Ray,
I hope you're not upset by anything what I have said or posted (if so, please accept my apologies), since I haven't received any reaction, but I was interested in the method you have mentioned and have busied myself somewhat with it. Thus please let me come back on this and try to transfer what I mean to having understood of this procedure.
First of all I have recalled myself what is understood by using the technical term "Residual Stresses". These - at least as to my best knowledge - are stresses, existing within the component or workpiece without further outer influences, e.g. forces.
Next question to be clarified should be: "How are residual stresses generated?" which can be replied by knowing that any mechanical (e.g. grinding) or thermal (e.g. welding) influence can create these stresses. The disadvantage is with all these stresses, that - in worst case - they can initiate the failure of the component while processing (e.g. cracking), or - at least - can negatively influence the component in a life time reducing way.
Now one can distinguish different grades of residual stresses as follows (I beg your forgiveness for not knowing the exact English terms):
1. "1.Grade" Residual Stresses. These are acting over longer macroscopic distances and cause macroscopic and thus visual detectable deformation or warpage, respectively. Important is by the way, that height and spatial direction of "1.Grade" residual stresses are equal.
2. "2.Grade" Residual Stresses. These are acting over smaller material distances (~ 1 grain or partial areas of a grain). Macroscopic deformation is rather less possible but can be caused by a strong perturbation of the inner material's balance.
3. "3.Grade" Residual Stresses. These are acting only over atomic and thus invisible distances. Macroscopic deformations are not observable.
As one can see by now, the most "dangerous" ones are the 1.Grade residual stresses, since they can reach heights of deforming the workpiece in a macroscopic scale.
The general problem in measuring residual stresses existing within a welded workpiece is the fact, that this is only indirectly feasible. And this again either by non destructive testing methods or by mechanical methods. The first named group includes amongst others e.g. X-Ray diffraction (XRD) used to measure the atomic scaled distances of a stress afflicted material vs. those ones of a stress free material. The second group includes e.g. the hole drilling method, you have spoken about. The physical principle behind this and the other mechanical stress measurement methods is, that by e.g. drilling a hole into the stress afflicted workpiece a stress displacement occurs in the adjacent area of the drilled hole. This displacement again causes minimal deformations which can be measured by using especial strain gages to be adjusted in a particular configuration around the hole. By the measured deformations again one can calculate the original stress condition as it has existed within the workpiece before the hole was drilled.
As I could find out, the major advantage of the "hole drilling method" lies in its mobility and flexibility, due to the testing equipment's low size, please see also the attached Hole_Drilling_Device.jpeg.
But now... as our appreciated fellow Jeff said once, the whole item becomes a "black hole". Since from now on, one has to use the calculating methods relating to even the intricate details cohering to the displacements, originated by the drilled hole, and the measured stresses by using the strain gages. The math to be used for even these calculations is on a rather higher level, just as I have stated another time. But however, since I suppose that you or some of the appreciated other fellows here in the forum are quite interested in the way of calculating the stresses been measured by the strain gages, I will attach the formulae as a separate document, see also Drilling_Hole_Residual_Stress_Calculation (only for information!!!).
However, besides the necessary mathematics I guess it might be interesting to have a short view upon the physical coherences between the residual stresses (within the material) and their relaxing - or even displacement - by the drilled hole may appear, please let me try to describe what it means to relax the inner stresses by drilling a hole into an imagined plate.
Let's imagine that there is the easiest case for residual stresses to be detected. This case means the strains within the material are acting single-axial. This condition is designated by the Greek symbol "sigma" and the suffix "x", which means that the strain is acting in the "x"-axial direction. Now let us imagine that we do have a plate (workpiece) afflicted with even the "Sigma_x" stresses. Please see the attached sketch X_Axial_Stress_Condition.jpeg.
To calculate the relaxed strains one has to prior resolve the KIRSCH'-Equation, see also the Drilling_Hole_Stress_Calculation.jpeg. After having resolved these equations, one can superpose (supposing always to have a linear-elastic material behavior), i.e. one can subtract the single-axial stress condition from the KIRSCH'-Solution.
Schematically this looks like the attached Superposition.jpeg. Here the left sketch shows the analytical solution for a (theoretically infinite) large plate containing a hole (after KIRSCH). The center sketch shows whereas the plate condition P R I O R the hole has been drilled and the right sketch shows the condition and the relaxations and strains after having drilled the hole. What does this mean in theory? The relaxed strain value equals even that strain value which is originated within a drilled plate even by subjecting the internal hole boundary with a boundary load (- Sigma_x). The boundary load has thus the same value as direction but the converse orientation.
And this is only the solution for the single-axial stress conditions. Three-axial stresses are quite harder to calculate, as surely imaginable. Thus I would like to avoid to continue here with the calculation of higher stress conditions (3-axial). You know, I have never learned to calculate algebra of matrices, which is however necessary to calculate more dimensional stress conditions.
Well, as you can see. It is as Jeff so wisely said once. Stress calculation and simulation might be a black hole for many of us welders, at least however for me, since I have never learned to use the necessary tools in a sufficient way. And to be honest, I do really not know, if I will ever learn these admittedly interesting but nonetheless quite tricky little things over the rest of my life. But who knows, nobody knows the day of tomorrow...
Best regards,
Stephan
Does anyone know where I can find some really good info on calculating residual stresses after welding?
Forum
