Vise Bending Stock? Pondering Custom Fixturing

sdinzy · Nov 8, 2020

Hi Guys,

I have only used my 5th axis vises thus far in my machining career of 1 year. But I would like to improve accuracy/consistency of my parts.

I currently square my stock (5”x4”x1.5” 316SS) and load it into the top portion of the vise jaws, which is only 0.1” tall, but holds rigid with the vise teeth (part is loaded with 5” side running along to X, 4” side to Y). I never exceed max torque of the vise, 75nM. Mill away, nothing overly aggressive, although I do use a 1” Helical end mill for the heavy roughing so it does need some clamping force, leaving about 0.150” plate remaining on bottom of stock, create some locating bores through the part. After I flip the part over, I find that the remaining “plate stock” is far from flat anymore.. all I can assume is that I am bending the stock when I clamp it in the jaws? Is that even possible with a 6” vise and stock of this size?

I am considering creating some custom fixturing with Mitee Bite pitbull clamps on all 4 sides of the squared stock so that there is no bending or lift.

Any wisdom to what’s wrong with the current setup or what would not work in the proposed custom setup would be appreciated!

TheBigLebowski · Nov 8, 2020

sdinzy said:
Hi Guys,

I have only used my 5th axis vises thus far in my machining career of 1 year. But I would like to improve accuracy/consistency of my parts.

I currently square my stock (5”x4”x1.5” 316SS) and load it into the top portion of the vise jaws, which is only 0.1” tall, but holds rigid with the vise teeth (part is loaded with 5” side running along to X, 4” side to Y). I never exceed max torque of the vise, 75nM. Mill away, nothing overly aggressive, although I do use a 1” Helical end mill for the heavy roughing so it does need some clamping force, leaving about 0.150” plate remaining on bottom of stock, create some locating bores through the part. After I flip the part over, I find that the remaining “plate stock” is far from flat anymore.. all I can assume is that I am bending the stock when I clamp it in the jaws? Is that even possible with a 6” vise and stock of this size?

I am considering creating some custom fixturing with Mitee Bite pitbull clamps on all 4 sides of the squared stock so that there is no bending or lift.

Any wisdom to what’s wrong with the current setup or what would not work in the proposed custom setup would be appreciated!

How much stock are you removing from the sides? Is the part just a rectangle or is it complex? The thickness remaining on the bottom is not all that important, what matters is the ratio of the remaining flange thickness vs how far it sticks out.

Given that 316 is far from stable, I would hazard a guess that anything more then 1/4 inch removal in a given area (at 0.15 remaining thickness) is going to result in some major stress relief and part movement.

Edit:

If you want just to check how much the vise bends the part, You can put the part in the vise without clamping, indicate it flat, then clamp it and check for bowing. There will be some, but it may be negligible.

sdinzy · Nov 8, 2020

TheBigLebowski said:
How much stock are you removing from the sides? Is the part just a rectangle or is it complex? The thickness remaining on the bottom is not all that important, what matters is the ratio of the remaining flange thickness vs how far it sticks out.

Given that 316 is far from stable, I would hazard a guess that anything more then 1/4 inch removal in a given area (at 0.15 remaining thickness) is going to result in some major stress relief and part movement.

Edit:

If you want just to check how much the vise bends the part, You can put the part in the vise without clamping, indicate it flat, then clamp it and check for bowing. There will be some, but it may be negligible.

Hi El Duderino,

Part is very complex, and sticks out nearly the full stock height from the remaining flange (~1.25"). 1st roughing pass with 1" EM profiles the part all the way down to the remaining flange, so a full 1.25" DOC. I haven't noticed any heat issues.. I'm running everything pretty conservatively, coolant, chips are a nice light champagne color, and nothing that would lead me to believe there is part movement. Not sure what else could cause deformation of the bottom side of the flange, but again, I'm fairly new to all this.

But yeah you are right, I should indicate the stock after clamping and see where it's at.

Ianagos · Nov 8, 2020

Different material but I’ve had similar problems with thin cold rolled material.

I was clamping a 4”x4” 1/2 flat bar and had to put an indicator on top as I tightened the vise. The vise would bend the part and when I loosened it would bow. But I also came to find that while the vise was part of the problem the rest of it was caused by the unstable nature of cold rolled material. After a stress relieve in the oven and careful clamping a part that went from .003 or so of bow was under .001 and in tolerance.

