| JADHAV RESEARCH |
| Web pages of Ajit Jadhav |
| Home |
| Training |
| Consultancy |
| Work Experience |
| Academics |
| Research |
| Publications |
| Miscellaneous |
| Links |
| Miscellaneous
-- My Photos -- Photos: COEP and University of Pune -- My B.E. Project: A Personal Recounting... -- Some Random Thoughts on Examinations, Marks, Ranks and Education -- Future Updates -- Rights |
|
My Photos:
Very Important Notice: It is this particular sub-section in this entire site about which adverse comments shall not at all be tolerated :) (Click on the thumbnails to view larger size images.)
|
Photos: COEP and University of Pune
Main building, College of Engineering, Pune. COEP completed 150 years in 2004. Guess this building itself is some 140 years old. 
Main building, University of Pune. This building used to be the British governor's residence... Give anything to Pune people and they will turn it into something for education! (Yes, as a Puneite, even I do like this trait of the people from Pune!) |
|
My B.E. Project: A Personal Recounting
As I said on the Academics page, the idea in my BE project was to see if formability (or ductility) of metals can be estimated by bending metallic bars until they started cracking, and measuring the inner radius of curvature at the onset of cracking. The smaller the curvature at cracking, the higher the ductility. I wanted to verify the analytical equation which related the two using mechanics of solids. In the actual experimentation, I found many flaws with this idea--how to put it to work, except for, may be, only a narrow range of materials. Firstly, metals have some real spring-back, making the measurement of the radius of curvature real hard. The spring-back makes those curvature measurements more and more imprecise exactly where you need to distinguish them most critically: at small radii. (We are talking about measurements possible on a shop-floor with a few circular disks of various sizes and with unaided eye, not using a magnifying profile projector.) Secondly, unlike sugar-cane or bamboo, metals are far too ductile to begin any cracking at all. If you don't believe me, try to break a paper-pin in a single bending operation. Here are a few rules. You are not allowed to press sideways on the bottom of the U shape that the pin will assume after bending it through 180 degrees. Unlike a match-stick, the metallic paper-pin won't break even after a free 180 degrees bend! That's what ductility of metals means. Another rule. It has to be a single bending operation. You can't bend the pin repeatedly in opposite directions in which case it becomes a fatigue failure, not bend-cracking. Thirdly, apart from their ductility, metals also are far too strong to manually bend the bars made of them to very small radii, as required by the theory. Steel bars become very strong after heat-treatment. And, a wrench-bar of Archemedian proportions is not all that practical once you actually enter a typical academic lab. So, the bars refused to crack. So, to somehow finish my project, I ended up cracking those steel bars in a 3-point bend arrangement on a nice and sturdy universal testing machine. This was not the free form bending as the theory required. But it was the next best thing that could be attempted to induce those cracks--again, as the theory required. There were problems in using that kind of a rather stout machine too. The main problem was one of controlling its strain rate near the "point" of the "just onset" of cracking. You see, the machine would certainly break the bars. But the bars would crack very suddenly, because the cracks wouldn't first form only at the outer "fiber," as the simple solid mechanics theory predicts, but virtually anywhere near the three bending rolls... Sometimes right beneath one of three bending rolls. And that would make those already tiny cracks fully covered up by the roll, making them invisible until it was all too late. And then, very suddenly, a large crackling noise--and the testing machine itself would be in danger. (Only the servo-hydraulic controls of the machine, coupled with the nimbleness of the fingers of yours truly at the control panel, could somehow ensure that the machine itself didn't break down after that sudden bang!) The "simple" elasticity theory that I had mastered so well, I learnt, goes for a toss because the friction between the bar and the rolls is very significant. Friction, in turn, leads to both local tri-axial stresses and local stress concentrations near the point of their contact. Now, for stronger materials like steel, you need quite high levels of overall loading to actually break the bars. The much higher normal loading at the bend rolls straightway implies much higher frictional force over the locally deformed surface too. The local traction (loading) also becomes uneven if the bar surfaces are of commercial quality. Now, it was not known to me beforehand, but fact is, metal bars possess this tricky personality trait to mess up all the niceness of their surface once they are heat-treated by an undergraduate--all by himself. In such a case, some dimensional distortions and surface unevenness due to scaling simply refuse to go away. No lubrication during bending--whether oil or graphite--then actually helps. Grinding the bars to a smooth finish is not practical either--they are too long and slender. If you attempt grinding, the diameter becomes uneven, the surface acquires grinding marks, and bingo, you again have stress concentrations at all the unpredictable spots! All in all, the end result was that the data sets ended up going hither and tither. Remember, for each sample, there was this whole sequence of heat-treatment, then taking a small metallographic cut to see if the heat-treatment was right (say, without surface decarburization), then testing elongation in tension (with a scatter in those values too), and finally finding the bend-cracking radius. And then, all the above complications in bending to cracking! So, despite using a number of samples, the data sets showed more scatter than trend. They perhaps could be made to fit any theory one felt like. It took a certain purest form of courage to draw a comparative graph showing both the theoretically expected curve and the experimental data points, and also include this graph in the final project report. The theoretical