ME 310: da/dN Testing

Hey guys, another research update for you all! So last time, I talked about fatigue pre-cracking my rail specimens and how the fatigue process worked. Basically, I needed to fatigue pre-crack my specimens in order to get a crack that would mark a path for the rest of the crack to follow. This process is to help with da/dN testing, which is a test that measures the change in crack length per change in number of cycles. Since I had fatigue pre-cracked a dozen specimens, I then began conducting the da/dN tests.

So da/dN…. What does that mean and why is it so significant? Well, as I mentioned before, it’s basically the rate at which a crack grows for a given cyclic loading on a specimen. What we are essentially trying to find from this type of test are 2 things:

1. The cyclic loading at which there is no longer a crack growth rate- this is important because it allows us to find the Kmin Threshold value. This is the minimum stress intensity factor (proportional to load) that a crack starts to grow at. In other words, a what load do you stop creating a crack on the specimen.

2. The slope of the da/dN vs Delta K graph. Basically, as long as the increase or decrease in crack growth rate maintains itself steady and moderately linear, then we can characterize the behavior of the crack for a given load and predict what the crack growth rate will be at a given time.

So the past month has consisted of these tests, but just like the fatigue pre-cracking, it has been a little tough figuring out the best way to go about these experiments. To give you an example, I ran a crack growth rate test a several weeks back and obtained the following results:

da_dN Vs Delta K

Now compare this experiment to one that was performed by a certain railroad company.

A36 steel da-dN test

So as you can see, there is definitely something wrong with what I was doing and it took time to figure out what was wrong. Apparently, the reason why my data was so distorted was because I had stored the data points was based on the number of cycles and not on the crack length. So in some instances, I would receive negative crack growth rates because the strain gauge would measure a smaller crack length at certain instances (remember it’s cycling, so there is sinusoidal amplitude that can give you positive and negative numbers), which would cause a negative difference. So although you may think, “Duh, isn’t that obvious??”, small little issues tend to occur very often and one always has to be on the lookout for small problems like these in order to obtain accurate results. Nonetheless, I think I’m starting to get the hang of all this da/dN testing and can now start running these tests with ease. Here’s a graph of one of the good tests that I ran:

L112 dadN test

So as the change in stress intensity factor Delta K increases, the crack growth rate obviously increases because the specimen’s area gets smaller  and because your putting larger load on it as well.

So that’s where I’ve been essentially with these tests. I’ll be sure to get a fracture toughness test in soon enough and hope that the tests are valid. More updates to come!

ME 310: Fatigue Precracking

Hey everyone. Here’s the latest update with everything thus far:

This past month consisted of fatigue precracking of our compact tension (CT) specimens and getting the software of the hydraulic Instron to work. Might not seem so much, but when you have other students wanting to use the Instron…. Well your in line to face delays. But anyways, things are at least much smoother than they were last semester and we finally got the hang of this fatigue precracking.

So what exactly is fatigue precracking? Well it is essentially placing a specimen under a cyclic load for certain amount of time and propagating a small crack at the end of a notch. In our case, we usually fatigue precrack our specimens until the crack grows a certain length and then we place them under the microscope to make sure that the crack on each side of the specimen is of equal length.

So how is this all done?  Well check out this photo:

img_5044

The above picture shows one of the CT specimens attached to the 8801 hydraulic Instron. The specimen is connected at the holes by Grade 8 dowel pins that are also attached to the clevises made in house. The ends of each clevis have 1/2-20 holes for the same size bolts, which are then attached to the hydraulic grips. And last but not least: the clip gauge extensometer. This devices measures the crack growth of the specimen and sends it to the da/dN software that calculates the crack growth rate per increasing amount of cycles (a= crack length and N=number of cycles) along with other various parameters. While fatigue precracking a specimen, the following screen is shown:

img_5133

This screen displays various parameters and plots that are changing with each ongoing cycle. One of the important things to know when fatiuge precracking is the way you go about doing this procedure. That is, are you letting the Instron run until you click Finish Test or are you setting a limit to the number of cycles or crack length that you want it to stop at? What types of loads will the specimen see during the fatiguing: increasing, decreasing or maybe even constant? Without getting into too much detail, we basically ran our test at a constant delta K, which basically means that our loads were decreasing as the fatigue precracking went on.

