practical manual 4th axis/indexing for several parts at once?

[laggeddag]

So I am working on a project where there is a part about the size of a paperback.

It has several holes and other geometries that would require several angled setups, although thankfully, it would only need to be rotated around one axis.

This makes me think I could make a simple indexing jig that is essentially an indexer, but with an extended platform and a support bearing on the side opposite the indexer. This would allow me to make a fixture that would hold several parts (they are only about an inch thick) at once, and rotate them manually so that the various angled holes could be drilled.

My question is, is making a fixture like this practical? It seems like it should be. Two bearings blocks with some nice bearings, and a manual indexer that would allow me to rotate the parts.

I'm concerned about accuracy, however given I could justify using some pretty good bearings and making the support blocks pretty much as thick as I like, and since the parts are only about 3" radius from the axis I'd rotate them around, I'm hoping I got lucky here and this is a practical solution.

I want to be able to use the fixture without any indicating. Just hit a mandatory stop in the program, then index manually, then continue the program. Of course any amount of indicating while setting up the fixture is fine. But I suppose overall this means I need it to have some pretty good repeatability.

I suppose, you could make a tenths dial indicator part of the fixture, and as you manually index it, it could tell you how far you are from intended zero and maybe even have a little thumbwheel you could fine tune it to. That wouldnt be time consuming and so thats okay. (It sounds like I should use a good rotary table instead of an indexer).

My CNC does not have the ability to use a probe and I think this could be done without it anyway, this seems like a very straightforward thing to do, although I have not done it before and I'm hoping its a routine sort of thing.

I cant afford a CNC 4th axis and even if I could I suppose I'd still have to build the platform and support bearing on the other side anyway.

I definitely cant afford a 5th axis and I think I'd need a giant one to fit as many parts as I'd like anyways. This part would not justify the cost of that anyway.

 

[cameraman]

So I am working on a project where there is a part about the size of a paperback.

It has several holes and other geometries that would require several angled setups, although thankfully, it would only need to be rotated around one axis.

This makes me think I could make a simple indexing jig that is essentially an indexer, but with an extended platform and a support bearing on the side opposite the indexer. This would allow me to make a fixture that would hold several parts (they are only about an inch thick) at once, and rotate them manually so that the various angled holes could be drilled.

My question is, is making a fixture like this practical? It seems like it should be. Two bearings blocks with some nice bearings, and a manual indexer that would allow me to rotate the parts.

I'm concerned about accuracy, however given I could justify using some pretty good bearings and making the support blocks pretty much as thick as I like, and since the parts are only about 3" radius from the axis I'd rotate them around, I'm hoping I got lucky here and this is a practical solution.

I want to be able to use the fixture without any indicating. Just hit a mandatory stop in the program, then index manually, then continue the program. Of course any amount of indicating while setting up the fixture is fine. But I suppose overall this means I need it to have some pretty good repeatability.

I suppose, you could make a tenths dial indicator part of the fixture, and as you manually index it, it could tell you how far you are from intended zero and maybe even have a little thumbwheel you could fine tune it to. That wouldnt be time consuming and so thats okay. (It sounds like I should use a good rotary table instead of an indexer).

My CNC does not have the ability to use a probe and I think this could be done without it anyway, this seems like a very straightforward thing to do, although I have not done it before and I'm hoping its a routine sort of thing.

I cant afford a CNC 4th axis and even if I could I suppose I'd still have to build the platform and support bearing on the other side anyway.

I definitely cant afford a 5th axis and I think I'd need a giant one to fit as many parts as I'd like anyways. This part would not justify the cost of that anyway.

It's certainly possible, but also depends on what other equipment you have to make your rotating fixture + your tolerances.

^^^ Peter Stanton made about six videos about how he built a 5 axis mill-stop fixture so look at the two or three videos before and after this one. But he did that mostly in his home shop garage [With an older HAAS toolroom mill (8000 rpm spindle) and HAAS toolroom lathe ]rather than on his bigger MAZAK 650 series H i Integrex thing / machine.

BUT also he did use a HAAS 4 axis rotary to drive the other half of the trunnion he built.

You'd have to build a full manual trunnion preferably with fine adjustments and critically you would have to design a breaking system that would handle the in cut torque and be sufficiently rigid for decent surface finishes/ cuts.

