[bowderynman]

I have read posts on the forum about rounding the (outer) edge of a disk. All good advice, so far. Concerns I have arise due to need to round over all edges of a square section ring to create a hoop; cfi curtain ring. My project, however, is not that simple: 1) it is not about making curtain rings per se. 2) there are 10 rings to make - all of different diameters; starting from about 0.75” bore to about 2.875” bore. 3) to create a circular section the internal bore edges, by definition, will need to be rounded as well. 4) earlier attempts to create these hoops on a wood turning lathe failed due to the hand held tools lacking the ability to reproduce - consistently - the order of accuracy demanded by this project; specifically and crucially in respect of the internal diameters and profile: these need to be accurate both in terms of internal diameter and radius. Perversely, the external profile only needs to “look” right!!
On the plus side, all the rings/hoops have the same circular section, which will be, of course, exactly two times the magnitude of the radius needed to turn the (already prepared) square section into a circular section. But this again poses it’s own issues: using a roundover cutter with a bearing results in there being no face to run the bearing against when the hoop is flipped over: making a “dummy” ring would be time consuming and waste material as the “dummy” disks would not profit any other projects or applications once discarded.
Using a ‘notched’ piece of (say) 3/4” ply to create a “nest” for the hoops whilst rounding the inter edges is a very good idea; it will need to be advanced/repositioned to cater for the different hoop diameters so a stop mounted on the router fence might suffice? But alignment and increment magnitude will need to be carefully managed. My solution would be to use the bearing as the initial setting stop - for each hoop - making sure the hoop is nested snugly in the notched plate. The notched plate, which should be in contact with, as well as guided and controlled by, the router fence at all times, must then be clamped firmly to prevent cutting forces displacing the hoop.
Biggest problem is, feeding the cutter: my router table is not in the least ‘user friendly’ and accessing the spindle adjustment and stops, underneath the table, is extremely difficult; with help I think adjustments could, in the event, be achieved. However, plunging a cutter (as in this case) into the workpiece could result in snatching. I don’t know how best to prevent or otherwise avoid this: the material is mahogany (of unknown pedigree) and could readily chip if not treated with respect. Flipping the hoop over also requires care as the router spindle will need to be reset to allow gradual advance of the roundover bit for this cut.
As there is not a great deal of purchase (grip) possible on the smaller hoops, preventing these from rotating, due to the cutter rotation, could prove very difficult so a positive means of safely preventing this - but at the same time allowing the workpiece to be rotated through the cutter - really is a must: I am, at this point, fresh out of ideas. A simple, manageable way of doing this safely and effectively would be a boon………..if anyone can help?

 

[John Smith_]

welcome to the forum, Bow.
if you can, could you post some sketches, drawings, or sample photos of what you want to do ?

 

[bowderynman]

Hi John…great to hear from you…..thank you.
Herewith images of project in progress:

7

 

[John Smith_]

yep - that helps a LOT.

 

[jdonhowe]

What an interesting challenge! First, a caveat: I've never done this myself, so just theorizing. Also, I hope I don't state the obvious to you. Anyway, FWIW, I think your "nest" idea (I assume a 90 degree notch) is the easiest way to go. Clamp the notch guide closest to you on your table (at 6 o'clock), such that the router bit cuts your ring at the 12 o'clock position. Start with the outside of the ring being a perfect circle, and have that outer edge bearing against the notch as you rotate your ring. For the inner ring surface, first make an approximately sized hole, which you will trim away to create your finished dimension. Rather than repeatedly having to adjust your bit upward into the cut, adjust the bit to the final height, and move your nest (notch) guide. Start with the nest close to the bit, and remove the first bits of wood from the ring center. Remove more of the inner surface by moving the nest farther from the bit. To keep your fingers safely away from the bit as you rotate your ring, you might cut a disc of polycarbonate the diameter of your ring, and attach to it using double sided tape at the perimeter. You probably need a spacer ring so the disc will clear the bit bearing. Again, I stress that this is just thinking out loud, and not based on having done this myself. One uncertainty with this is whether you can complete one half of your inner semicircle, and then flip the ring to finish the other half. Please post how you solve this- I have some projects in mind that could use this sort of profile. Good luck! Addendum: On re-reading your post, it appears that you want to make all your rings from one piece of wood by making multiple concentric cuts with your bit. If that's the case, perhaps begin with a disc the size of your largest ring, and make a center hole sized for your smallest ring. Once you've finished the smallest ring, cut it out (I'm intentionally vague here!) and proceed to the next larger ring, etc., until you finish with your largest ring.

