Can this part be saved? Internal stress

[matthew-s]

I could use some wisdom and experience here. I’m rather internally stressed about this internal stress.

This is the radius rod from a 3/4 scale Kozo A3 (Kim’s build for reference).  There are other similar parts I’ve machined, and this did not happen …

The material is annealed 303 stainless.

You can see after cutting the part from the block with a slitting saw, it bowed by 0.020 over its span.

Can it be saved?
My current thoughts are:
1. Attempt to bend it back … somehow. Clearly I’d have to support it to keep it from twisting.
2. See what happens when I machine the taper, try to favor one side or another? Technically only the dimensions between the three to-be-completed holes matter. That may not work, and/or look like junk.
3. I guess I could anneal it, but I don’t think that would return it to a different dimension, it might just freeze it as is? OTOH, maybe that would make option (1) easier / more feasible?

For the next part(s) …
There is still the rough second rod that has not been sliced from the block yet. Should try to anneal this in the kitchen oven ( )  before using the slitting saw to separate it from the block? Any other advice?

Thanks in advance!

[uuu]

Although it seems to be a sledgehammer to crack a nut, I have found the fly press to be a very effective tool for straightening things.  Guided by my friend John, who definitely has a knack for this, it's possible to use a vee-block and blade to tap the part back into shape.

You can build up a feel just bouncing the tool along the part, within the elastic limit, so it stays as it is.  Then little by little you increase the bounce until you just, and only just, take it past the limit, and the magic happens.  It's amazing how sensitive it is.

Wilf

PS - but I've not done it on stainless, where there's clearly a risk of hardening.

[crueby]

If it had been milled on one side from a flat bar, then milling the same amount from the other would have it straighten out again - if done in a couple shallower cuts, alternating sides. But, if I read it right, this was cut off the side of a wider block. In that case, machining the side away from the slitting saw cut will likely not have it change the bow. But, if the part is thick enough to trim some more, you could mill one side flat, then take the other side down parallel to that. If its not thick enough to do that, then bending it could work, 303 will take small bends without cracking. Support the ends, then press down on the center, past straight and see where it returns to. If you dont have a press, a wide vise would work too, tape blocks at either end of one jaw and one in the middle of the other jaw to get the supports needed. Worst case is to make a new part from flat bar, and mill alternately from either side to get it down to thickness without warping.

[matthew-s]

Thanks to the input so far. People are converging on bending it.

As an aside. I did mill equal amount off either side of the block to get it from .250 to .200 thick.

The problem here is the part itself is not symmetrical. The long thin part is offset from the centerline, aligned with the “back” of the part.

Anyone ever try to anneal 303 at home? What happened?

[crueby]

I'm not sure  I understand  how it was cut from the bar. The offset on one 3nd would warp from stresses to the side towards the fork, but it seems to have warped in the other axis?

[uuu]

I look forward to how this all turns out, as I have some parts to cut out of a 48" long rectangular bar of 303.  I know the stuff machines OK, as I've used some of it already, but not long parts yet.

Wilf

[Jasonb]

Won't hurt to try and bend it. Mill vice would probably do, just use feeler gauge to pack the "high" point so you can bend it beyond straight as there will be some spring back.

It's times like this when Hot Rolled mild steel has it's advantages over shiny stainless and why stainless is not usually used for much on a steam engine except piston and valve rods.


Chris. I imaging it was cut from wider bar held on edge in the vice. The profile cut and then a slitting saw used to cut it from the stock that is still in the vice. so basically the top edge of the piece has expanded once cut from the stock and caused the bend. In other words the cut is unbalanced vertically not horizontally

[matthew-s]

Yes. It’s generally as Jasonb drew it. It so happens that Kim approached this part differently than Kozo suggests, so my link to his post only really shows what the finished part is like.

I’m going to try the “vise as a press” method. Can’t make it worse!  I shall report back here with the results.

[Charles Lamont]

Not having any kind of press in the shop, I find light jobs can be accomplished with a fair amount of control using the pillar drill. The spindle bearings are good for a fair amount of force, and after all, another name for it is drill press. Things sit on the table while adjustments are made, rather than falling out of the bench vice.

[Kim]

I'd go with the bending route. You don't really need too much movement there to get it back where you want it.

Sorry you're having to deal with this

I would think annealing stainless would work similar to other steel, though I don't know as I haven't really done it before.  I have successfully work hardened stainless though! 

Good luck!  Make sure and let us know how it comes out!
Kim

[matthew-s]

Well, that was far less drama than I anticipated. Bending it worked. I got it within 0.002. I’ll take it, especially for this part.

For the curious, I added a picture from Kozos book, showing how he suggests the part be made.

There was mixed info on the internet about stress relieving 303. And whether that is the same as a full anneal. One source implied stress relief can be done at broiler temperatures. Others made me think that person confused F and C and perhaps I need to approach 2,000F.

I guess we’ll find if a kitchen oven soak helps when I slice the second radius rod off the block!


I did purchase what was described as annealed 303. I was hoping that would prevent this. Oh well. Forward, march!

[Kim]

Glad the bending went well, Matt! That's great! 

So nice not to lose all that work for such a simple thing!

Interestingly, Blondihacks posted her video of making this part just yesterday.  She used Kozo's method: https://www.youtube.com/watch?v=JLyd7DCfT7Q&t=1s

Great work, Matt, and good luck with the next one!
Kim

[crueby]

Stress relieving and annealing are not the same thing. Annealing softens the metal, stress relieving relaxes the rolling stresses from the manufacturing processes. Annealed metal can still have stresses.

[rklopp]

Stress relieving and annealing are not the same thing. Annealing softens the metal, stress relieving relaxes the rolling stresses from the manufacturing processes. Annealed metal can still have stresses.

No, annealed metal should have less residual stress than stress-relieved metal. Annealing is done at a higher temperature and should be cooled slowly enough to avoid thermal stresses.

[crueby]

Stress relieving and annealing are not the same thing. Annealing softens the metal, stress relieving relaxes the rolling stresses from the manufacturing processes. Annealed metal can still have stresses.

No, annealed metal should have less residual stress than stress-relieved metal. Annealing is done at a higher temperature and should be cooled slowly enough to avoid thermal stresses.

Annealing can be done quickly. Stress relieving takes a lot longer. And a lot of the stresses causing warps when cut are from working the bar, not just from thermal stress. Things vary a lot with the alloy too, the recipes for stress relieving and  annealing  vary a lot too, not just a simple heat and cool for all metals. My experience  with annealed metal is that it still has stresses and still warps when cut. The 303 in the part shown in this thread was annealed 303, and it warped.