There's a myriad of ways to do that (again depending on what you have access to). + $.

I know here on PM forum there are probably twenty different methods that could be offered / suggested.

Maybe throw up a sketch of rough dimensions (doesn't matter how bad / rough) for what you have in mind and some sense of the kinds of cuts you want to make to get an idea of the forces / torques that may be needed to take into account.


Peter Stanton's video covers one half of what you want including bearing fits, just needs the mirror image and an indexing, breaking/ rotational clamp and part clamping system you might be happy with...

[Interestingly the breakings systems (depending on design) of 4th and 5th axis (CNC) trunnions are a source of pretty significant error in some cases + backlash that has to be mapped out / compensated ~ So theoretically a really accurate manual indexer / trunnion can be built.]. For a manual indexer (depending on materials used a split hub on a large diameter shaft design could be rigid and precise enough as a breaking / clamping system and it would be theoretically possible to incorporate a fine adjustment using a large micrometer (if you needed it). Alternatively you could build a curvic coupling (tricky) or a myriad of other things (off the shelf) that you can hack for 90 degree rotations - again depends on $ you have to spend.

OTHERWISE ~ Hardinge does actually sell manual indexing equipment. like a 4th axis / trunnion. There are other variants from different companies too.


Maybe tomorrow I dig up a catalog / PDF. [Depends on your part tolerances / rotational tolerances, materials, spindle / machine capability and geometry of your parts.

Do even post Millennials even know what a "Paper back " is lol :-) (even if they are fan of the Beatles .). (Smaller than an i-pad but larger than a smart phone ? ). Trade paper back 5.5” x 8.5” x how many pages thick (like 1" ?) .

 

[laggeddag]

It's certainly possible, but also depends on what other equipment you have to make your rotating fixture + your tolerances.

^^^ Peter Stanton made about six videos about how he built a 5 axis mill-stop fixture so look at the two or three videos before and after this one. But he did that mostly in his home shop garage rather than on his bigger MAZAK 650 series H i Integrex thing / machine.

BUT also he did use a HAAS 4 axis rotary to drive the other half of the trunnion he built.

You'd have to build a full manual trunnion preferably with fine adjustments and critically you would have to design a breaking system that would handle the in cut torque and be sufficiently rigid for decent surface finishes/ cuts.

There's a myriad of ways to do that (again depending on what you have access to). + $.

I know here on PM forum there are probably twenty different methods that could be offered / suggested.

Maybe throw up a sketch of rough dimensions (doesn't matter how bad / rough) for what you have in mind and some sense of the kinds of cuts you want to make to get an idea of the forces / torques that may be needed to take into account.


Peter Stanton's video covers one half of what you want including bearing fits, just needs the mirror image and an indexing and clamping system you might be happy with...

OTHERWISE ~ Hardinge does actually sell manual indexing equipment. like a 4th axis / trunnion.


Maybe tomorrow I dig up a catalog / PDF. [Depends on your part tolerances / rotational tolerances, materials, spindle / machine capability and geometry of your parts.

Do even post Millennials even know what a "Paper back " is lol :-) (even if they are fan of the Beatles .).

Nice thank you for all that!! That looks like what I'm thinking (the 4 axis one), except I'm hoping to get away with a manual indexer of some sort.

So, as far as being able to resist cutting forces how about:

-bearing blocks like that video

-platform supported on the bearings that allows workpieces to be mounted

-on the other side of one of the bearings, an axle stub that is attached to a plate. The plate has bosses machined at the exact locations they need to be so that as you rotate the platform, the plate rotates, and the bosses hit a stop (flat machined surface onto flat machined surface). The bosses have hole through them and the stop has a threaded hole so you can actually clamp the bosses to the stop, providing the cutting-force resistance as well as the precision location. If this idea ends up not working I can just unmount the plate and try something else at least.

Not sure how the bosses can rotate without hitting the stop if you need to pass a boss...but lets just put a pin in that.

Or, instead of that, a manual rotary table that has a clamp? I havent used one of those before, I'm guessing the clamp introduces some error?

The cutting forces shouldnt be too bad. The material is 7075-T651. Most of the operations are literally drills under 1/4" diameter. There will be some milling but its going to be end mills under 3/8" diameter and not much material is going to be removed.

