Please Help: First Time Using Fixtures (w/ Photos)

[M30]

I'm new to machining and I have been searching around about fixturing techniques for CNC milling. Unfortunately, there does not seem to be a lot of solid information on it. The basic idea as I understand it is to make part to hold the workpiece in a clamp or vise where otherwise you could not. It's also become apparent to me that fixtures can be used simply to process multiple parts in one setup.

So far, I am convinced that there may just not be a lot advise to give on fixturing since the fixture will vary greatly with the part. I think I am on the right track, but if anyone would like to give some advise on my current project, I would appreciate it.

Currently, I am working on a prototype of a gerotor pump for compressible fluids:


I will only make a few of these as I designed it simply as an exercise in milling. I have a TM-1 mill and the 4th axis but don't yet feel comfortable using the 4th. I'll be leaving that adventure for another day. Also, I am using machinable wax at the moment but plan to make a couple out of aluminum once I get the setup figured out. It's good thing I am using wax considering how many bad attempts I have had now.

My approach to milling the rotor was to make 2 halves that I could use to clamp the part in a vise. The halves do not touch each other, ensuring all the clamping pressure is on the part.



The problem I am encountering is that the part vibrates loose and comes up out of the fixture. I suspect this may be compounded by the slipperiness of the machinable wax and that the vise jaws do not reach the top. Also, the fixture and part is a very tight fit, I'm concerned that will be a problem with aluminum.

But for now, what I was considering next is to redesign the fixture to be held with counter-bored screws. Since the height of the vise does not reach the portion of the part with the most surface area, I was hoping this would keep it from coming loose.



My next challenge is the pump housing. I designed it so that it could be milled from three sides. I came up with a fixture to hold it for the 2nd and third sides. In this case, the part itself will be bolted to the fixture:


Also, does anyone have any tips about material stock? Currently I have been facing the stock on all 6 sides to fit the precise bounding box of the part. But it seems I can't get the stock truly square. It will be off by as much as .15 under at times and the stock is not very square to begin with. I suspect it is because the stock is not sitting square in the vise even though I faced the other sides. Perhaps I should be leaving extra material for that reason rather than facing the stock to a precise 2x2x4 for example?


So am I way off base here or am I right on track? Thanks!

[Geof]

You are on track, really very well on track with only one error that I see on a quick read; this is profile jaws that you hold in the vise where the part climbs out. It seems these are held in the vise jaws and not attached to them and if this is the case you need to have the vise jaws come all the way to the top so your clamping pressure from the vise is inline with the part which is negin held. Probably what is happening is that the vise jaws are closing the bottom of your profile jaws more than the top so they are assuming a wedge/taper position and squishing the part out.

[M30]

Thanks Geof. Wedging caused by the vise definitely makes sense - I think intuitively I knew that. So it sounds like the second design of the rotor fixture held together with screws might work?

held in the vise jaws and not attached to them

How do you attach a fixture to the vise jaw? I have a Kurt vise but don't know the model #.

[HuFlungDung]

Are you making the shaft an integral part of the rotor? This is asking for trouble, because the shaft is almost guaranteed to be eccentric when you flip the part over.

Its easy to spit out the rotor (minus the shaft) if you make it in a conventional manner, for instance, as the boss on top of an over-thickness plate, complete with accurate bored hole, then use your vise fixture to hold the part when you flip it to face the excess thickness off.

[M30]

Yes, the shaft and rotor are all one piece milled in two steps... or that was the plan. I didn't mention it, but my only successful try with this setup the shaft on the other side was indeed eccentric. I thought it was because I was rushing to get home that day and I just wanted to try out the fixture.

So what you are proposing is to have the shaft and rotor 2 pieces and then assemble them later? I had not considered that, but it would take acceptable compromises to the design.

Somewhere in the rotor there would have to be a set screw to hold the shaft in place. I suppose I could make a boss on the rotor so that the screw sits outside any of the meshing/sealing surfaces. I will have to ponder it a bit more on how this can be done.

Thanks!

EDIT:
Hmmm... Thinking about it a bit more this could be tough. If I have a boss only on one side of the rotor for the set screw, then that means I can't use the same housing for both sides as one would need a relief. One of the design goals for this pump was to ensure the housing halves were symmetric to ease setup and programming.

