Could someone kindly explain to me how inclusions in steel do not constitute a crack?

Robby

Dear Matt:
A couple of semi-related questions:
1) Would an indication/inclusion deep beneath the surface of, say, driving axle on a steam locomotive be problematic?
2) Would indication (s) of forging flow lines beneath the surface of, say, a driving axle of a steam locomotive be problematic?
J.David

Inclusions and indications are terms that tempts this metallurgist to stick his head up and reply... Nondestructive inspection techniques genrally have a sensitivity to inclusions that is proportional to the thickness of the section being evaluated. Very thick sections will need to have larger inclusions before they are detected and become indications. The opposite is true unless things get to be so tiny they drop below the threshold of what you can see. (There are volumes of exceptions and limits and so on that make the last two statements very general at best).

Now what J. David asks is really a much more profound issue. The effects of defects is a horribly complicated topic and it stretches far beyond the relatively simple issue of detactability. The complication comes from how the indication reacts to applied stresses. The specific case of an inclusion in steel is a good one to consider. All steel objects have inclusions. Really good quality steel has tiny ones and poorer grades have more and/or larger ones. Depending on where they are in a part made from the steel and how that part is loaded in service may be a boring or dramatic story.

Let's consider J. David's two questions. If the inclusion is at or near the neutral axis of the axle, it can take train rides for many many years before it casues the wheel to fall off. If the inclusion is near the surface and is not parallel to the main forging grain direction, it will be the place the fatal crack orginates from. In general the closer to the surface the worse the conditions in terms of stress. Forging flow lines are usually not a big problem except in very highly stressed parts. Most locomotives are designed to have a generous factor of safety . That factor is, among other things, there to account for material inconsistancy. The flow lines originate from minor alloy element separation during the casting of the ingot. The differing regions of chemistry get pushed and spread out by the forging hammers / rolls and leave the charactistic grain structure of forged steel. Unless it is determined to be extreme segregation it isn't normally a problem. In NDI it can show up in magnetic particle evaluation and can confuse some evaluators.

Sorry for the long answer. It isn't often metallurgists get to spew on this board.....

Fred

Here's the rather pessimistic (and really funny unless you work with foundries) definition of the word "Casting" given to me by an old foundry engineer co-worker of mine, he was quite a character!

Casting; "A series of voids and inclusions held together by metal of questionable integrity".

Sir Fred did an excellent job of outlining a “horribly complicated topic.”

To elaborate on J. David, Esq.’s original questions, it is necessary to get specific on definitions. In the engineering world, words that the general public uses interchangeably have very specific meanings. In the NDE realm, indication and inclusion are not synonyms. An inclusion will give an indication on a UT scope, a crack will give an indication on a UT scope as will a machined keyway or a change of cross-section. Indications are divided into two realms, the relevant and non-relevant indication. If a trace or echo shows up on the UT screen where a keyway is known to exist, the trace is analyzed to verify proper geometry of the keyway, noted and the inspection continues.

If an indication appears where none is supposed to exist, then the detective work starts. The trace shows something that is not supposed to exist, so it is a “relevant” indication. Location, size and the type of defect that is generating the indication are all important now. A linear forging inclusion will show an indication both axially and radially. A crack may show an axial trace, but be nearly invisible when shot radially by a UT probe.

To be specific to your questions J. David, Esq.:

1. Any relevant indication that exists near the radial edge of a round shaft with shear and torsional loadings, which generate tri-axial stresses, such as exists in a steam locomotive driving axle, would be failure critical and problematic to diagnose. The indication could be located a specific distance from the end of the axle, but if the beginning of the indication is located in the wheel fit, it would be impossible to judge axial length because radial examination would be impossible without pressing the wheel off the axle.

2. Forging flow lines due to casting segregation as Sir Fred has identified, should be invisible to both radial and axial UT scans. Wrought iron and even cast iron are difficult to UT due to dispersion and absorption of the sound waves by the slag and carbon nodules, respectively.

Referenced are certain sections of ASTM/ASME A-388 Practice for Ultrasonic Examination of Heavy Steel Forgings

It is also good to note that any score, scratch or machine mark on the surface of a stressed shaft such as an axle is a location where a fracture may initiate. Something as simple as a brake rod riding upon an axle is enough to cause an axle to break in two at that point.