I am still working on the toolpath and list of angled operations so I am not sure yet what I'm facing here. I would be concerned that a long platform would resonate and some cutting frequency, even if its a short operation. So its probably going to have to be pretty stout.

I bet there are some clever ways to do this I am not even considering. I dont actually need the general purpose nature of infinite-angle-adjustability. I would actually prefer some method that clamps only to the specific angles needed. I haven't thought up how I'm actually going to hold the workpieces (they are paperback sized but are not rectangular, they are work-in-progress when they get to this rotary setup and will have an odd shape. I would like them to be quick to fixture. In needs to be a precision method of locating them and allowing them to be rotated, all in sync with each other)

 

[dandrummerman21]

Or, instead of that, a manual rotary table that has a clamp? I havent used one of those before, I'm guessing the clamp introduces some error?

TOUCHDEX - KAWATATEC CORP.

YUASA 710-230 - yuasa-intl (you would need to look at their catalog for more "rapidex" versions)

We have a touchdex fd-200, works pretty good. extremely rigid.

 

[laggeddag]

Maybe throw up a sketch of rough dimensions (doesn't matter how bad / rough) for what you have in mind and some sense of the kinds of cuts you want to make to get an idea of the forces / torques that may be needed to take into account.

Yes I'm going to work something up. I'm just getting familiar with this part. I might need to work on how the part will be fixtured to this trunnion first, since I'm not actually sure how I need to do that yet.

I definitely do not have a big budget for this workholding. At least the machining shouldn't be too demanding. But this is going to have to be effective, fast, and solid. At least it does not need to be automatic/cnc, 5 axis, infinitely adjustable (beyond fine tuning position manually.

 

[DavidScott]

What angles do you have to index to? My solution was a Phase II 5c collet indexer but it indexes no finer than 15 degrees. Since I DO NOT load parts into fixtures while the machine waits I made a tailstock to hold the other end pinching the fixture between it and the indexer currently with up to 360lbs of force. This has been my setup for a little over 20 years now. Here is a video showing it in use with a shorter fixture and a 4th axis, I still have the indexer but haven't used it since I got a mill with a real 4th. The tailstock is the same one I made 20 years ago but updated as it was adapted to 3 different VMCs and rebuilt when it got close to .001" of slop. I have never had a problem with stiffness while machining with this setup, and what I am doing in the video is middle of the road for the parts I make with it.

One part I have been making for 20 years is 3/8"x3"x3" with a cross hole that I drill down .5", index 180 degrees and drill the same. They have to be in line with each other within .005" total deviation, any more and the part is rejected. Even the indexer didn't have a problem holding this tolerance.

 

[kazlx]

You should be able make a tombstone that clears your parts and drill/ream/pin the end at the required angles. Even if you had to back off the programming to take lighter cuts, you could make up for it with not having to indicate subsequent ops.

 

[laggeddag]

You should be able make a tombstone that clears your parts and drill/ream/pin the end at the required angles. Even if you had to back off the programming to take lighter cuts, you could make up for it with not having to indicate subsequent ops.

okay so to be clear this would be tombstone mounted so it rotates on its long axis, and the drill/ream/pin is for alignment of the tombstone with mating features on the bearing support blocks that lock it to a certain rotation right?

yes I really must avoid indicating ops unless its super fast like step 1: MAJOR rotation to an alignment point 2: look at dial indicator and twist a thumbwheel to get into tenths alignment 3) press green button

I cant be doing a) major rotation b) install dti in spindle, do a bunch of jogging and get the plastic hammer out to get things aligned..nope nope nope nope

 

[Delw]

So I am working on a project where there is a part about the size of a paperback.

It has several holes and other geometries that would require several angled setups, although thankfully, it would only need to be rotated around one axis.

This makes me think I could make a simple indexing jig that is essentially an indexer, but with an extended platform and a support bearing on the side opposite the indexer. This would allow me to make a fixture that would hold several parts (they are only about an inch thick) at once, and rotate them manually so that the various angled holes could be drilled.

My question is, is making a fixture like this practical? It seems like it should be. Two bearings blocks with some nice bearings, and a manual indexer that would allow me to rotate the parts.