I was using a similar method as you but only leaving .1 on the floor of the first operation.

macds · Nov 9, 2020

If you do a lot of similar work, and the volume justifies it, the dimpling machines and matching jaws are the cats arse.

Regarding torquing the vise, put your finger in, and turn the clamp by clamp. Youll see how hard those suckers bite.... I grimace when i see guys putting all 200lbs of their weight on my Gerardi vises (the first and only time) thinking they are holding the shit out of the part.

sdinzy · Nov 9, 2020

macds said:
If you do a lot of similar work, and the volume justifies it, the dimpling machines and matching jaws are the cats arse.

Regarding torquing the vise, put your finger in, and turn the clamp by clamp. Youll see how hard those suckers bite.... I grimace when i see guys putting all 200lbs of their weight on my Gerardi vises (the first and only time) thinking they are holding the shit out of the part.

I totally get it. I’ve had worpieces slip out with that 1” end mill though, $500 gone.

I’m going to try annealing some stock, squaring, putting it in the vice so that it’s not bent/bowed from the get-go, and see how that plays out. May not get to that for a week or so, but I will report back. Any other comments/suggestions in the meantime are welcomed!

TheBigLebowski · Nov 10, 2020

You could also try milling a part without ever clamping it down. For testing purposes, you could bolt the part to a plate and mill it using the same program (leaving the same 0.150" on the bottom)

I would wager that when you flip it for the second op, warping would be near identical to the parts done in the vise. That would tell you that stress release from machining is the major issue, not clamping pressure from the vise.

cb750chris · Nov 10, 2020

Most material can warp considerably after you remove a lot of meat from a blank, as was said above, regardless of how its clamped. We use a lang press at work, so a lot less clamping torque is necessary, but a lot of the time we clamp firmly for roughing, then loosen the vise (to the point of just keeping the part from moving) then reclamp at a much lower torque. Then run finishing passes. It takes a lot of the stress off the part before finishing passes. Also, I wouldn't expect to have a finished surface on the bottom face after hogging out a bunch of material on the top. Maybe leave a little extra stock to be faced off after the flip, then it won't matter. You could also try using a smaller endmill with HEM or high feed milling to reduce radial load so you won't have to clamp as hard. Plus 1/2" endmills are a LOT cheaper than 1" ones.

Sent from my SM-N960U using Tapatalk

g-coder05 · Nov 11, 2020

Part is very complex, and sticks out nearly the full stock height from the remaining flange (~1.25"). 1st roughing pass with 1" EM profiles the part all the way down to the remaining flange, so a full 1.25" DOC. I haven't noticed any heat issues.. I'm running everything pretty conservatively, coolant, chips are a nice light champagne color, and nothing that would lead me to believe there is part movement. Not sure what else could cause deformation of the bottom side of the flange, but again, I'm fairly new to all this.

316 just sucks. It is terribly unstable and just pulling the scale off makes the stuff conform to a pringles chip. I try to avoid bidding jobs with 3XX stainless but if I have to, I tack on a headache fee. In your case I would

OP 10: Face mill the scale
OP 20: Flip and Face mill the scale
OP 30: Flip, semi finish features and semi finish profile
OP 40: Flip, Finish all features
OP 50: Flip back to OP 30 and finish features

You gotta get the stress out or you may as well be trying to make a ski jump. Even with your PitBull setup the stuff needs somewhere to relax, I don't think that will solve your problem either.

sdinzy · Nov 11, 2020

sdinzy said:
I totally get it. I’ve had worpieces slip out with that 1” end mill though, $500 gone.

I’m going to try annealing some stock, squaring, putting it in the vice so that it’s not bent/bowed from the get-go, and see how that plays out. May not get to that for a week or so, but I will report back. Any other comments/suggestions in the meantime are welcomed!

I’m actually happy to hear this, as I’d rather just stress relieve the stock than figure out a whole new fixturing solution. I’ve got some foundry tongs arriving today so that I can try stress relieving the 316 properly and try it out. To be perfectly honest, I didn’t even know what stress relieving was before this thread. Sounds like maybe I should read the Machinist’s Handbook or something.

sdinzy · Nov 11, 2020

cb750chris said:
Most material can warp considerably after you remove a lot of meat from a blank, as was said above, regardless of how its clamped. We use a lang press at work, so a lot less clamping torque is necessary, but a lot of the time we clamp firmly for roughing, then loosen the vise (to the point of just keeping the part from moving) then reclamp at a much lower torque. Then run finishing passes. It takes a lot of the stress off the part before finishing passes. Also, I wouldn't expect to have a finished surface on the bottom face after hogging out a bunch of material on the top. Maybe leave a little extra stock to be faced off after the flip, then it won't matter. You could also try using a smaller endmill with HEM or high feed milling to reduce radial load so you won't have to clamp as hard. Plus 1/2" endmills are a LOT cheaper than 1" ones.

