Need Help!- Workholding advice

[StalkDaddy]

Hello, I am new here but I have a good amount of CNC experience. I am a U.S. Marine Machinist stationed overseas in Japan. I have been out of the CNC machining industry for the last 3 years because the location I am at has only had manual machines until recently. I convinced the command here to purchase a Haas TL1 and TM1. Anyways, I am attempting to fabricate a sort of anchoring plate. The finished plate size needs to be .250" thick + or - .020". I am cutting the plate out of .500" 304 stainless steel. The plate machines fine, but when I flip the plate over to face the excess material off the back side, I am left with a .020" taper from end to end. To hold the workpiece for facing off the back side, I cut the profile of the part into some aluminum vise jaws I made so that I have a good hold. Some pictures are below that show the way I am doing it now. I assume that my taper is coming from the way I am holding the workpiece. I can't think of anything else that would give me a taper. Any advice would be greatly appreciated. (The finished plate size is just under 14 inches long.)



Above is a picture of the finished product. (it's two pieces, I am only concerned about the base plate)



Above is a picture of my vise jaws. Its hare to see because of the flash, but the jaws contour to the arcs of the workpiece.

[Caprirs]

When I do something that thin, I will rough off most of the material and leave .002"-.030" to finish. Use an M0 to pause the program, loosen the vise, use a torque wrench to tighten the part again and use a rubber mallet or dead blow hammer to seat the part into the soft jaws. Then finish the part. This allows for adequate clamping force for roughing without pulling the part out of the vise and light clamping to reduce bowing for finishing.

[Delw]

+1 for what cap says

Were you at in Japan? (marine base by mt fuji?) tonzowa?(sp)

I was born and raised there till 15, dad was Navy 24 years we were in Yokuska, Misawa yokohama and chi chi jima. Mom Just left tonight for guam , she hasnt been there since 1961

Thanks for your service.

Delw

[StalkDaddy]

I am currently stationed on Marine Corps Air Station Iwakuni.

[tarkus01]

my guess is that you have a little bit of play in the moving vise jaw. when you tighten the jaw it will lift slightly. causing the part to be tapered.
I would have held it on the boss with a small round of soft steel. see picture (pls excuse simply picture, no CAD @ home). then as long as the fixed jaw is true the machined top should be true.

Robert
Melbourne Australia

[DMF_TomB]

some times i use a fixture plate to hold parts down to it.
.
sometimes holes are bored and then hold down clamps are put on.
.
my most used fixture plate has 1/4-20 tapped holes every inch so hold down clamps can be put almost anywhere.
.
as attached picture shows i often use kant twist clamps to directly clamp plate to fixture. fixture is shown in a kurt vise so it is high enough off table to give clearance for clamps.
.
some glue plate down with loctite 430 metal bonding adhesive to another plate. this lessens curling from vise clamping pressure and metal deflecting away from cutter that is not fully supported.
.
some use thin 1/32" double sided foam. i have seen this tape work so well it can be difficult to remove metal from tape.
.
if you use glue, adhesive or tape sometimes heating part will weaken bond. in the old days they used melted shellac or certain waxes as a hot melt glue and then put part in alcohol to break glue bond.
.
by the way normal hot melt glue does not do well gluing aluminum to steel. i have seen parts that appeared to be strongly attached and when the slightly warm part was put under cold water the parts fell apart. i believe a glue with no give will break free when part expands bigger when it is hotter.

[christinandavid]

Face 1st side.

Turn plate over, clamp around edges. Face to size and finish holes.

Clamp through holes and machine profile.

As previously mentioned, this method would require a sacrificial plate that you can mill/drill into.

Plus there is little chance it would end up flat - but you do not mention a flatness tolerance.

DP

[dertsap]

Originally Posted by tarkus01 my guess is that you have a little bit of play in the moving vise jaw. when you tighten the jaw it will lift slightly. causing the part to be tapered.

that was my initial thought , if thats a kurt vise then that problem can easily be solved by snugging up the set screw on the sliding jaw

[handlewanker]

Hi, going down from .500" to .250" means you're going to be releasing a lot of rolling stress from plate manufacture.

I would have made first side cut at about .030" depth just to get the skin off, then turn over and cut second side, again .030" deep to get the skin off, then turn over and cut half what's left, leaving .010" to clean up, then turn over and finish the other side, finally turning over to finish to size.

This does sound a bit "long winded", but when you are handling difficult material, (stainless) and plate work even in soft hot rolled mild steel, becomes difficult when you machine only one side of it, then sometimes to get the material to behave you have to "go with the flow" or work with the problem not forcing it.

