Help with warping material

[snaggletto]

Hello,
I'm trying to make a gang type fixture for machining a slot on the end of several (10)1/2" diameter steel shafts. My idea was to take two 1/2" x 3" x 13+ long steel flat bar pieces and bolt and dowel them together. Then drill ten 1/2" holes along the 3" width so that each plate has half of each hole, or a 1/4" radius. Then mill about ~.010" off of one side so that when the plates are bolted together with the 1/2" shafts they are captured and clamped in a 'sandwich'. This assembly will then be doweled/mounted on the mill table so that 10 of these shafts can be slotted with a slitter saw using the power feed etc... A second duplicate 'sandwich' fixture will be made so that while the machine is cutting one set of shafts, I can be unloading/loading another set of shafts. Obviously, these are production parts.

The trouble is that when you remove all that material off of one side of the plates, they warp up like bananas. I first tried to drill the holes by bolting each plate together and drilling ten 3" deep holes, no luck, both plates warped badly when unbolted. Then I tried using a .5" ball end mill to machine the 1/4" radius in each plate while held flat in the vice, still warped up. I'm stumped, I've been using coolant and low RPM's to keep the material from getting hot while machining, but no luck.

Other than redesigning the whole fixture and using 1"+ thick plate rather than .5" plate, I don't know what to do. Any tricks/ideas on how to keep this material from warping? Thanks for any tips.

[murphy625]

Sounds like heat to me...

Try keeping the whole thing much cooler.

What about just using a thick piece of steel and drilling the holes all the way? Then use set-screws to keep the shafts still. The set screws will also give you some measure of adjustment. (If my picture of what your doing is correct in my own mind).

It could also be the type of steel your using. But I could not tell you what would cause or fix it with the steel grade. (I'm not much up on metallurgy)

I know this, with some steel (cold roll), it doesnt take but a little bit of heat on just one side to cause things to become warped. (the key word being "one side" because if you heat up the entire piece evenly, they wont warp unless they cool down un-evenly).

Even just drilling holes can cause a piece of steel to warp if there is enough heat localized within the piece, or if that heat is allowed to travel naturally (over time) to the top side of the part during any lengthy drilling process.

Working with what you have, I suggest you try this:
1. Dont attempt to drill the hole all at once. Drill just a 1/4 inch and back off to let it cool.
2. Monitor the heat in the immediate area around the tool. Keep it below 120 deg F. (I dont think 120 will warp anything).
3. Take your time and let the piece cool down completely between each bit of drilling. Dont try to drill a hole all at once.

I could still be wrong. Im just a newbie..

Hope this helps.
Murphy

[Ken_Shea]

Here are a couple of thoughts,
1) Stress relieving them prior to milling,
2) Use a single 1" piece drilled and reamed .501 or .502 and set screws to hold the .5 shaft

[psychomill]

For what your fixture is for, I'd use aluminum plates. Take (2) 1." plates and build this fixture. First, face one side of the plate in a relaxed state like light vise clamping or mitee bites. Take a skin cut to clean up the surface. Then flip it over and face the other side. Now take a ball mill and cut the .250 rad in but only go .235 to .240 deep. Drill and tap your 1/2-13s and maybe a couple dowel pins to locate on your table. This should come out relatively flat, definately not a banana and cheap, quick and easy.

The job your running is light machining. Aluminum plates will be plenty stable and hold up to what your doing for quite some time. The beauty of it is, if you need to remake the plates, all you do is face the groove side a little and take the same amount with a ball mill. You'll be back up in minutes. The steel plates sound like a good idea, but its overkill.

HTH

I've done the same thing with vises. I mount 2 vises with aluminum soft jaws, one on each end of the table. Then I drill/ream holes in the soft jaws for the shafts with a .03 shim clamped in between the jaws. Using the power feed I can run one set of parts while loading the other set in the other vise. Works great and its a quick set up. Keep in mind with this set up, if you run more than two parts per vise, the parts need to be real close in diameter. I've done this with 8 parts per vise but all of the shafts were within .001 of each other. If your shafts vary by a few, you'll have to sort them otherwise you might lose from spinning or vibrating. Or just go with the aluminum plate method with screw in between the parts.

[snaggletto]

Thanks for the responses.

I think the 1" solid material with thru holes will be the way to go. Just have to order material.

The actual shafts are 12" lng, 1/2" 4140. The material is not turned true, is a cold finish and mill tolerance on the 1/2" diameter. They will be getting a .0625" slot, ~1" deep on one end. Again, I want to be doing 10 shafts at a time.

