Temperature Effects of Elastic Modulus for copper

[FerrariEngineering]

Hello-
I am running a simulation of a simple brazed copper assembly; brazed at 1050°C. I believe the Young's Modulus will drastically change after brazing at this temperature since the annealed state is "dead soft". My analyst colleague is using an annealed value for E of 15E-6 psi, which isn’t much less than cold-drawn. My deflection values are nowhere near the actual measurements I am empirically seeing, and I’m thinking E must be much less. I have located some graphs that do show E decreasing with temperature, but none close to 1050°C. Qualitatively speaking, I can vigorously swing a length of copper annealed at this temperature and see it bend like a noodle, so the E has to be lower. I am comparing to a behavior of something like lead??
Since my simulation model really only requires E in terms of the material properties for deflection, I can’t think what else is causing such a discrepancy?? Other boundary conditions are quite simple.
Anyone have knowledge of material properties for REALLY annealed copper at 1050°C??
Thank you in advance.
Chris

 

[FerrariEngineering]

@FennLane
The assy is a rectangular cavity with specific features (openings) at given wall thicknesses so we can measure deflections with typical QA instruments. As I mentioned, we also confirm the deformation with frequency measurements as the cavity volume is sensitive to frequency shifts.

The simulated model display deflections smaller than empirical measurements by a magnitude of ~15.

I agree with you about the level of residual stresses expected, especially in Cu, but we do see this happen with larger parts that are not symmetric during cooling but more so with materials like SST and especially dissimilar metals. I don’t expect much if anything, but I’m curious if there are any, and if so at what location they reside.

We are measuring pre-braze, post braze and after pulling the assy under vacuum (the load). We are basically measuring a cavity spacing uniformity by measuring the flatness of specific features. With cavities under vacuum, the deformation is more “oil-canning” like. Again we confirm the measurement deltas with frequency measurements.

Please note at this point I am not trying to prove any theory about reduction in modulus. Understanding the FEA model outlining, I just explored and discovered that E significantly decreases at elevated temps, and at this property difference at these temps mathematically made my model became much more accurate so I really thought I was onto something. Some other analysts speculated the same, but I am learning from others including yourself that E will recover upon cooling. That’s what beautiful about problem solving and researching with available sources.

So I am continuing to explore as having an understanding of a simulation baseline for brazed (H2 furnace) copper will be very helpful for people in my field. I am inquiring with some other contacts from the national labs and hopefully will learn where I am making the disconnect.

 

[btrueblood]

Is it possible that the flat areas that are "oil canning" are actually bowed outwards very slightly under no load, and then experience snap-thru buckling when the vacuum is drawn?

Another thought - what is the braze alloy used? Are you modeling the changed modulus and section properties of the joint (due to braze/parent metal alloying)? Could the thermal contraction of the braze during solidification be setting up the residual stresses you hypothesize?

 

[FerrariEngineering]

We are measuring the surfaces pre and post braze, and there isn’t much different, and the flatness is pretty good; basically about the same as the machined components. As soon as we apply vacuum, some critical surfaces are displaying deflection.
We are using a copper-gold alloy and although the diffusion of the alloy at the copper will definitely cause some material property deviations, these joints are located at the far outside dimensions and the deflections are occurring in the center. Think of something like an 8X2X1in rectangular box made of ¼ plate with a 1 ATM difference between the outside and inside surfaces; the oil-canning would take place in the center of the sides with the greatest surface areas, furthest from the supported ends as expected.
I really don’t expect much, if any, residual stresses after braze. I am just curious if running the model with the thermal loading from the braze would indicate any change. Definitely wouldn’t make up the magnitude of difference I am seeing with an E value of 15E6. I’m going to play with my fixture boundary conditions again….

 

[MagBen]

Since deflection occurs as soon as vacuum is applied, does that mean you use 1 ATM (.1 MPa) as your load for simulation? Is the weight itself also taken into consideration?

 

[1gibson]

Does this thing have a preferred position, meaning can it "pop in" and "pop out" ? Maybe some minor dimensional change after heat treatment is causing the diaphragm to want to move, like a hot cookie sheet?

 

[MagBen]

One last point to make: for an anneald soft material, a load much smaller than YS can cause deformation off the linear portion. Attached is a stress-strain curve for annealed Cu indicating non-linear charactertics when the stress is only 1/6 of YS.

At high temperature (1030C) when YS is very small, the load for a linear elastic deformation would be very very small. 1 ATM load could be surfficent to lead to non-linear deform in which case defelction would be greatly underestimated even using the true E (half of RT E) for simulation.

 

[MagBen]

oooooops, here is the attachment