I suppose I may have to learn how to use the lathe we have. Then I could turn both shafts and mill the rotor. But then the fixture would only hold the rotor by the shaft. I would think this would not be rigid enough to keep the part from flying away.

[pointcloud]

I agree with HFD.. But if you insist on haveing it be 1 part... Make the vaines on the 4th axis, then turn it out.. Put the center holes in on the lathe and then use the same one's in both processes? That's the way to have it one pcs and close to streight...

That's a good looking pump... It'd be a great one used externally for turbo applications... I made one simular for my sond go-kart, it was probablly smaller.. I had a tiny turbo, worked ok for a little.. The head gaskets started going...
I used a toyota pump.. I did not wat to go to all the trouble hardening it, if I had made it.

On the houseing? Why not start the other way? Do the inside first? then no need to make some funky thing to hold it down? Do the inside then it is flat? Hold it down with the shaft hole that you drilled to line up the oppisite side with? Get your planges and the rest all finished up, the remove the bolt, and place bolts into the newly dinifhed flange holes.. Fix up the top next?
Good luck.
M

[pointcloud]

No need for set screws? Use a MOON key?

[Mitsui Seiki]

How do you attach a fixture to the vise jaw? I have a Kurt vise but don't know the model #.

Easy,you remove the vise jaws(if it's possible) and replace them with the halves of your fixture.

[Geof]

Originally Posted by M30 ...How do you attach a fixture to the vise jaw? I have a Kurt vise but don't know the model #.

Make the jaw the fixture. Get a piece of cold rolled steel, drill and counterbore for the bolts to attach it to the vise jaws and then machine your profile in these 'soft jaws'. Grip a piece of1/8" aluminum when machining the profile so the jaws are not closed together.

This is the way to do it for production work so every piece goes in to the same position.

Regarding making the shaft and rotor in one piece not only is it difficult to get the shaft true it is wastefull on material. Make the shaft separately and have a larger diameter in the center that fits in the rotor and maybe also have a shoulder on this with a matching recess in the rotor. Cut a Woodruff keyway in the shaft and broach a keyway in the rotor and size everything for a press fit. This obviates the need for a setscrew. If you are worried about doing a press fit make a slight recess in the rotor on the side opposite the shoulder and have a circlip on the shaft to retain the rotor.

I think you are being ambitious making a gas tight gerotor type pump; I don't know of any other type of compressible fluid.

[M30]

pointcloud,

Not sure if you saw the edit after my last post. I may consider turning the shafts and milling the rotor.

The pump may be smaller than it looks. The inlets have a diameter of .91" to give you a reference point. The OD of the housing is 3.5".

The inside of the housing is flat, and I considered doing what you said (and maybe I should reconsider). I would still need something to stand off the part so I can profile around the mounting tabs. I was also concerned that using one bolt to hold it down would not be enough to keep it from rotating around the bolt. Maybe I am overestimating the forces applied during milling?

[HuFlungDung]

I've seen several pumps like that, that do indeed use a woodruff key (half-moon) with thin spiral rings on each side of the rotor to keep the shaft from falling endwise out of the pump. The rotor can have a shallow counterbore on each face to keep the spiral rings a little below flush.

Another method would be a simple interference fit. If this is a low pressure pump, it may not have to transmit a large amount of torque from the shaft to the rotor anyway.

Still another method would be to counterbore the rotor about halfway through on one side, and fabricate a flanged shaft to fit the counterbore. Then, provide a couple of small screws or pins to drive the rotor.

[M30]

Geof,

I'm liking the press-fit idea. But I was avoiding the keying because I was unsure of how to cut one in the shaft.

I think you are being ambitious making a gas tight gerotor type pump; I don't know of any other type of compressible fluid.

Indeed But I wanted to try something that required me touching on a lot of different principles. If I can successfully build this, then I can make just about anything else I would want. My other projects will not likely exceed this complexity anytime soon.

This design should double for incompressible fluids by using a different valve plate on both sides. Basically it will not compress the fluid rather just push it through avoiding hydro-lock.