Re: Challenging O-ring Groove

Is this close to what you need?

Mike,

Thank you for your help and quick response. Unfortunately I'm still running

Mike,

Oring grooves are always square. Try it with a rectangle

Regards

Mark

Cubs,

I emailed you an example of what I think you want in 2001+. Real simple, no sweeps or surface cuts.

Regards

Mark

here's the trick the groove

.090 long. The long side of

bottom and top of the

and the top of the groove

an angle around the

one o-ring, rotating the

groove. I created an angled

Now that I think of it, how the hell are you gonna cut that groove ????? It's impossible !

You'll have to make it out of three pieces, a central cylinder, and two slash cut rings. You'll also have to make a fixture to press, or pin, the three pieces together.

Regards

Mark

Well,

My email got bounced back

If you want the model, email me privately with a good return address

(remove the obvious from my address to reach me)

Mark

here's the trick the groove

.090 long. The long side of

bottom and top of the

and the top of the groove

an angle around the

one o-ring, rotating the

groove. I created an angled

Regards Darryl

You'll have to make it out of three pieces, a central cylinder, and two slash cut rings. You'll also have to make a fixture to press, or pin, the three pieces together.

Regards

Mark

I see! OK, here's another try in 2001Plus...

Mike,

That one's allot more machinable than what I came up with. The walls are normal to the bottom of the groove. There is something wierd going on though, (at first glance). It looks like the bottom doesn't have a constant width.

Regards

Mark

Mike,

OK... mine doesn't have an equa width bottom either, it's just harder to see. So, it looks like this really is challenging. I "KNOW" that if I write a CNC program for my fourth axis, using axis substitution (where the "Y" is swapped for the 'A") I can machine it. If the cutter has a dead flat bottom, and if I follow the center line of the groove with the axis of the cutter right on center with the cylinder.

I've been trying for about an hour to duplicate this in SW, no luck. Think sweeping a cylinder, not a 2D profile along the curve. The axis of the swept cylinder is allways pointed right at the center of the cylinder being cut. Don't think the tools are in there to do this.

Regards

Mark

That's funny, It's definitely not as easy as I first thought. Maybe third times a charm?

Does anyone else have a solution they would like to share? I would be interested in seeing it!

Mike Wilson

yes I did a hybrid solution-used a circular profile to cut sweep around an elliptical path to define the correct edge profile at the cyl surface-this is not actually a straight line in profile-and then used the 2 outside surfaces to thicken into 2 body rings, made into a whole with a merging cylinder-hid the original body. hope you followed that folks

Mike,

I think this approach looks pretty close to the angled grrove.....much better than mine.

Thanks

Mike,

Very good ! I was working on something similar when I ran out of gas last night.

The cross section perpendicular to the groove at the center is about .086, but it's real close.

This will go into the tips and tricks folder on our server

Regards

Mark

Thanks Matt! I think I like your method the best. I knew there was a way to do it with a sweep.

Regards, Mike

"Mike J. Wilson" wrote in news:4Nudne51x_ snipped-for-privacy@giganews.com:

Mike:

I was going to try to make a point about it being better to do it using solid features rather than surface features ("that old chestnut" to quote Dr. Evil), but I found that mine (solids) rebuilt in about 2 seconds and the second one you did (surfaces) rebuilt in 0.8. Your first one rebuilt in 3.4. I'm curious as to why the thickened part rebuilds so much faster than the other methods. Any ideas?

matt

"Corey Scheich" wrote in news:bhr3n4$28tkb$ snipped-for-privacy@ID-200385.news.uni-berlin.de:

Yeah, you're right, I made that mistake the first try. In the model I posted, I used a dummy point to make the pierce to the GC and used a dimension for the width.

matt

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

I would except that I don't have 2001+ on my machine anymore. Would a parasolid (no features) be of any use?

Matt