I'm concerned about accuracy, however given I could justify using some pretty good bearings and making the support blocks pretty much as thick as I like, and since the parts are only about 3" radius from the axis I'd rotate them around, I'm hoping I got lucky here and this is a practical solution.

I want to be able to use the fixture without any indicating. Just hit a mandatory stop in the program, then index manually, then continue the program. Of course any amount of indicating while setting up the fixture is fine. But I suppose overall this means I need it to have some pretty good repeatability.

I suppose, you could make a tenths dial indicator part of the fixture, and as you manually index it, it could tell you how far you are from intended zero and maybe even have a little thumbwheel you could fine tune it to. That wouldnt be time consuming and so thats okay. (It sounds like I should use a good rotary table instead of an indexer).

My CNC does not have the ability to use a probe and I think this could be done without it anyway, this seems like a very straightforward thing to do, although I have not done it before and I'm hoping its a routine sort of thing.

I cant afford a CNC 4th axis and even if I could I suppose I'd still have to build the platform and support bearing on the other side anyway.

I definitely cant afford a 5th axis and I think I'd need a giant one to fit as many parts as I'd like anyways. This part would not justify the cost of that anyway.

your over thinking everything.

In the past ive used a Hardinge manual 5C collet closer for a 4 axis tombstone, still have it in case my full forth takes a crap or is tied up on another machine. I use a tail stock to support it like David, or I have a a-frame thats alot more rigid with a bushing.
if you want to make fixtures so you can take one out load parts while other is running, or what have you. just get a over sized collet like a 2"dia or so. bore it out to 1.5" and and drill and ream a 1/4" dowell pin in the collet face. then cut your tombstone fixture to 1.5 round and pop a 1/4 hole slip fit for the pin. (use 2 pins for better alignment and repeatability)

for the support. bearings are not necc due to the low speed it will turn, a bronze bushing is more than enough or a tail stock.unless you press the the bearings on to the fixture(light press) are going to give you more accuracy issues than a bushing. (unless you buy good ones)

as far as accuracy, its only as good as the equipment you use and make for it. ie shit indexer will give you shit accuracy

 

[laggeddag]

your over thinking everything.

In the past ive used a Hardinge manual 5C collet closer for a 4 axis tombstone, still have it in case my full forth takes a crap or is tied up on another machine. I use a tail stock to support it like David, or I have a a-frame thats alot more rigid with a bushing.
if you want to make fixtures so you can take one out load parts while other is running, or what have you. just get a over sized collet like a 2"dia or so. bore it out to 1.5" and and drill and ream a 1/4" dowell pin in the collet face. then cut your tombstone fixture to 1.5 round and pop a 1/4 hole slip fit for the pin. (use 2 pins for better alignment and repeatability)

for the support. bearings are not necc due to the low speed it will turn, a bronze bushing is more than enough or a tail stock.unless you press the the bearings on to the fixture(light press) are going to give you more accuracy issues than a bushing. (unless you buy good ones)

as far as accuracy, its only as good as the equipment you use and make for it. ie shit indexer will give you shit accuracy

I'm not trying to make it more complicated believe me, just trying to figure out what problems I will run into.

The idea of using a 5C collet 'something' is appealing because the accuracy should be there, and it would also let me clamp a test bar end-to-end to align the fixture before replacing the test bar with the trunnion/tomsbstone (and axle "stubs").

I do have a 10 arc second (or 30 arc second?) 6" rotary table that is worm drive. It doesn't have a clamp though. In a perfect world I would use it as the indexer and support on one side, and the other I'd have a tailstock with a center. The less work I have to do here making this the better! Thing is i'd have to deal with the table clamping issue, maybe come up with something simple that will lock it so it cant bounce around its 10 arc second backlash during a cut.

I am not trying to have a project either. If I could buy affordable off-the-shelf stuff that will let me get there then so be it. I still have alot of work to do making a precision fixture to hold several parts at once.

If a bushing is something I can put together and will work thats great. I was thinking bearings because it would make it easier. I wouldnt think pressing them in would be a problem. Basically just making a simple spindle with two back to back angulars right?

Maybe I'll set up my rotary on the mill bed and see what this is starting to look like
.

 

[dieselpilot]

I would guess you want a super spacer. I was looking into a similar idea. I have limited space and also considered the 5C indexers like Hardinge H-4.