[Bradyz]

I don't have very much experience with lathes especially when it comes to threading and doing curves. That being said;

I make and sell hundreds of parts that are exactly the same, one is a rod that is about 10" long, 8mm diameter and has about a half inch of 5/16" x 24 threading on each side. The other part is the part in this picture:

What would be the most efficient way to make these parts? I am currently farming out the work but I would really prefer to do it on my own. I have a VMC so I am familiar with cnc work if that would be the best solution. I would really like to stay under $20k. Any help would be really appreciated. Thanks.

[Geof]

If your budget is only 20k you may be better off continuing to farm them out. That part with a thread and a ball is trivial to make on a screw machine and quite practical to make on a small CNC lathe equipped with a bar feeder, but this type of equipment costs several times 20k. You could get a small CNC but then you will be dealing with a manual chuck or collet and you would have to advance the bar stock by hand. Doing hundreds of small parts this way teaches you the meaning of the word tedious and your time is probably too valuable to spend hours in front of a machine getting brain dead.

[wizard]

My first thought is what is wrong with farming them out? A modern shop would do several hundred of these in a day and that would include setup. They should be able to do those rather cheap too.

Of course you may have good reason to do them yourself. As has been pointed out though this type of work would require a bar feeding machine to be fully automated. Even then I'd have to wonder about that hex and the ball. Ideally you would want to feed hex bar stock with the idea that the ball fits within your hex size. If the ball is bigger than the hex minor diameter then you have an expensive secondary operation or you will find yourself in need of a fancy lathe with advanced capabilities.

If at all possible I'd make sure that the part above could be machined from commonly available hex stock. Obviously surface finish and tolerance may be an issue but it would make for a job that can run on a much cheaper lathe.

The thing is this, you might be able to find a simple used CNC lathe with a bar feeder in your price range or maybe if you are inclined you can make a bar feeder and integrate it into the controls. That can only be done cheaply if the lathe isn't required to mill the hex.

I don't have very much experience with lathes especially when it comes to threading and doing curves. That being said;

I make and sell hundreds of parts that are exactly the same, one is a rod that is about 10" long, 8mm diameter and has about a half inch of 5/16" x 24 threading on each side.

This is certainly where CNC shines but the problem is you are only talking about hundreds of parts per month. For simple things like this that is hard to justify when it comes to new CNC lathes with all the bells and whistles to automate this type of job.

The other part is the part in this picture:

What would be the most efficient way to make these parts? I am currently farming out the work but I would really prefer to do it on my own.

That is a funny question because what do you mean by effciemt? The reality is that no matter what you do, new used or something improvised the machine will sit around for a good portion of every month with no work to ripun on it. That isn't very effcient if you ask me. Now in some industries you can get away with that, but with these little parts you would be wide open to competition from other sources.

I have a VMC so I am familiar with cnc work if that would be the best solution. I would really like to stay under $20k. Any help would be really appreciated. Thanks.

The problem is a few hundred parts a month puts you in a difficult position. It isn't really enough work to pay for an expensive machine as you may find yourself only running it one or two days a month. On the other hand if you are manually loading a cheaper CNC lathe with your parts you will be wasting a lot of your time with very tedious and possibly error prone work.

There might be acceptable new machine options such as: Web which is marketing a small lathe with a collet closer option. I have no idea what they charge for such a machine though. However the machine is small enough that manually feeding might not be a huge issue. Remember though the ability of such a limited lathe to do the job is dependent upon the parts design and as it is the lathes capabilities.

The real issue with respect to making this all work is your programming skills. With the right programming effort you can get a simple lathe to do much for you. With a bit of creativity with tooling, even manual chucking of the parts might become viable but this would require a bit of tooling engineering and a bit of custom programming.

[metlcutr55]

if your vmc has rigid tapping, or if you have a releasing type die head you can make the rod on it from sawn blanks. using an extended er collet holder in the spindle to hold the rod, do 1 end, flip, repeat (use lathe threading tool in vise, and a facing tool if needed). multiple passes for the "rigid tap" method or just 1 for the die head. use a big dia collet holder as possible for rigidity if you do multiple passes. shouldnt matter much with a die head. as an afterthought, im not sure of the effect of dragging the lathe thd tool backwards on the return using rigid tap would be good, so multiple helical interpolated passes might be a better choice if your control has the ability to "time" them.

depending on matl toughness and criticality of lengths, maybe reverse this setup holding the collet holder (and rod) on the table, just thread mill ea end would be sufficient. maybe raising a 5c collet closer up off the table with a spacer to usable height would be better for rod holding.

the ball part id also saw a blank, from hex if possible, interpolate the thd dia & thds (and mill the hex if no hex stk available). 2nd op using your spindle as a lathe, and a holder with a female threaded insert in the end (that you would make), snug the part in there with a wrench, use left & right handed lathe tools in the vise to turn the ball end. all this provided the part is not so large as to overwhelm your machines rigidity, or made of super tough matl that would do the same. multiple light cuts will help reduce the tendency for the part to overtighten in the holder. you may need to experiment a bit. perhaps an occasional dab of never-seize in the thded holder.

efficient? depends on if you have enough open time. if i have a job i must make on the cheap, the son in law will take $10/hr in hand, i just have to make the setup and cycle foolproof and provide enough easy checks to verify quality.