapers, regardless of how heavily built or how much horsepower, is the fence, and if the fence is not reliable, if you can't trust it to be at 90˚ to the table surface, then you will have frustrations.

Another feature to look for is the capacity to easily interchange spindles.

[Brian]

John, How about cutting the rail to accept the support? It would seem easier to me than to have mating parts with a rounded inside profile which of course when the RH changes it will gap around the outside of the beam at the edges of the profile.

[John Whitley]

John, How about cutting the rail to accept the support? It would seem easier to me than to have mating parts with a rounded inside profile which of course when the RH changes it will gap around the outside of the beam at the edges of the profile.

That's crossed my mind, Brian. The photos above are the first draft of a redesign of my rail supports, breaking away from the simple, functional "square with rounded corners" in the first prototype. I'm essentially feeling out a simple design language that will unite the rail supports, rail profile, and the stand's torii. Which is a long winded way of saying that everything's still up for grabs. Since this version is intended as a regular production piece, I need to be mindful of the time-cost of various design choices. Likewise, an important design goal is that the stand be easily disassembled for storage or transport. (A requirement common to virtually all larger modern hand-work fiber arts equipment.)

[John Whitley]

Do you have ripping table saw capability?

I make most rounded parts with a combination of table saw, hand planes and sanding block. I believe every shop could use a designated ripping table saw and for me its a favorite tool in the shop.

I do! Well, I did and I will, to be pedantic. I'm currently building my own shop after having worked in co-op shops for years. As it happens, I largely used the basic table saw approaches you describe on the first prototype, excepting the angled cuts needed for this new profile.

[useful hand-finishing advice]

Thanks, that's also helpful. As I mentioned earlier in the thread, I may or may not go with a hand-planed approach once I settle on a profile for the production stand. However, I'm eagerly plotting out designs for stand variations which will involve quite a bit of handwork.