And thus so far, we’ve been able to fatigue precrack about 12 specimens. Some of the problems we encountered through the past month is that we initially had no idea how to work the da/dN software when we had finished machining the specimens. I had to contact Instron and other places to get answers to some of the questions we had. We luckily got some help from Professor Vinci in the Material Science department and he helped us a ton on how to work the Instron itself. Another problem we faced was that we broke 3 specimens during the fatigue precracking. We figured out that the software can sometimes go haywire with the some of the calculations it performs and that it can then go at full load and break the specimen as shown in the picture below.

img_5046

If you examine closely, the specimen is broken at its centerline where the notch is formed. This isn’t too critical of a problem since we have 36 in total, but I was honestly amazed on how much of a load the Instron can output. We never got to see at what load this specimen broke, but it was definitely higher than 15kN for sure. In order to debug this problem, we just defaulted the coefficients of the calculations then we never had the problem again.

But anyways, that is where we are so far. Next, we will start da/dN testing where we would like to obtain a value for Kmin threshold, which is basically the minimum stress intensity factor needed to begin propagating crack. Stay tuned for more updates!

ME 310 Research Update: Good News!

Hey guys! Now that it’s back to school, it’s time to rock and roll, but before we talk about the future of our Fracture Toughness Testing, let me give you a little update of what went down during winter break.

So while I was back home, Dick Towne, who is one of the staff members of the ME Department, finished making our clevises for our Compact Tension (CT) specimens. The clevises took some time to make considering they were made out of 17-4 Stainless Steel where the material was heat treated to an HRC value of 40, which basically meant it was really hard stuff! Nonetheless, the fact that we now have them helped us get closer to begin testing…. But wait! I have another update!

img_5015

Clevises are attached at both ends with 1/2-20 bolts that connect to the grips of the hydraulic Instron machine. We are using the 8801 Instron located in Whitaker Laboratory.

While the clevises were being fabricated, the 36 CT specimens that we sent out a while back finally returned with their wire EDM notches. 3 of the specimens were also fatigue pre-cracked, which basically means that we can conduct a Fracture Toughness test straight away. We still have to fatigue pre-crack the rest of the specimens, but this is something we don’t mind doing because we want to learn different ways to fatigue pre-crack. Fracture Toughness testing hasn’t been performed at Lehigh for several years now, but relearning how it all works including the preparation, setup, and testing procedures are the things we as a university want to how to do.

img_5017

Here are 8 out of the 36 specimens that have returned. The clip gauge attaches to the mouth of the specimen and the wire EDM notch has a thickness of .012″.

The next steps towards this research project will be to fatigue pre-crack our first specimen. A few days ago, we went to the Instron machine and examined the software that both measures K1C (fracture toughness) values and conducts fatigue pre-cracking. After a few hours of playing around with the software, we think we have it down. Tomorrow will be the moment of truth and hopefully, we get some good results.

ME 310 Final Semester Update

Hey everyone! It’s been a while since I last posted an update on my ongoing research, but that’s been quite the trend for me, huh? Updates every few weeks. Anyways, here’s the latest scope with what went down to finish off the semester:

So I had mentioned last time that we had finished machining all of the specimens. Well, all of the specimens were then sent out and were supposed to be wire-EDM, but unfortunately that has not happened. The company that we sent our specimens to got confused with some of the dimensions that we wanted in our notched specimens and didn’t notify us until three weeks later! We thought we would get some of the specimens fairly soon, but that was not the case. We eventually talked everything out and they are finally now wire-EDMing the notches.  The company is also now wire-EDMing out the 0.2” slot that is shown in the figure below.

lti-specimen

The rest of the dimensions remain the same, however, from our previous machining. We are including a slot into our specimen because this allows a more intact, sturdy method of holding the clip gauge while also allowing for a better crack to propagate. We at first were going to use knife edges to hold the clip gauge, but decided to avoid that method because having a slot surprisingly comes out much cheaper than the original plan and because we can easily fit the gauge clip within the specimen.

I had also mentioned a while back that we would fabricate a clevis in house to avoid paying a large amount. The clevis however, has also been delayed but this is because our shop manager has been really busy trying to accomplish 3 or 4 projects at the same time, so he hasn’t had enough time to fully create our specimens.

It’s unfortunate that my team and I could not run a fracture toughness test. I was really interested to see what the results would display, but the wire-EDM and clevis fixture situation are things that I personally cannot control.  I am consistent in asking for updates from everyone, but I can only give a reminder for these projects to be completed.

Next semester, however, I will continue my research project and be sure to get these tests done as quickly as possible.  There are 36 specimens that need to be tested, so figuring out how I will get them all done will be quite a challenge, but I hope that with some help from graduate students, I can get these done quickly and efficiently.

