What's new
What's new

Cutting an o ring seal path on a face on a mill

CrustyJuggIerz

Plastic
Joined
Jan 24, 2020
Hey everyone. Im using a manual mill and I'm looking to cut an oring path thats not a simple circle on a face. I'm trying to cut a groove around a water block for a PC, and while the straight edges are dead simple im throwing ideas around about how to do the radii.
My ideas so far are:
An XY table on top of a rotary head to move the cutter to beginning of the rad, offset it accordingly to the centre of the rotary head and then rotate it as desired (a lot of articulation but could be done)

Or, if such a tool exists that allows you to offset the cutting head eccentric but allows it to be indexed around Z, think mill boring bar head but instead of the whole fixture rotating, the fixture can be indexed to whatever angle and the cutting head rotates at a certain offset, if this makes sense to people.

Does anyone have experience with this issue? Any help would be appreciated.
 
Well a CNC would make this pretty easy. I don't have one however. Some years ago I had to cut an o-ring groove in a pump housing that I had surfaced on the mill with a flycutter. It was a bronze end plate. I used a ball end mill and very carefully and very slowly hand fed the X and Y feeds to cut the groove radius. It wasn't perfect but very much good enough. The path of the o-ring in this instance was not critical, it just had to be the proper depth and width for the squish factor. Still running and not leaking after two years.
 
You need a Volstro head. It’s function in life is to cut radii on a mill. Not seen much since CNC is more prevalent, so you may be able to pick one up for a reasonable amount.
 
Depending on the radius size, you could do as previously mentioned and slowly make single axis moves to generate the overall radius. I've done this but while on a lathe, to make a couple new crank handle knobs for my mill. Plot out the X and Y for, say, every 2deg increment of said radius, and move to each point. Won't be perfect, but will likely be fine for the application.
 
That's how I did these on a manual HBM. This was an o-ring groove for a steel mill part, worked great. Everyone at the shop that subbed these to us thought we had gotten a CNC, haha.

I used the DRO's hole pattern function to generate the numbers for the X and Y moves. Didn't take but a minute or two per radius. I fed the entire tool path in one go, so fed straight along the straight parts then used the generated numbers through the radius, then fed straight, then generated numbers through the radius, etc. I forget the increments now but they were barely visible (used a ¼" endmill) unless you got up close and personal.

KJ2T3798.jpg
 
Yeah, that looks great, can't even tell it was fed in increments!

On a related note, this was the handle knob in question I referred to. I drew it on Solidworks, offset the profile by the tool radius, then "sliced" it .015" lengthwise (tool rad), and plotted the X (radial) and Z points all the way down, touched off the tool and essentially treated the tool X and Z zeroes off the rad center, and plunged in stock diameter minus target diameter, then filed, sanded, scotch brite. In this photo you can see the before and after of plunged profile and filing/sanding to smooth finish.

2014-08-19 20.42.52.jpg
 
That's how I did these on a manual HBM. This was an o-ring groove for a steel mill part, worked great. Everyone at the shop that subbed these to us thought we had gotten a CNC, haha.

I used the DRO's hole pattern function to generate the numbers for the X and Y moves. Didn't take but a minute or two per radius. I fed the entire tool path in one go, so fed straight along the straight parts then used the generated numbers through the radius, then fed straight, then generated numbers through the radius, etc. I forget the increments now but they were barely visible (used a ¼" endmill) unless you got up close and personal.

Youz a gots smooth hands
 








 
Back
Top