Other than installing a 'set screw' or equivalent in each of the 10 locations, anybody have any ideas on a quicker, but still secure locking mechanism for parts that won't be dead nuts on the diameter?

Thanks for the feedback.

[lerman]

Use toggle clamps instead of set screws (think vise grips).

I suspect that your setup time for ten will be more than ten times the setup time for one.

I'd setup a pair of vee blocks with a toggle clamp to hold the rod in the vee and a stop to set the position. Run it past a slitting saw blade, run it back, remove the part, insert the next one, and do it again.

Sometimes the simple solution is the best. (So, what am I missing?) :-)

Ken

[HuFlungDung]

I'd try to avoid building a fixture where you might have the odd "loose one" in the middle of the bunch.

A stack of rounds will self stack in a "tirangular pile" with one clamp across the top bar on two equal stacks. All you would have to provide is a "base fixture" which would be a simple channel with the proper internal width to hold a certain number of rounds snugly. Then lay another row on top of the base row, and so on, until you build a pyramid of rounds which you can then apply a simple bar clamp on the top one.

Create two fixtures to create two stacks and clamp with one clamp across both piles.

If you take the time to measure the centerline height of each row in the stack, you can cut many more pieces in one clamping, just by changing the saw height for each cut.

[snaggletto]

Thanks for the ideas.

I ended up going with the 1" thick plate, and drilled 10 holes through it. There are 2 set screws on each hole to hold the parts in place. It will be a pain to load/unload the parts, but I did make 2 of them. So, while the machine is cutting one set of parts, the operator can be changing the next set of parts. Thanks.

[Mcgyver]

I know found a solution, but for prosperity I thought the underlying problem should be exposed! The issue is not heat but the inherent stresses in the metal. Cold rolled steel is the worst. The stresses come from the rolling process and are mainly concentrated the outer 20-50 thou layer. When you remove the outer layer from one side, the forces are not equal anymore and boing! (technical term)

The warping is more pronounced when you take a cut on a long spindly piece but is present in any piece of cold rolled and will effect dimensional stability. This is one of the reasons why machines and tooling are made from cast iron. The warping is there, you just may not notice it on smaller pieces

When you are doing work that real matters, you must either 1) stress relieve the piece (I send it out because I don’t have an oven) as Ken Shea said or 2) remove the outer layer of metal, or 3) design changes as you have done.

With 2), basically you have to remove the outer layer then take the steps to square and dimension the piece. Be careful with that you are not perpetuating the problem by clamping – the vice straightens the banana while a second cut is made, and of course the banana is back when the vice is released Even a magnetic chuck on the surface grinder will pull things out of shape.

Also, when you do send it out for stress relieving, you’ll back a piece of black scaly metal looking pretty much like hot rolled, not the nice shiny CR you sent out. Hot rolled is a better choice and has less stresses, but it doesn’t seem to be available is the nice free cutting steels that CR is.

[DareBee]

BINGO for Mcgyver

[Rekd]

I know found a solution, but for prosperity I thought the underlying problem should be exposed! The issue is not heat but the inherent stresses in the metal. Cold rolled steel is the worst. The stresses come from the rolling process and are mainly concentrated the outer 20-50 thou layer. When you remove the outer layer from one side, the forces are not equal anymore and boing! (technical term)

The warping is more pronounced when you take a cut on a long spindly piece but is present in any piece of cold rolled and will effect dimensional stability. This is one of the reasons why machines and tooling are made from cast iron. The warping is there, you just may not notice it on smaller pieces

When you are doing work that real matters, you must either 1) stress relieve the piece (I send it out because I don’t have an oven) as Ken Shea said or 2) remove the outer layer of metal, or 3) design changes as you have done.

With 2), basically you have to remove the outer layer then take the steps to square and dimension the piece. Be careful with that you are not perpetuating the problem by clamping – the vice straightens the banana while a second cut is made, and of course the banana is back when the vice is released Even a magnetic chuck on the surface grinder will pull things out of shape.

Also, when you do send it out for stress relieving, you’ll back a piece of black scaly metal looking pretty much like hot rolled, not the nice shiny CR you sent out. Hot rolled is a better choice and has less stresses, but it doesn’t seem to be available is the nice free cutting steels that CR is.

^ nailed it.

[gibbsman]

When material is rolled the outside surface cools quickest causing uneven stresses. When you remove material from one side it causes the other to pull in the opposite direction causing the banana effect . You need to remove the 'scale' from one side then flip the part over and take the material off the other side and repeat this for all sides to make a flat part.