More updates to come next semester!  Winter break is here!

ME 310 Research Update

Hey everyone, it’s been a while since I’ve last made a post, but here are some of the things that my group and I have done so far:

All of the specimens have been machined to their correct dimensions.

img_4640

The specimens inside the box are half of what we machined out.  There are 18 total specimens inside the box and they are from the crown of the rail.  We still have to put 2 holes on each one, but I was able to drill some of mine out.

img_4643

This is how the specimens should look like before they are Wire-EDM.  There are two 6mm (drilled and reamed) holes and the specimens have been polished so that when we perform the fracture toughness tests, we can see where exactly the crack begins to propagate. I just sent out the yellow box yesterday to get the specimens wire-EDM and now we’re in a waiting game.  The clevis is currently being built and until then, there’s not much for me to do.  We asked for some specimens to arrive back early so that we can get them tested, so we’ll see how fast they get back.

Stayed tuned for more updates!

ME 310 Research Update

So some progress has been made from the previous time I posted about my research project.  Again, there’s been a lot of machining from my part and I have practically finalized some specimens that are ready for wire EDM (Electrical Discharge Machining).  What’s wire EDM?

awedm_04

By looking at this picture, it’s basically a way to cut small notches into different parts or materials by exerting a very high voltage between two electrodes separated by a dielectric liquid. Once the voltage is high enough, current passes through the diodes and cuts the whatever is in between. Sure don’t want to get my fingers between there!

But anyhow,  there are 3 of us working on this research project and each of us has a different type of rail with different material properties.  We all have been working on the crown of the rail (the contact surface that the wheels make with the rails) and we each have three specimens that are almost ready to be wire EDM-ed.  My crown parts look as such:

img_4557

Cut 2 and Cut 3 are going to look like the three “cubes” on the top.  All I need to do for the top cubes is place a smooth surface on each piece because when we perform our Fracture Toughness test, we will need to observe the crack that propagates with each load that is applied to the specimen.  It may seem as though this specimen preparation is a very simple task to do, but it takes a lot of time and if you mess up once, well…. It’s REALLY bad.  This isn’t a project that you just make up on your own.  In my case, Lehigh is essentially a contractor to the Federal Railroad Administration that is running these tests for them to extract the necessary information needed in order to improve that quality of these railroad tracks.  Sure we have funding and could always get more rail, but starting all over takes too long and progress needs to be made.

Most of our work is done on a Bridgeport mill where we do most of the precise material cutting and finishing, but we sometimes use the bandsaw to cut large excess material.

mill2

This is a typical Bridgeport mill that we use at the student shop in order to do our work.  One must be careful when milling.  Things such as feed rate, RPM, and material wear on the tool must be noticed at all times so that you can get the best results.  Don’t forget about keeping track of how much material you take off!  You’ll be toast if you don’t!

More updates to come!

ME 310-Independent Research

As a senior here at Lehigh, you can choose different elective classes to take in order to fulfill your major requirements.  Some people fulfill their requirements by taking classes related to Aerospace to complete their  Aerospace minor or classes to complete their Energy minor.  Whatever it is, you can choose to take any 300 level mechanical engineering course to complete your bachelor’s degree.  I’m actually not pursuing a minor because frankly, I’m not interested in either of these minors so I’ve chosen to take a different path relating more to strength of materials and fluids.

But going back to what I was saying, one of the 300 level courses that you can take here at Lehigh is ME 310, which as the title of this blog mentions, is an independent research class where you pair up with a professor and do research for him or her.  My research project is with Professor Herman Nied and I’m working on the Mechanical Testing of Hardened Railroad Rails.  I was actually involved in this project during the summer where we performed tensile tests on many specimens with the Instron testing machine and we had to calculate several of its properties.  Now I’m working on creating Compact Tension (CT) specimens from several plates of railroad that were cut off the cross section of the rails. Once we make these specimens, we’re going to perform Fracture Toughness tests for more analysis.

These are the rails that we started off with:

railroad1

These are the main rails that came from Mountaintop and were brought down to Whitaker.

railroad2

After cutting flat sections through the cross sectional area of the rails, this what each “slice” looked like. There are 3 slices in total.

railroad3

This a picture of the crown slice of the rail where we measured its height.

railroad4

After cutting 3 sections for each slice that I had, I labeled everything to make sure I knew where everything was.

railroad5

Here I’m basically using a face mill to machine down the rust off the crown.

So that essentially how far I’ve gotten so far, but I’ll hopefully be posting more of this with weeks to come.