 

[jw2170]

Welcome to the forum @bowderynman

That looks like an interesting project

How many of these do you want to make. If making multiple toys, I would look at templates with roundover cutters?

 

[old55]

Welcome to the forum.

 

[bowderynman]

Good try John! However, just to clarify: all but two of the ten rings are ready to have both internal and external edges rounded over. The remaining two are, shall we say, work in progres.
Yes, the “notch” would be an accurately made 90 deg “nest” within a square (or rectangular) plate of 3/4” ply - ctr line of notch must be parallel to the edge of the notched plate which would be held against the fence of router table.
Your approach, I note, may not quite work at the point I am presently at as I need to, as-it-were, ‘hit the ground running’. That is, there is now little or no opportunity to adopt a softly softly approach.
I did smile wryly at your comment: “what an interesting challenge”……..it has got a few scratching their heads, not least mine!
However, it is interesting to see you agree with the notch idea for controlling the work-piece (hoop). My biggest headaches from that point on, as I see it, are feeding the router bit into the work and a means of firmly and safely holding the hoop to: 1) keep it in contact with the notch and: 2) enabling a good firm grip on the hoop in order to rotate it against the cutter in a smooth continuous motion; the latter being the biggest “challenge”.
I’ll post how I solve this once recovered from my hospital procedure, which could take several weeks during which I have been advised to rest as much as possible.
As a qualified, mechanical engineer the principles involved here are not difficult to fathom; the challenge is adapting woodworking techniques to what, in reality, is an engineering challenge! In fact, I do have an effective and workable solution, using engineering machine tools found in almost every tool room. Problem is, I don’t have access to suitable machinery, hence the struggle to ”make do” with a woodworker’s routing table.
On a rotary table, with a chuck mounted on it, and set on a universal milling machine, the task would be “simple” and very quick. The only downside with the mill is that spindle speeds tend to be just a fraction of those achievable with a routing machine so, finish would not be as fine snd would likely require dressing by hand.
Anyway, I appreciate your interest and hope through perseverance, if nowt else, I’ll eventually succeed in solving all the problems discussed here.

 

[bowderynman]

Welcome to the forum @bowderynman

That looks like an interesting project

How many of these do you want to make. If making multiple toys, I would look at templates with roundover cutters?

Hi jw. Sadly, now I’m into my 80’s, such ambitions are mere pipe dreams; time not really being on my side; in addition to which I am having to cope with underlying health issues. In short, I simply saw the project as much an intellectual challenge as a craft skill challenge and intend to make just the one. Hopefully, given a successful outcome it might become a family heirloom?

 

[MEBCWD]

One problem I see is your 0.59-inch diameter for the ring cross-sections. You will need around a 19/64 radius roundover bit and that might be really hard to find.
I would recommend changing the diameter so you can use a standard size roundover bit.

 

[bowderynman]