Good posts on an important topic that is little understood outside of the materials and testing world. I for one got carried away and neglected to address Robby Peartree's orignal post. An inclusion in steel is a solid material that is very initmately connected to the surrounding metal. If it were large and exposed so you could see it, it would be nearly impossible to remove it witout tearing out some of the steel surrounding it. It really is part of the metal body. Having said that much, it is also very different interms of strength charateristics. The steel surrounding it has ductility, the ability to stretch and deform before breaking. Steel is also moderately stiff. Most inclusions are a glassy substance that has little if any ductility and is very stiff. These two characteristics, stiffness and ductility, are why an inclusion can generate a crack. When the steel part with the inclusion is given a load to carry it will stretch in porportion to the load. Because it is stiffer material the inclusion won't stretch as far. The difference is at the border between the inclusion and the steel. With enough load the steel and the inclusion may part company. Or if the load is in the right direction the inclusion may break. Now the part has a tiny crack along or through the inclusion. If the loads are high enough or repeat often enough, (cyclic loads like an axle turning under a load), the crack will grow. It may become large enough for an inspection technique to find it and/or for the part to break.
This crack initiation process doesn't need an inclusion. The previous post mentions scratches and grooves. In high cyclic loading, like an axle, the outer"fiber" of the shaft is terribly sensitive to notches. The area just inboard of the journal bearing sees the highest bending load and must be free of this sort of thing.

Enough...

Fred

Wow....A real discussion on RYPN!

Robby P.,

Is there a forged axle in the steam industry that is FREE of any type of inclusion? Based on what we read, and based on what Sir Matt A has posted, it is clear that UT inspection of a forging will reveal some type of "artifact" for lack of a better word. Read 10.1 of the ASME section posted by Mr. Austin.

The reality is Robby, if we discard every steam locomotive as a result of finding any "artifact" in the UT process, we are doomed to reject them all. Fred, whom I do not know, has made it very clear that there is no such thing a perfect steel....or forgings for that matter.

Fred,

Thank you so much for your time and support in the matter (at least I read it that you supported the idea of minor inclusions in a forging and note it as a some what normal condition....am I at least on track?

Mr. Matt A.

I take exception to only one thing you mention. A Level II Inspector is capable of reading the machine to determine what is "real" and what is not.

To Sir Conrad:

I am in your boat.....as I read your questions/comments. In the real world...what is the end impact of these "inclusions"? Though Matt has some valid points, I side with Fred in the fact that it will take a very long time for any of these so called inclusions to produce a real problem.

All:

Fred, you seem very qualified. Matt A. You have seen plenty of UT inspections, as have I. J. David you have been around the block. I would be very interested in hearing a brief explanation of what a defined CRACK would look like on a UT screen. I would be further interested in hearing what an INCLUSION would look like on the same screen. I am further very interested in hearing what the teaching guidelines are in helping a Level II or Level III inspector understand what they need to look for and how they define inclusions or artifacts from CRACKS and definable defects. Further, on an axle that is four feet long, what recomended angles would be used to properly UT the axle....or would you attempt to shoot straight through the axle?

After the given explanations, I will provide the procedure that was used to determine the extent of defect in the No. 40's axles...or should I say the lack of defect.

As a brief background on me. I studied NDE as part of my welding degree. I strongly considered becoming a level II inspector. I even took the classes and I had two job offers to become a qualified inspector, however, I opted to stick with my welding and love for trains. For those that question my experience or training, I hope the above explanation will suffice.

