So in an effort to try to improve my armature designs, I am experimenting with milling ball seats into ball and socket joints. Normally, I just drill a 1/8th inch jobber hole through two pieces of mild steel stock which has worked just fine. However, when I try to make a joint with the milled ball seats, I find that I am getting a lot of drift with the movement in the joint- obviously not an improvement from my previous technique. Now I am sure I am getting really fine tolerances as I am doing this on my lathe and I am using a dial indicator with a manual adjustable four jaw chuck.
So I am just curious to see what the other ball/socket machinists experiences are. Are ball seats made with a ball end mills much better than those made with the jobber drill? Does anyone have tricks that I may be missing here? I would be happy to give more info on my technique if needed.
Let me share my experience with this topic. When I first tried milling sockets for an armature, I was using a ball end mill with a table top mini mill. What I was finding was that the socket I milled did not seem to be the exact same shape as the ball even though I was using a 1/4" ball mill for a 1/4" steel ball. It seemed to be bottoming out, leaving the ball at the top area of the hole rather loose. Not a good fit at all. People advised me to drill a larger hole in the bottom of the socket and that would solve the problem. It helped but it was obvious that the socket was somehow not the right size. I took a good look at the hemisphere shape on my end mill and realized it was not the correct radius. For a 1/4" ball end mill the radius should be 1/8". (1/2 the diameter) but it looked like the rounded end had a larger radius to it. My end mills, I should mention, were purchased as a set and were very inexpensive. I had a theory that a more expensive end mill would have the proper radius so I bought a 1/4" ball end mill from McMaster-Carr. It cost as much as the entire set I had purchased before. Just looking at it I could see that it looked like a better hemisphere shape. I then tested this by milling 2 sockets side by side in a piece of steel. I dropped a ball into each socket. The one made with my old end mill rattled around in the socket as had happened before. The ball in the socket made by the new end mill fit snug and tight. It was a perfect fit! I then proceeded to make a simple sandwich joint with the new end mill and it worked absolutely perfectly. No wandering around and no need for additional holes in the bottom of the socket. I don't know if this is exactly the problem you are having but it is definitely something to check. I brought this up in another thread a while back, but nobody really seemed to pay any attention to me. They just kept talking about drilling holes in the bottom of the socket which is not a solution of any kind. If the socket is not the same diameter as the ball fitting into it, then you may as well just drill large holes and use that to hold the ball. But that is my experience.
Well, I'm paying attention. :-) I asked a similar question a long while back and got similar answers - drill out the bottom of the hole. Effectively all I was doing with the ball mill was recessing the open hole surface. My holes are the same as your experience: loose at the top
Was there some characteristic (measurement, tolerance, brand, etc.) about the good bit that would help identify it as "right"? For example, if looking at two bits in an online catalog, is there something other than price that lets you see that it's actually a proper shape?
I really don't have any particular measuring system that told me it was not a true hemisphere shape, although I did hold up a circle template for a 1/4" hole and it did not look the same. This was all strictly by eye, but it just seemed obvious to me that this was the problem and because it was a slightly larger radius than the ball, explained why the ball was loose. I actually talked with someone who runs a cnc mill about this issue and he told me that the end mills he uses have a true hemisphere on the end. He was using more expensive end mills so I simply took a chance that the pricier item would be better, and in this case that was the truth. Paying $25- $30 for one end mill may seem like a lot to us amateur machinists but in this case it will save me a lot of head ache trying to make joints work that simply were not machined correctly.
This is very interesting! Since posting, I have tried making a joint without drilling a through-hole first, and it did make an improvement. It was similar in performance to just drilling the 1/8 inch jobbers as per my original method. I am sure it would work well during animation; however, I am developing an almost obsessive-compulsive curiosity to see if I can improve a the movement in the socket. I am very curious to see if buying a better end mill with do just that. I have used Mcmaster Carr (granted the endmill I am using isn't Mcmaster carr) and was quite satisfied, but now I wonder if there is another company/brand that may make end mills to an absurdly high quality?
I have to look back in Tom's book,he covers "drifting" issues with ball socket joints.I can't recall the culperate at the moment,but it think it might be a location issue or possibly the depth of the ball seat(or both?).I take it you are using a "ball endmill" to mill the ball seat,correct?Are you using a ball endmill that is the same diameter as the ball bearing?How deep are you maching the ball seat,and is the location and depth machined identical on both mating parts?That is crucial.Also,the distance your tensioning screw is to the hole can affect the ball joint performance.The location and depth of the ball seat has to be "dead nuts on".I think if it is off by just a thousandth or two,you will have drifting problems and maybe even "ratcheting".As per your querry on quality endmills,YES!You can buy high quality endmills both in HSS,Cobolt and solid Carbide.MSC Industrial supply carries all of these items(plus the crappy imports).Sta away from crappy imports,like I've always said..."You get what you pay for".