I would also have a look at the "fit" of the part in the soft jaws, as even a couple of thou' on the contour can make all the difference as to if'n it's being held or just at one end.

When you work with rolled plate you have lots of stress built in and "making bananas" is the term we always used for people who thought they could machine one side only and get perfectly flat results.
Ian.

[DMF_TomB]

part curling is worse the longer the part and if metal was cold rolled. Hot rolled and or stress relieved or annealed metal is more stable.
.......yes you should machine even amounts from both side or figure on a roughing cut, unclamping and deliberately clamping the metal again but NOT trying to flatten it out in a vise or clamping flat to the table. This may require using shims if clamping to a table. I have seen many a machinist clamp a curled metal piece flat take another cut and when unclamped the curl got worse. Pretend you want to hold a banana and want to take cuts to straighten it out.
........Basically clamp the rough machined part with what ever curl it naturally has when unclamped and taking finishing cuts to take the curl or warpage out.
........long metal parts can easily curl, warp and or twist many mm. so make sure you leave enough stock to take finishing cuts.

[handlewanker]

Hi, I quite agree with NOT reclamping a piece of flat plate stock after the first cut when it warps, and then reclamping DOWN to flatten it prior to subsequent machining.

All you achieve is to lock up the spring in the metal by bending against the natural pull from the stress.

Many years ago, about 40 as far as I can remember, I had to machine a piece of cold rolled 50mm X50mm (2" X 2") square steel bar stock, about 250mm (10") long, on a Cincinnatti vertical mill, with the part held in a vice.

The drawing called for two cut outs in one side about 50mm wide and 35mm deep spaced about 125mm (5") apart on centres.

Everything went to plan, rough out the centre of each cut and then take a skim to clean up.....I wish.

When the vice was opened the work piece jumped up like a hump backed bridge.

The foreman said, "No worries, just do as we always do and stick it under the press to flatten it and remachine it to final size"..... ooooooops, it was, past tense, already on size, so I learned a valuable lesson about working with cold rolled steel to flatten and remachine to final size.

In the case of a piece of flat plate, I'd cut the thing both sides, and before final finishing either heat it up or flatten it to as flat as possible and skim both side again, and if'n it still takes on a bit of a bow, go outside to the boy's room and have quiet cry.

I've seen some people who were faced with the flat plate stuff warping all the time, placing shims under the middle of the bow and machining the bowed face flat testing as you go, it's going to move no matter how you hold the stuff.

A preheat to get the material to relax is probably the best solution, but shimming the centre and remachining, although a lot of fiddling about generally, does get most of the warp out eventually, but don't wack it down before cutting.

If'n I was designing this job I'd go to material .250" thick and lightly grind not mor'n about .03mm (.005") off one side that mattered and was seen, or maybe both sides if the customer wanted shiny all over, that way the stress would still be evenly tensioned in the heart of the plate.

That centre boss is a worry, as soon as you cut out or weld in the boss the plates going to go bananas...LOL.

Incidently, the most important point is what is the job worth....that dictates how much time you spend mucking about, and some jobs don't even pay, and the only profit is what you get for scrap value from the swarf....LOL.
Ian.

[DMF_TomB]

straight parts from curved ones is time consuming.

1) stress relieving or annealing parts helps

2) think of it , the curled or warped part as a piece of wood or lumber. it can be cupped across width, curled in length and can be twisted. but you start thicker than desired thickness and take a little off on each side trying not to effect it's natural straightness or lack of by using so much force as to flatten it against something flat. using shims to clamp it down to a table usually requires a dial indicator and magnetic base. with the proper shims as you clamp it down to a milling machine table the dial indicator should not change readings more than 0.001" / otherwise use 0.001" more in shims.

3) the tighter the tolerance the more the need for stress relieving and the need for more roughing and finishing cuts flipping and checking for straightness or lack of.

4) still think metal is easier than plastics. i have had nylon under so much stress from it drying out (it absorbs water) that i have had 6" round nylon crack or split like a wood log every time i drilled a hole down the center. i tried multiple time using more different size drill bits but always from this same piece of nylon when it had a hole down the center big enough it made a pop noise and split like a wood log every time. i have had nylon parts with a arbor pressed in bearing extremely tight but sit in a drawer for a year drying out and the metal bearing would fall out because nylon bore got bigger and od got smaller. at least metal does not change 0.010" per inch over some water.