I look forward to a future discussion on the matter,

Respectfully,

John E. Rimmasch
Wasatch Railroad Contractors

jrimmasch@wrrc.us

Greetings All:
My experience with scope type UT devises has been in the inspection of axles and crank pins for cracks. Typically the little "T" gauges are only used for thickness. They might find a crack, but since the distance into the material is limited, the more powerful scope type is used for longer objects, say steam locomotive axles. Also the scope makes it possible to interpret what is "inside" the object.
The tester is adjusted so that it will "read" to a certain "depth" into the axle (somewhere past the wheel seat and bearing surface).
Inclusions and the like show up as "blip" on the screen which is repeatable but small, when the transducer is moved to one side or the other, the "blip" disappears.
A crack shows up as a line on the screen at a given, repeatable location. If you place the trasducer at or close to the radial center of the axle at one end and move it towards the outer edge, it will show up prior to the transducer reaching the outer diameter of the axle. Depending on the diameter trasducer used, you can get "good" readings right up to the edge. It is essential that the geometry of the object is know (diameter of axle at wheel seat, diameter at hub, etc.) so an educated interpertation can be made.
Note: We found two cracked axles, later confirmed by dye-penetrant testing, both on Valley Railroad No. 40 in 1988. This discovery precipitated the replacement of all of No. 40's driver axles (another was nearly at it's limit as to minimum diameter, the last one "just because").
Contrary to a posting above, I beleive that "forging flow" can show up as an indication. In the mid-1970's I hired Pittsburgh Testing Laboratory to assist with a pre-purchase inspection of a locomotive for a client. One axle showed repeated "indications" deep within the axle, which did not come close to the surface. The "indications" showed up as a regular, repeatable series of "blips/lines" in circular pattern. They extended neither to the center of the axle or near the surface. The UT reader said that he often saw indication of this sort while testing new forgings for steel mills. On this axle we used both a 1/2" and a 3/4" transducer. Using the larger transducer, the "indications" were minimised.
Be well,
J.David

I am not sure if it was stated elsewhere, but this thread seems to reflect on the recent return to service of Nevada Northern 4-6-0 #40.


http://nevadanorthernrailway.net/throttle/throttle092309.htm

Dear Mr. Rimmash,

Please reread my question as I asked how is an inclusion not a crack? It is a simple question. I was expecting an answer similar to what a metallurgist hears from a welder that the metal crystallized. It is a description often used by people who have not had the opportunity to educate themselves in metallurgy. By understanding what the intent of the used words are you can sometimes find out what is really going on. Understanding what the problem is the first step in solving a problem. This practice has served several friends well and I thought I would try it. Fred’s answer to my question is appreciated and I can only wonder how many minds he educated by his posts.

You will find that I have made the comment that every piece of metal (no matter how it is formed) has cracks. What is most important is whether or not they are growing. A crack not actively growing should be of no concern however they should be monitored.

I would like to suggest a few books "Manufacturing Processes and Materials for Engineers" by Doyle/Keyser/Leach/Schader/Singer publisher is Pentice Hall, "Mechanical Metallurgy" by Dieter and published by McGraw Hill and ASM METALS Handbook V11 "FAILURE ANALYSIS AND PREVENTION" which can be purchased at the ASTM web site. Manufacturing Processes and Materials for Engineers has a discussion on forging and it discusses in modern processes that you oversize any rolled forging so that the inclusions introduced from the forging process are removed before the steel is in final product form. To me the real question is how is the base metal going to behave? With either a crack or an inclusion you have a separation of the base metal an effective reduction of the cross-section and resulting increase in stress levels cause a move of the materials behavior toward failure behavior.

Cracks can initiate in areas of surface change such as the transition from the smooth bearing surface to the rough surface particularly where bearing are at or beyond their turning limit. There are limits to testing equipment and the advantage to each time has to be known when choosing the appropriate method.

The inability to test for an unknown defect is the most frightening unknown we face. The courts have not given a free pass to such ignorance. In fact one accident can instantly and irrefutable change lives in an instant that can affect the whole industry. Inspectors for the FRA is watching both operating and maintenance standards, or the lack their of, of each operation in the tourist/museum industry. If we do not make sure that we are as safe as reasonably possible we will see a far more restrictive operating environment in the future.

Respectfully

Robby Peartree

I had an uncle who lived near Pittsburgh and did quite a bit of work over the years with Pittsburgh Testing/Technical Labs, or as he and everyone else called it at the time, PTL.

Some of you can already see where this is going.

Well, he related to me that, years later, he saw someone in a grocery store with a t-shirt saying "The PTL Club". He remarked something to the wearer to the effect that "Boy, I knew Pittsburgh Testing Labs was big and popular, but you mean to tell me they have their own CLUB now?" He said he was rather dourly and haughtily informed of the TV show of that name starring Jim and Tammy Fae Bakker, and the "proper" meaning of the acronym. It was rather apparent to my uncle that this individual had no sense of humor, and was offended by the insinuation.

As he said, "I never really thought much of that show and their fans after that....."