Announcement

Collapse
No announcement yet.

cutting o-ring groove in stainless--advice?

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • cutting o-ring groove in stainless--advice?

    Hi, I'm working on a project that requires cutting an o-ring groove in the surface of a few 304 stainless disks to seal against a pipe. The o-ring groove width comes out to be .104 +/-.003, the depth .052 +/- .002, and the diameter is around 3.5". I'm curious as to how people here would approach this. I have done similar things before, just not on stainless. The disks are expensive, so I really would like to get this right the first time (and I have no extras)!

    Tools at my disposal include an asian 12X36 lathe and a large cnc knee mill. I am not that confident in the ability of the mill to hold the required tolerances or I would just cut them on that via interpolation. Plus, the scratches in the groove would be the wrong way for good sealing, thus requiring even better surface finish.

    Ideas so far:

    1. chuck disk in 4-jaw, work HSS tool ground narrower than slot (say maybe .060) to make slot full width/depth. Am I going to have problems with work hardening the stainless if I try to take too small of a cut when finishing, etc.?

    2. chuck disk in 4jaw, plunge tool ground to full width into slot and hope for the best. I don't think that this is a very good idea...

    3. Rough out slot using cnc mill and small cutter, say to .010 undersize. Then, using boring head with specially ground tool, finish off groove. Is this going to chatter or what?

    Any additional ideas or comments are very much appreciated!
    Last edited by mwp; 04-30-2006, 10:51 PM.

  • #2
    Given that your CNC mill is questionable (too bad), I would do it all in the lathe in the 4-jaw chuck, provided you have something stiffer than a lantern-style toolpost. The key to success is going to be all about stiffness. I'd get a piece of HSS tool blank at least 1/2" tall by 3/16" wide. You could of course get a wider piece, but then you'd have to do more grinding. You want as much vertical stiffness as you can get. Grind the business end to look from the top a bit like a stubby dovetail. Make the width 5 or 10 thousandths smaller than your desired groove width. Put on some side clearances (try 5 degrees max in a front view and 3 degrees in a top view) and end clearance (try 10 degrees). Test the end view against a 3.5-inch circle template to check your clearances. With the template, check that the "heel" of the tool doesn't rub on the OD of your nascent groove. Also check with a calipers set to 0.104 inch that you can get all the way to the right depth without hitting the shoulders of the dovetail. However, don't make the dovetail any longer than necessary.

    Put a good 5-degree or so hook on the top of the cutting edge. Most o-ring grooves call for a radius at the corners, so put that on if the drawings say so, guided by a radius gage or a drill shank. After grinding, stone everything really sharp with an oilstone or similar. I have an Accu-Finish grinder, so I'd use that, but it's just a convenience, not a necessity.

    Given that you have no extra pieces, I'd take a practice run on a hunk of aluminum or whatever, to get the starting radius set right, and to check that the tool has clearance enough to cut, and makes a nicely curled chip. Once the radius is dialed in, do not lose the cross-slide position! With a DRO, just zero it out. Otherwise, remember the dial position, or just don't move the cross slide as you change to the real workpiece.

    When cutting the stainless, go at a slow surface speed (try 60 FPM, say 110 RPM), continuously goop on lots of sulfur oil or your favorite cutting oil, and plunge right in. DO NOT pause in the cut; keep the tool cutting a continuous chip. Try to push it as hard as as you think the machine and setup will stand. Use your ears and listen to the gears and motor. To control depth, you can use an axial DRO, a dial indicator, a carriage stop preset to the right distance ahead, or set the compound slide at 90 degrees and feed with that. If you want to back out before you're finished, do it quickly, so you don't pause and rub the tool in the cut. Get back into it fast, too.

    Once you've got the first plunge over with, check the groove width, then move the cross-slide whichever way you need to, to shave the groove wall to final width.

    Comment


    • #3
      If this is a face groove wouldn't it be easier to use a 3-jaw chuck, especially if multiple parts need to be made? The groove circle may be off axis a few thou but that shouldn't make any difference.

      I did that recently in acrylic and it worked just fine. The run was 40 parts and I didn't want to mess with the 4-jaw for that many.

      Mike
      Mike Henry near Chicago

      Comment


      • #4
        Machinery's hanbook has a writeup on trepanning or face groving that explains the procedure. I would turn the part slow or flood coolant to prevent heat. I prefer a four jaw chuck or a 5c collet but a three jaw would work.
        my two cents worth
        re
        Herm Williams

        Comment


        • #5
          Had a similar job in 304 years ago. This was a production run with several other operations in addition to a 1/4" wide X 1/4"-3/8" deep X 5" D. groove with full radius bottom. Rigidity will be the key in this operation. I did this on #3 W&S, using a Valenite trepanning tool and carbide insert of the dog-bone style. Don't remember much about the job except that the tool had to be set over center, and flood coolant was used.
          Harry

          Comment


          • #6
            Thanks for all the replies. I will definitely look into the machinery's handbook write up. Today I'm going to go grab a chunk of scrap 304 stainless to practice on. I may even give the cnc mill a try. It has the advantage of having coolant, plus I have to drill a hole circle so that will avoid another setup. I think I have my friend talked into using the next size up in o-rings, so the tolerances are a little looser and the cutters for the mill not so fragile.

            Comment

            Working...
            X