Geometry question

From the rest of the thread, a picture or sketch would be very helpful.

People already make screws with O-ring seals: .

And, Kaiser Tool Company makes lathe tool bits for cutting circular grooves at various angles. I think that their ThinBit line is what you seek.

Joe Gwinn

Reply to
Joe Gwinn
Loading thread data ...

The customer wants the O-ring to seat high up the taper. He wants the groove normal to the angled surface. I think though that I may be able to talk him into a groove that is normal to the screw axis. Eric

Reply to
etpm

The o-ring will be sealing against low pressure-basically water a few inches deep. So the sealing cannot be helped by pressure or vacuum and instead must rely solely on compression. The customer would rather use stock o-rings that are available everywhere. Eric

Reply to
etpm

The groove is square cross section.

Reply to
etpm

Greetings Ed, The tool is only .06 wide and the groove radius is small so I need circular contours for strength. Remember, I have made tools like this before just because of the needed strength. Eric

Reply to
etpm

Greetings John, I know I can use HSS. I need to do 500 of these at a time and need to do them fast. Eric

Reply to
etpm

Greetings George, I already know how to hold and locate the screw. It will take approximately 4 seconds to remove one screw and put another in the collet. What I really want to know is if my way of finding the radii of the walls of the groove is one correct way. Eric

Reply to
etpm

Greetings Joe, Zago does indeed sell those screws. $2.90 each in quantity. I need to be able to make these for about 25 cents. And I will. Thinbit does not make the groove tool I need. They will though, for about &175.00. I have been through this before. Eric

Reply to
etpm

o-ring is overkill -- use flat rubber washer for face compression seal.

Reply to
F. George McDuffee

wrote in message news: snipped-for-privacy@4ax.com...

On Thu, 16 Apr 2015 20:35:10 -0400, Bob Engelhardt wrote:

The groove is square cross section. ==============================================================

I hope you meant rectangular, not square. The oring is deformable but not compressible, so the cross-sectional area of the groove needs to be bigger than the cross-sectional area of the oring else you extrude and cut it on tightening. As for the geometry you could do it with a sketch and trig but I would prefer to just draw it in a cad program. Being a lover of run-on sentences here's how I would do it in autocad. Draw an angled line to represent the angled side of the flat head, draw a line perpendicular to that for the end of the oring groove closest to the threads, make an offset copy of the first line offsetting by 80% (or 75% or whatever you choose from the Parker oring handbook for your oring and pressure and oring life but it almost always works out to be between 75 and 80%) to be the bottom of the groove, draw a circle for the oring tangent to the side and bottom of the groove, and finally an offset copy of the first side line offset by 125-150% of the oring diameter to be the second side of the groove (this creates the extra cross-sectional area for the oring to deform into, and isn't critical so long as it is large enough). Draw a line parallel to the screw axis offset from the center of the oring by the radius of the ring (which also be the centerline axis of the screw), trim all the original groove lines to just leave the rectangular groove, then reflect everything around the centerline axis. Now add the top of the flat head and the threaded section, trim the excess from the angled lines, and you can pick off the dimensions you need. That's the stuff I'm sure off, the actual machining I'm less help with. On a manual lathe the way to do this would be to use a tool with a squared-off end like a parting tool and rotate the compound to the right angle and just plunge the compound. On a cnc couldn't you grind the end of a boring bar to the parting tool shape at the angle you need, and then make the angled plunge cut to simulate the straight plunge of a compound link you were turning a taper? Either way I don't see where you need any kind of a radius on the cutting tool, except a tiny break on the two front corners to leave a rounded chamfer at the bottom corners of the groove for stress relief. Hope this helps some :-).

Regards, Carl Ijames carl.ijames aat deletethis verizon dott net

Reply to
Carl Ijames

I'm assuming the screws are going to be tightened, which means you can get much much more than enough compression of the oring. The trick will be to have the right amount of compression and design it so it doesn't tear, cut into or shred the oring when the screw is tightened all the way.

Reply to
jim

Not for nothin', but it seems to me that for $125 total project income (500 pieces at $0.25 each), you're going to have a tough time making a profit on this. You've probably spent more than $125 worth of time just thinking about the tooling.

After 30+ years in business, I've learned that there are some jobs you just ought to walk away from. Tell the customer he can buy the parts for $2.90 from Zago.

Just sayin'

Reply to
rangerssuck

I assume that Zago has competitors by now. Given competition and low the production volume, this may be a good price.

One can mine thinbit's catalog for ideas. I've made face grooving tools by grinding HSS. The hard part may be holding the screw to be machined.

And, maybe the better approach is to have special screws made, complete

screw machine would make short work of this. The minimum order may be

10,000 screws.

Joe Gwinn

Reply to
Joe Gwinn

I wonder if you could use a rotabroach type cutter as used in magnetic drills, I have quite a few and they go down to 10mm hole size in metric, maybe 3/8" in inch as that is close. The pilots are 1/4" IIRC so they may be ideal if the end geometry is suitable and you can control the depth. They are available in carbide but the only ones I've used are HSS. I cut 304 with my HSS ones frequently and they go fine but I haven't cut anywhere near 500 so couldn't comment on longevity. I suspect a maker of the cutters could comment on the likely lifetime before re-sharpening if required.

Reply to
David Billington

And yes, I would definitely consult with the o-ring manufacturer on this. Then you have so real, informed backup for your decision, other than Gunner's "Ayup."

Reply to
rangerssuck

Greetings Carl, I think you described in more word what I said in my first post. Eric

Reply to
etpm

George-The customer wants an o-ring. I said this. Plus a rubber washer won't work because the screw needs to bear against the aluminum. If there is a washer in between then the countersink needs to be changed. And all the parts out in the field would need to be returned to be re-machined. If just the screw is modified then my customer can offer this screw to his customers if they want an upgrade. Look, I appreciate all the advice about how an o-ring is not the best solution but it is what my customer wants and in this case I agree. What I really wanted to know is if my method for finding the radii of the sides of the o-ring groove will work. Thanks, Eric

Reply to
etpm

I get paid for tooling and setup and programming and engineering. Then I get paid per piece. The job will come up several times a year. Each time it does I get paid for setup and tool attrition. Eric

Reply to
etpm

So my post should have just been "yes"? :-) :-) :-)

Regards, Carl Ijames carl.ijames aat deletethis verizon dott net

Reply to
Carl Ijames

Fair enough.

Now, for a possible solution: After checking with the o-ring manufacturer who will almost certainly recommend a rectangular groove, why can't you do this with a rectangular grooving tool set parallel to the screw head and fed at right angles into the head?

I see in your original post that you said the the sides of the groove are curved, but then where does the o-ring squish to when the screw is tightened?

I haven't seen this much discussion about o-ring grooves since Feynman dropped one in a glass of ice water.

formatting link

Reply to
rangerssuck

PolyTech Forum website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.