Hello Mike….Welcome to the discussion.
Yes indeed, I agree wholeheartedly with your sentiments! However, there is light at the end of the tunnel! After a great deal of searching I did find a suitable “hybrid” cutter: you are absolutely correct; the design calls for a non-standard radius, that is, it does not fall within industry standards. That said, you would be right to increase the size of the section to permit selection and use of a “standard” cutter. Problem is, due to the collective (knock-on) effect this would have on the final build of the toy it would mean, in order to retain the integrity of the concept, a complete redesign of the toy: a huge challenge in itself so late in the day. So far, most of the toy has been made: the cone (tower) is complete as is the base. The rings are well advanced: 8 waiting to have the edges rounded and two with some turning work yet to be completed. Because of this, I persevered with my search for a cutter with an appropriate radius. Although successful, eventually, it is still a compromise as it has a cutting radius of 7.8mm ( I won’t translate to imperial as there is no simple equivalent): the design actually calls for a 7.5mm cutter radius - the rings/hoops being 15mm thick or in other words have a cross-section equal to 15mm diameter.
Just for the record, I failed to appreciate the implications of this when I took on the challenge - something I now regret! There is a reason for this: until I retired some years ago I was actively engaged in 3D CAD technology in the aircraft industry and indeed spent over 20 years using advanced CAD drafting programmes. Had I spotted the design flaws in this designer’s toy design I not only have the capability, I have the tools (a suitable computer programme, i.e. Sketchup 2017), experience and, indeed, the skill to redesign the toy to ”fit” within the tooling available, routinely, within the industry. But, as I say, back tracking would mean scrapping all that has already been made and obtaining a fresh supply of suitable material: demanding, costly and with personal health issues to contend with, impractical.
Mind you, purely as “an exercise in design” I did reproduce the designer’s design on my computer snd made some interesting discoveries, which enabled me to determine how much leeway there was for tweaking the section radius.That is how it came about that the cutter I found will meet the assembly criteria imposed by the peculiarities of this design.
With over 40 years as a Design Engineer (20+ years on the drawing board and 20 years on CAD system) my view of this design tells me that the designer - at the time he/she created this toy design - did so without the benefit of a background in any commercial enterprise involved in, engineering type, functional design: more than likely, he/she is a talented university undergraduate or an ‘A’ level student at a college for Advanced Education creating a project for course-work assessment.
Regrettable though it has been, I did miss the chance to do a completely remodelling of the design to make it, not only easier to manufacture, but compatible with accepted, industry standards for design and manufacture.
As I recorded elsewhere on the forum, a milling machine would make life easier: I am waiting on news in this case as a machine, currently out of commission in need of a new part - an internal power trip device - may be serviced in the near future - however, no rotary table is an issue as these can be very costly ( equivalent to around $500 US) which is, frankly, more than my personal resources can stretch to, especially as I would have no further use for it once the rings are finished.
”Back to the drawing board” is sadly, not an option. Just hope I can get there anyway even if it will mean using a good deal of ingenuity …….and, dare I say, luck!

 

[jdonhowe]

Oh, now you mention that you're a qualified mechanical engineer! I'm clearly not qualified, but fools rush in.... Anyway, I'm having trouble visualizing (I gather you'd say 'visualising') the notch orientation, vis-a-vis the router bit and the hoop- "ctr line of notch must be parallel to the edge of the notched plate which would be held against the fence of router table". Is the arc of the hoop between the notch and the bit, in which case there is no room for error. FWIW, my notion is that the notch be located such that the hoop bears against it at points on the opposite side of the hoop. So, the bit would be at 12 (inside the hoop), while (whilst?) the vertex of the notch would be exactly at 6, and hoop contact points would be at, say, 4:30 and 7:30. My thought was that any deviation of the hoop away from the notch contact would shift the hoop away from the bit, or, from the bit's point of reference, the bit moves toward the center. So, any imprecise contact with the notch would result in less wood removal, rather the too much, remedied by another pass by the bit. The "softy soft" I envisioned would be by incrementally moving the notch board toward the bit. My notion of sticking a polycarbonate (or acrylic) disc to the hoop was akin to using a glue block on a wood lathe to grasp an awkward piece. The thought was that with a transparent "block" attached to the top of the hoop you could see the bit; a simple handle could be attached to the disc for easier rotating. Just random thoughts from a rank amateur. BTW, is that an Isle of Man flag on your label? I always thought it one of the most interesting- if a bit peculiar- flag designs!

 

[bowderynman]

Hi John. Good call!👏 However, simplicity is everything; less scope for error. Problem is; how to make it as simple as possible and yet effective in achieving the desired result safely……watch this space…

 

[bowderynman]

Oh! the flag design by the way is called a Triskellion……featured on the national flag of the Isle of Man.

 

[woodworker47]

Bowderyman,

Is this your own design? I played with a similar toy as a kid. The only difference was that the center post was one diameter, making the hole size in each disk the same.

 

[bowderynman]

Hello Frank. The simple answer is no. The design is one I stumbled upon on the internet and thought it would be a great way to learn wood turning skills. Yes, the one size fits all idea is not uncommon for this kind of toy. I believe the cone idea was aimed at testing the ability of a young child to work out which ring should go on to the cone first and work out the sequence to complete the stack. Regrettably, the design, as published, fails to observe the fundamental principles of good design by not using easy to achieve dimensions and geometry for ease of construction, e.g. the angle of the cone flank wrt the base is 78.2 degrees! Why didn’t the designer make it at least 78 degrees dead or, better still 80 degrees? Numerous other ‘ failings’ are present making the manufacture awkward at the very least and certainly unconventional. By using standard sections and rounded, easily measured angles, the whole concept would have been much more user friendly, more effective and easier to make.
I am thankful none-the-less for your interest.