Sent from my SM-N960U using Tapatalk

Hi Chris,

Thanks for the info, I am going to stress relieve and see how it goes. FYI I am using HEM, it’s just such a big tool that the cutting forces are pretty strong, although I’m running it somewhat conservatively (125% ADOC, 5% RDOC). Sounds like once this is annealed, it might be possible to machine this a little more aggressively.

sdinzy · Nov 14, 2020

UPDATE: Ok, so I annealed a few blocks of 316; heated to 1950 degF, maintained for 2 hours (2" thick stock), water quenched. Machined my first op, measured the remaining flange. There is about 0.020" lift from one end of the plate to the other, over a length of 4", and bending/bowing is between the two edges that are held by the vice; the other direction was flat. I am not disregarding the stress relieving, sounds like I should have been doing that all along, but there is still an issue here.

Looking at the vice specs, the clamping force is 20,000N (4500 lbs) at 100Nm of torque; I tightened the vice to 50nM in this test, so assuming linear clamping force, there was 2250 lbs clamping force. Is that enough to bend a 4.75" x 3.875" x 2" bar of annealed 316? The top part of the flange was perfectly flat, so it seems any bending would have come from initial clamping. Let me know your guys' thoughts!

50BMG DUDE · Nov 14, 2020

sdinzy said:
UPDATE: Ok, so I annealed a few blocks of 316; heated to 1950 degF, maintained for 2 hours (2" thick stock), water quenched.

I think the water quench would be the problem right there. I'm not a metalurgist, but my experience is anneal or Stress relieve is usually an sill air cool process. I've even heard of guys banking under vermiculite or usually just letting cool down in the furnace. Also stress relieve is usually quite a bit lower temp than annealing.

Also you may want to soak the part at temp longer as it sounds like a fair chunk of metal. I think the rule is at least 1 hour for 1" of cross section.

I'd give it another shot.

YMMV

sdinzy · Nov 14, 2020

50BMG DUDE said:
I think the water quench would be the problem right there. I'm not a metalurgist, but my experience is anneal or Stress relieve is usually an sill air cool process. I've even heard of guys banking under vermiculite or usually just letting cool down in the furnace. Also stress relieve is usually quite a bit lower temp than annealing.

Also you may want to soak the part at temp longer as it sounds like a fair chunk of metal. I think the rule is at least 1 hour for 1" of cross section.

I'd give it another shot.

YMMV

I did quite a bit of research on annealing 316, sounds like water quenching is the universal way. I am definitely ok with trying annealing for a longer period. On another note, I was testing to see if I could bend a piece of stock during the squaring process, and this is what happened:
*Did first op squaring; removed 1/4" from the top with a 3/4" end mill, final facing, contoured sides down to just above the jaws.
*Measured everything, flat as a pancake (pancakes aren't really flat though?)
*Flipped it over, tightened the vise 5nM at a time up to 45nM, no permanent bending (after each tightening, removed stock and measured the faced side)
*Put the stock back in the vise at 45nM, did the final squaring op, which consists of removing a little stock (about 1/4") off the top with a 3/4" end mill, then a final facing.
*Removed the stock from the vise, measured, it's square/flat

I'm going to run the first op on my part, and if it comes out warped, then the annealing may not have been done long enough as you suggested. From what I understand, annealing is supposed to remove all residual stresses, whereas stress relieving at a lower temperature removes most (85%) of residual stresses.

Ianagos · Nov 14, 2020

I think the water quench was not the best way but acceptable in 316 whe I stress relieve I let the parts cool in the oven.

I know you are doing some heavy roughing on the part and I think the issue is not that it bends on the initial clamping but once you rough out the majority of the material there is very little left to support the part and it causes flex.

You may try taking a op1 finished part and loosening it then put an indicator on the top and tighten till you get .001” of lift. Then at that point use your torque wrench to determine about how much torque that amount of bow takes. Its usually surprisingly little force. Trying something like morrer bite talon grips in your jaws can hold your parts much better with less force from the vise.

