Stress relieving mild steel weldments

Pete F · Feb 9, 2016

I have some parts I'm trying to decide how to assemble, and am considering doing them as TIG weldments. The material is bog standard hot rolled and will be ground after assembly. Once assembled I'd like it to remain stable as it will need to house mirrors I'll autocollimate off.

I can fire my oven to max 1200 C, so I wonder if these weldments were heat treated after welding, could I expect them to remain stable? If so, what is the suggested temperature?

Thanks

billzweig · Feb 9, 2016

I stabilize steel at about 500C for a few hours and let it cool slowly overnight in the furnace .

Pete F · Feb 9, 2016

Thanks Bill. One thing I'm not sure of, and I guess that's part 2 of the question, is just how stable is "stable" after heat treatment. I've never done it where I really don't want these things to be moving about due to residual internal stress. Will a heat treat cycle(s) really get rid of all the welding stresses, or just how thorough is the process? I ask because the alternatives I have are to bolt the components together, or silver solder the components. My preference is to TIG, but I'm entering uncharted waters for myself in terms of thorough stress relieving post welding.

billzweig · Feb 9, 2016

Pete F said:
Thanks Bill. One thing I'm not sure of, and I guess that's part 2 of the question, is just how stable is "stable" after heat treatment. I've never done it where I really don't want these things to be moving about due to residual internal stress. Will a heat treat cycle(s) really get rid of all the welding stresses, or just how thorough is the process? I ask because the alternatives I have are to bolt the components together, or silver solder the components. My preference is to TIG, but I'm entering uncharted waters for myself in terms of thorough stress relieving post welding.

I work mostly with smaller parts - maybe 4-5 kg max. The results are very stable over the period of checking which can be a few months. For example a 150x150x 12mm plate will stay whin a few of microns after surface grinding. For best results I sometimes repeat the heating cycle the next day.
One thing to keep in mind is that the annealed steel is quite soft.

Street · Feb 9, 2016

Of all the Mirrors i have seen mounted they have been in Cast iron, slightly different to a weldment.
Where stresses are in the plate and due to weld shrinkage.
A cast item it will be due to differential cooling when cast.
Weld distorting items can be quite powerful, as to which is better, i'd go with cast as joins are full thickness as cast, time to make full penetration joins will eat into savings from not casting.

We used to anneal cast by heating it till light red then placing in tub filled with Powdered lime and covering it with a good coverage ( several inches of lime ) pull it out a couple of days later.
A furnace brought up to temp and lowered over days would be ideal solution.

Look up dull red as to what that equates to temp.

As for steel weldment you may have to go higher in temp.

Best bet would be get hold of a old heat treatment book..

Pete F · Feb 9, 2016

Thanks guys, these are going to be steel and I'm not interested in cast iron. If it were a production run I'd make a pattern and have them cast, but mild steel is the way I'll go with this for a number of reasons I won't go in to here. They will sit on 3 hardened feet that will be lapped to bring the mirror in perpendicular.

I'll now go ahead and plan to weld them. They're a very simple inverted T shape with a ring to hold the mirror.

Thanks

Forrest Addy · Feb 9, 2016

Plain vanilla mild steel stress relieves quite well but there will always be residual strains. You say you will use your fabrication for structural elements of fine optics - which is a test for the stability of any material. Once you've completed and tested your instrument, I suggest you calibrate it carefully incorporating test points you generated is process with 20 micro resolution and set up a calibration log..

My suggestion is to start with short calibration intervals - say 30 days - and keep a careful log of how the instrument drifts over time. If the beast seems stable over short intervals increase them.

Mild steel bruises easily and these bruises usually involve raised metal which will have to be carefully remedied if they lie within or near a reference surface. Mild steel instrument frames will not tolerate the abuse cast iron, or HT steel, or hard bronze could take in stride. You will have to handle steel fabricated instrumentation with extra care and include a slip stone in the protective case with the other ancillaries.

Stabilized mild steel will work for very precise instrument construction but it will have to be babied and calibrated accordingly. Stick lots of those rubber bumper things on them in likely places.

snapatap · Feb 9, 2016

My heat treatment book says 400-650C for stress relieving low carbon steel. The higher temp will relive more stress than the lower, as will longer soaking times. It would be a good experiment to do regular calibrations on it and see if or how much it moves over time after it is finished.

Pete F · Feb 9, 2016

It's not the hot rolled steel that will need to be stress relieved, it's the welds. The heat affected zone (HAZ) and the welds shrink tremendously upon cooling, and will set up a lot of stress in even a simple weldment like that. It's why I initially considered alternative methods of assembly.

These are autocollimator mirrors and wear and abuse won't be a factor. They are double sided and will be self-proved for alignment. In fact for one of the designs, perpendicularity isn't essential, as autocollimation measurements are typically relative from one position to the next. Nevertheless I will make the mirror accurately perpendicular in any case. For the other design it will matter (in that case the mirror is parallel to the base, but it can again be self-proved, also there's no welding in that design.

malcolm rypauf · Feb 9, 2016

Pete, just a wild idea: could you assemble these with a high-tech adhesive? It would eliminate the stresses from heat and be about as quick as welding.

billzweig · Feb 9, 2016

I am making mild steel assemblies by tig welding, silver soldering and soft soldering. For precision parts I think silver soldering will be better than welding and I did have good results. Soft solder is another joining technique I have used with mild steel with good results, especially when the end products are not exposed to high temperature.
For both hard and soft soldering I allow about 15-20 micron clarence for the solder.
A machine element I have made out of silver soldered mild steel (and stress relieved) over a year ago is still as accurate as can be (in this case) practically measured: +/- 2 microns.
Recently I've encountered a bad batch of hot roll mild steel. After aneling it became harder and brittle (!) and the part came totally distorted. For critical work it is important to get a material certificate.