 

[DesertRatTom]

I like a challenge, but this just doesn't do it for me. I think you are right that it is a design failure. As to cuttng the rings, I think I'd turn them on a lathe with a stepped pattern. Each step would be the same width, plus an eighth inch for separating them. I'd then start at the top and separate each ring in turn from the stack. Separate the next down and cut it free from the stack. Then a roundover inside wouldn't be such a terrible task because in cutting each ring apart would roundover the outside, and you could roundover the exposed part of each ring while detaching it.

What would be left to do is roundover one inside edge. A nest for doing this using two 90 degree pieces clamped in place and using a small roundover bit with a bottom mounted bearing would be easy.

I would then stack the rings as carefully as possible, perhaps tape them in place, I'd measure the inside angle of the stack and turn the cone to that angle, and a little higher that the height of the stack. This is pretty much the reverse of what you're trying to do, but as we know, there are lots of ways to arrive at the same result. I'd also hollow out the cone before gluing it on the base to lighten it up a bit.

I don't have a lathe, which gives me the freedom to imagine doing this, but not the practical knowledge to actually do it.

 

[bowderynman]

Bravo Tom! We are on the same wavelength. Yes indeed, I did start off exactly as you describe. However, two things transpired: I was unable to reproduce the internal profile to the degree of accuracy demanded by the toy’s configuration and, secondly, my choice of material - MDF - was flawed due to it being carcinogenic: with significant dust being created during turning plus the absence of any form of dust extraction, the health risk was too onerous to accept. As a cancer victim I am more than happy I abandoned this first attempt! I do accept that turning hoops/rings in the way you describe is well tried; what is apparent is the indeterminate accuracy of the technique using hand held tools; an anomaly I could not accept for this exercise. OK, I suggest, for curtain rings, but too uncertain for my needs.
The status of the project is such that I have, essentially, gone past the point of no return: many hours of work have already been expended on this project. With all but two of the rings prepared waiting rounding of edges and the cone already finished you will understand backtracking, using a totally different approach, would take some doing, especially now, after a long search, I have actually managed to obtain a router bit with an appropriate (close) match to the specified design spec.
Due mainly to Covid, access to the workshop I have been using has been barred for over a year, but my health too has not helped. Hopefully, now the workshop is up and running again, I can look forward to revisiting the project and with good fortune on my side, actually finish it.
Thanks anyway for your comments and ideas.

 

[DesertRatTom]

Bravo Tom! We are on the same wavelength...my choice of material - MDF - was flawed due to it being carcinogenic: with significant dust being created during turning plus the absence of any form of dust extraction, the health risk was too onerous to accept. As a cancer victim I am more than happy I abandoned this first attempt!

I am another cancer survivor, cancer of the larynx. So I am very careful about sawdust. I use a large square of MDF occasionally as a backer to slide vertical pieces through the router. Made one project with it many years ago, a cabinet with insulated lining and a thermostat and small light bulb to keep my shop computer from freezing. Still use it as a tool stand for my drill press.

It will be interesting to see how this project turns out. I think the bit you're describing sounds like a small bullnose?

In turning, you use calipers to get sizes just right. Get the outside diameter right and then you can use them again to control the size of the opening. As you complete each ring, you make a flat cut so the shape is the size of the opening, minus double the thickness of your cutoff tool. It would look a little like a stepped drill bit. This is what you'd up with, which could then be smoothed to render the cone. Pix below.

I'm not an expert on wood, but it seems that Boxwood or some other very fine-grained species (apple?) would be a good choice for the rings. For the cone, placing the tool support at the angle of the cone's slope, and even attaching a stop to the tool's shank, should get you very close to what you intend.

Posting this not just for you, but I think a number of others are following this string.

 

[bowderynman]

I catch your drift. In the meantime have discussed my situation with my therapist and the advice is that, until I’m through the post radiation follow up - several weeks away at this time - I would be well advised to avoid struggling with this task as it does appear to be creating a degree of stress and it is impacting on my mental well-being. “Find something less demanding to do for a month or so and then look at the project again once you are in a better frame of mind”. In the circumstances, I have to agree with the therapist so, as we’re so close to the end of year festivities, I shall heed his advice and have a fresh look in the New Year. On that note, I shall sign off for now and just keep a watching brief in the event some breakthrough idea pops up. Thanks again everyone for the support…... Bowderynman ( ‘bowman’ in Manx Garlic)