I don’t see the need for a big ole 1” cutter either.

gkoenig · Nov 15, 2020

Being a 5th Axis vise, one of the nice things is that the gripping teeth also serve as a dovetail clamping surface. One method might be to do an Op 0 prep that faces the scale off the bottom and adds a set of dovetails along with a work-stop for easy loading. You can now bring the jaws closer together and remove the bending stress from the edges of the part.

It is the one thing I'll give 5th Axis over Lang - Lang wants you to use the stamping dingus and never dovetail a part. You can totally get away with it, but the gripper teeth have some funky angle to them that makes this a non-optimal and off-label use. The stamper is nice and all, but if you do mostly aluminum, it isn't really necessary and the dovetail trick works in those instances where it doesn't.

sdinzy · Nov 15, 2020

gkoenig said:
Being a 5th Axis vise, one of the nice things is that the gripping teeth also serve as a dovetail clamping surface. One method might be to do an Op 0 prep that faces the scale off the bottom and adds a set of dovetails along with a work-stop for easy loading. You can now bring the jaws closer together and remove the bending stress from the edges of the part.

It is the one thing I'll give 5th Axis over Lang - Lang wants you to use the stamping dingus and never dovetail a part. You can totally get away with it, but the gripper teeth have some funky angle to them that makes this a non-optimal and off-label use. The stamper is nice and all, but if you do mostly aluminum, it isn't really necessary and the dovetail trick works in those instances where it doesn't.

That is great thinking, I forgot the jaws had the dovetail slots in them.

SendItCNC · Nov 15, 2020

gkoenig said:
Being a 5th Axis vise, one of the nice things is that the gripping teeth also serve as a dovetail clamping surface. One method might be to do an Op 0 prep that faces the scale off the bottom and adds a set of dovetails along with a work-stop for easy loading. You can now bring the jaws closer together and remove the bending stress from the edges of the part.

It is the one thing I'll give 5th Axis over Lang - Lang wants you to use the stamping dingus and never dovetail a part. You can totally get away with it, but the gripper teeth have some funky angle to them that makes this a non-optimal and off-label use. The stamper is nice and all, but if you do mostly aluminum, it isn't really necessary and the dovetail trick works in those instances where it doesn't.

With the 5th axis vise you might want to check for jaw lift after tightening when using the dovetails at the top of the jaws. Still, the setup may still be plenty firm in spite of any jaw lift.

sdinzy · Nov 16, 2020

50BMG DUDE said:
I think the water quench would be the problem right there. I'm not a metalurgist, but my experience is anneal or Stress relieve is usually an sill air cool process. I've even heard of guys banking under vermiculite or usually just letting cool down in the furnace. Also stress relieve is usually quite a bit lower temp than annealing.

Also you may want to soak the part at temp longer as it sounds like a fair chunk of metal. I think the rule is at least 1 hour for 1" of cross section.

I'd give it another shot.

YMMV

Upon closer inspection, looks like the warping is less uniform than I thought (after a 2-hour anneal), so I think you’re right that it needs a longer soak. FYI, the reason for water quenching is to avoid sensitization in the stainless; sounds like a must-do. I’ll follow up after machining a part annealed for a longer period of time.

sdinzy · Nov 16, 2020

Ianagos said:
I think the water quench was not the best way but acceptable in 316 whe I stress relieve I let the parts cool in the oven.

I know you are doing some heavy roughing on the part and I think the issue is not that it bends on the initial clamping but once you rough out the majority of the material there is very little left to support the part and it causes flex.

You may try taking a op1 finished part and loosening it then put an indicator on the top and tighten till you get .001” of lift. Then at that point use your torque wrench to determine about how much torque that amount of bow takes. Its usually surprisingly little force. Trying something like morrer bite talon grips in your jaws can hold your parts much better with less force from the vise.

I don’t see the need for a big ole 1” cutter either.

If you stress relieve, you can let it air cool. But if you anneal, you have to water quench to avoid sensitization through the 950-1500degF range.

FYI, I changed up the squaring/ops a bit to leave a thicker remaining flange, about 0.5” thick, per your reasoning. Also clamped much lighter without issue. Still getting distortion, so I don’t think it was annealed long enough. To be continued...

Vise Bending Stock? Pondering Custom Fixturing

Aluminum

Aluminum

Aluminum

Stainless

Hot Rolled

Aluminum

Aluminum

Aluminum

Titanium

Aluminum

Aluminum

Aluminum

Cast Iron

Aluminum

Stainless

Titanium

Aluminum

Plastic

Aluminum

Aluminum

Similar threads