Orange Vise · Feb 9, 2016

550-600C is typical and could be done at least twice, with diminishing returns after the 2nd or 3rd cycle. Cool inside the furnace, not in open air. You could also do a full anneal but expect scale to form on the surface.

How stable the assembly remains depends on the average/minimum section thickness and the amount of welding - the less the better.

What's the overall LxWxH?

Matt_Maguire · Feb 9, 2016

I'd stay closer to the 500C if it were mine. PWHT is mostly to knock down those areas under restraint due to the welding & not lose any strength in the parent structure. Hot rolled still has an attitude after rolling and softens a bunch when you get near Ac1 temps in the furnace.

Now, if this were a welding or machining fixture - I'd be spec-ing a lower strength filler for this. Tig with E70s-3 is still a fair bit stronger than the hot rolled in any hardness condition. Mig-ing the s-3 with Co2 or using 6010 or 13 drops the strength 10-15% getting a closer match.

Good luck
Matt

gbent · Feb 9, 2016

You can't lower the induced stress level below the yield strength of the steel at the process temperature. I would full anneal your part, and then do a cryogenic treatment. I don't have any data on the cryo treatment beyond using it to salvage some tool steels years ago. If nothing else, I'd put the part in a cooler with (but not touching) dry ice for a couple of days after the stress relief.

4GSR · Feb 9, 2016

I would do at least 650-700 deg C (1200-1300 deg. F). Oven cool, I don't remember the rate of cooling, but slow. Hold at temperature for 1/2 hour per inch of material thickness before cooling. I would do this twice, turn the weldment between heat treats, then last do the cryogenic treatment.

Pete F · Feb 9, 2016

malcolm rypauf said:
Pete, just a wild idea: could you assemble these with a high-tech adhesive? It would eliminate the stresses from heat and be about as quick as welding.

I'm not sure what "high tech adhesive" you had in mind, but I did consider high tech cap screws to assemble the parts

I guess you could also consider silver solder to be an adhesive of sorts, so that's definitely an option if I don't think the stress relieving will work.

The mirror bases are roughly 120 x 60 mm and the mirror back is roughly 60 x 60 mm. Strength is obviously not a real issue as all they do is hold an aluminium mirror from dragging in the dirt. I won't be cryogenically treating the weldments, there's a limit to haw ridiculous I want to take this. If I were going to do that I'd silver solder it or bolt the components together.

Peter from Holland · Feb 9, 2016

Mild steel grinds real bad in my experience And that can generate stress again
I would take a somewhat higher quality weldable steel
The equivelent of our ST52.3 or so
Machines and grinds much better

Peter from holland

Pete F · Feb 9, 2016

Peter I've ground a bunch of hot rolled steel and never had an issue with it. Indeed I'll typically do that as it's often the fastest way to knock the scale off and bring it up to a consistent size. There's definitely nicer steels to grind, but at the end of the day I think some aren't getting the message that these are simple little fixtures and I'm not trying for a manned mission to Mars here. It would be ideal if they didn't move, and that's what I'm aiming for, but if they turn out to be rubber I'll change one of the fixed feet for an adjustable one and just prove the mirrors each time they're used. I was just curious as to how much the stress relieving would relieve the welds, as I didn't want to set myself up for certain failure while there's alternatives. I've been caught out by welds pulling things badly in the past and setting up enormous stresses. I normally work around them, but in this instance would like to be a bit more proactive in managing them. I think at this stage I'll give welding a go, maybe machine them all but just weld up one mirror and see how it goes, that way if it doesn't work I can always change tack mid-stream. Thanks for the help, advice and opinions for those who contributed.

tailstock · Feb 9, 2016

I agree with Peter. If you are going to make a precision instrument you need to start out with good certified metal. Standard hot rolled or cold rolled can be anything from a melted down crankshaft to a refrigerator door. :codger:

Pete F · Feb 9, 2016

tailstock said:
I agree with Peter. If you are going to make a precision instrument you need to start out with good certified metal. Standard hot rolled or cold rolled can be anything from a melted down crankshaft to a refrigerator door.

True. But what if it is? I'm not going to get a certificate for what turns out to be a few hundred millimetres of 8 mm steel plate! My steel supplier would throw me out the door! I appreciate the input, but I do wonder if some of the advice that is dispensed so freely here comes from practical experience! I'll run with what I have and make the best of it. It sounds as if Bill and others have done something similar and managed to pull it off. As I mentioned above, I can self-prove these mirrors and if it all turns to poop make them adjustable, so I don't think I have a lot of skin to lose here by giving it a go. I'll weld and give them a few cycles at 700 C and see what happens.

Stress relieving mild steel weldments

Titanium

Stainless

Titanium

Stainless

Hot Rolled

Titanium

Diamond

Aluminum

Titanium

Cast Iron

Stainless

Titanium

Stainless

Diamond

Diamond

Titanium

Diamond

Titanium

Hot Rolled

Titanium

Similar threads