curve was lying so far away from the data points, so disdainfully looking at the data I had so heroically obtained, that the two seemed to belong to two different pages--even two different project reports! The text of the project report manfully tried to maneavour around this particular difficulty, using all the skills of the obscure but hopeful writing that I could muster. It was the book-binder, and not the guide, who provided the necessary moral support at that stage of the project. (As per the official procedure followed at COEP, at the end of the project, the guide himself becomes a part of the evaluation team.) At that time, only he (I mean the book-binder) seemed a hundred and one percent sure, albeit in a peculiarly disinterested way, that I would definitely pass my project examination. The examiners didn't seem to have been all that impressed, much less swayed, by all the different things I had tried in the experimentation. More so, once they noticed that particular graph, and remained studiously fascinated by it for such a long time. Thereafter, they seemed to be only marginally satisfied that I had done some real work towards my project. One of them queried if the cracks I observed were not quench cracks... I knew the difference: the bend cracks would show a fresh surface but the quench cracks should show some thin layer of burning or oxidation products (after tempering), and soot and oil depositions (before tempering)... One could easily tell them apart. But somehow, it seemed, I couldn't convince the examiners. Thereafter, they seemed to begin to notice that I could easily explain the mathematical part of the theory. Quite well, in fact. Rather very well, and in much detail, they seemed to be concluding. And then, right in the middle of my explanation, I was asked by one of the examiners that I should tell them the theory, not teach it!... ...But, yes, on the day of the University results for the degree of Bachelor of Engineering, the book-binder's judgment turned out to have been an astute one. I got that degree in the Distinction class... My project didn't stand at the head of my class, to be sure. Yet, it was the good marks I got in the project work that just about pushed me into the range of the Distinction class. That was some 23 years back... If you ask me today, I would say that the single-most important reason the project work got such good marks at COEP was because I had refused to lie in my project. I had taken the harder way out. I kept the uncomfortable data right in the report. Sure, I wrote some text that concluded favorably towards the correlation. But the experimental results were a mixed case anyways, and so, the text could be given a benefit of doubt so long as the writing was an original and unedited one, coming straight from the undergraduate himself. But, far more important was that even if I was embarassed by the results at that time, I did make the experimental disagreement prominent by including that graph. In the project work and in the oral examination, I refused to disrespect my own experimentation in some sort of deference to a published paper. And then, I did proceed to attempt explaining the experimental results using every theoretical point I could find right then. By COEP tradition, they would want to distinguish that kind of an attempt by an undergraduate, despite any flaws there might be in his work... And the book-binder had some way of his own to know this! And come to think of it, I still haven't really understood how bending "moments" and shear "forces" and all really work, in a clear enough way. But before you start writing an adverse recommendation to my current Ph D committee, think of the following: Must straight sections remain straight--i.e. linear? How about the linearity of the distribution from the outermost to the innermost fibre? And then does that analysis (e.g. Timoshenko's) include the so-called couple stresses? Finally, just one more point: If you hit a thin and long glass road with a rubber-padded hammer axially--i.e. on one of its extreme ends--the rod breaks near the other end. Not at the end where you delievered the blow... How will dynamic/impact loading alter the "nonlinear" distribution you propose, assuming your analysis has managed to already include the aforementioned couple stresses?... You see, when it comes to not understanding stresses in a clear enough way, even in as simple situation as bending, we are all together in the same boat! |
|
Some Random Thoughts on Examinations, Marks, Ranks and Education
I never have studied regularly in a disciplined way, and never have believed in getting the highest possible marks at any examination in my entire life (except so as to just about cross the admission barriers when and as necessary). This is not an escape route to avoid confrontation with top rankers, nor an excuse for any failure. It is a personally grown, personal belief akin to a conviction. So much so that whatever good achievements I had in academics were actually only after almost a sort of struggle between wanting to understand (to the extent I did want to understand) vs. having to score (from a practical success viewpoint). I always have thought (right from the school days) that highest marks and ranks might make your case a bit more easily convincing to others. But they are poor substitutes to your own understanding. If I got especially good marks or ranks (say at the high-school scholarship examination or at GATE), it was not because of separate studies for those examinations. It was because I do like reading, and my extra reading took care of the preparation there was supposed to be, but in an informal way, and by itself. But, again, the reading never was done targetting the examination syllabus as such, except for may be for one or two days before the examination. I find it very hard to study for examinations. About educational ranks: Few people seem to share my beliefs, but here is how I see ranks. Take, for instance, my own case, say the fact that I was within 150 students out of about 50 thousand at the high-school scholarship. Most would say: wow! But there are many other perspectives by other, more achieving people, and then, some still other perspective I want to highlight. We will go in sequence. Take that rank, and ask the standard X or XII board rankers. They would say: "Yeah, fine. But why, this guy didn't make it to the board's merit list, did he? And say, even if he didn't study and so we have to give him benefit of doubt, even by his scholarship rank, he still is more or less just like one of us." Take it to the folks who took JEE successfully. They would say to us both: "What? But was it an all India rank? And then, those state board exams... Do they really test problem solving?" That is the normal reply. But here is another way to look at that "rank limited to just one state vs. all over India" argument. The central government of India runs another, all India level scholarship examination at high-school level, called National Talent Search Examination. It is quite prestigious in India. They are about 1000 in numbers. Surely it must be more competitive? But the fact of the matter is, over years and years, a clear trend emerges: About 1/4 to 1/3 of those all India level scholars come only from a single state: Maharashtra. Now, how do you level, say, a within 150 rank from Maharashtra by one scholarship examination with say, within 250 (or 330), again from Maharashtra? I mean the only hidden assumption here is that the sampling is statistically similar, which should be the case because both refer more or less to the same period, in the same education system, in the same state! You see, I got to know this fact about this trend of Maharashtra in the context of "more money to other states for more even development." But the same fact can also be used the way I did here! Real smart, heh? :) And then, take the same National Level argument with the folks who took GATE successfully. Now, they would say: "If those B Tech IITians are so good, why do so many of them (about 1/3) fail in GATE? Especially if the papers are being set by the same IITs? And how about the money the government spent on training them! And they still fail !!" These folks might even cite a case or two. For instance, there was this guy who got placed at about 250th place in boards but had cleared JEE. He went to an IIT, became the topper in aggregate terms over all years among all branches at that IIT. If he was so smart, and board exams so easy, why was he placed at number 250 at XII board exams? And remember, every year, there are these hundreds of board rankers, JEE rankers, GATE rankers. Throw in the people who go do an MS from an American university--another set of rankers. Throw in those who scored GMAT or CAT--yeat another set... And remember, there is a new set of rankers coming up every year. So, even much before the time you hit your middle age, there already are literally thousands of rankers out there!... How many people are you going to mentally score over? Sure, it's a fact that the premium on ranks gets placed all over the world. Sure, ranks are not bad! We will come to that in a moment. But for the time being, if you look at the arguments above, obviously there is a sort of bickering in there. Obviosly, we need a break there. I believe, we need that break without sacrificing the concern for achievement, distinctions, excellence--including actual efforts, purposeful work. Remove these last two and you would get into that usual socialist trap. I once heard this one in a grave voice (by a manager, not a labor union leader): "All people talk about Mensa and all, but you get Mensa membership if you are within 2%. At the rate of 1 in 50, that's a lot of people! Mensa calls them genius. So, how many geniuses are toiling only as workers in our factory? It's just that you never gave them a chance!" Then, he concluded: "That's the only differnce between you and them!"... What? No difference? None at all? Hello! I am not a Mensa member. Never attempted. Never will. But, what is the idea of that argument? Straightway give them management perks, heh? How about the fact that a high IQ guy could still be unmotivated to at all understand anything? Unwilling to take that effort to know? Or is all knowledge effortless? And what's the relation of Mensa problems to real problems, anyways? Outside the context of chess as a sort of sports activity, didn't you think anything at all of the story of the movie "Shatranj Ke Khiladi"?... So, that's the trap we need to avoid. My point is: all sorts of weird questions come up if you focus too narrowly on ranks in mere mental tasks, including those in education, without concern for reality. We need to ask: What are we doing here? Why? In what context? What is the purpose of education? I will not even attempt answering those questions. People do Ph Ds on questions like that. But I will just note down a few thoughts. More the element of statism (coercive government controls) in a mixed economy, the higher the emphasis the middle class places on achieving sheer ranks in examination. Obvious. With their own education, the middle class people think that going by the educational ranks is adhering to a just system of rewards. Lower classes can't grasp (for whatever reason--their circumstance or lethargy) anything like a system--statist, anarchist, or any system better than the two--any system as such. Elite classes (typically) couldn't care less for educational excellence--they already have what educational ranks would have given them. So, typically, it's the middle class that ends up ranting and raving about ranks in education. But, unfortunately, often forgetting in the process the sort of system that encourages that over-emphasis on ranks--thereby imparting a moral sanction to the element of statism. The notion of promotion by education ranks, without a thought to the actual performance in a task by an individual, suits only to a Platonist state. The middle class often forgets this. So, here are some speed-breakers to that idea: Any examination result is, at best, an estimate. In any examination, you get tested on what you know about the questions that were asked. Not about what other things from the same syllabus you did know. And that is apart from what things should (or should not) have been there in the syllabus in the first place! And then, in no examination you get tested on what questions you, in turn, could have asked--the questions that were proper (i.e. valid), and probed things that were unknown. And therefore, carried the promise of taking progress further... In the organized education system, Ph D is the only thing that comes close to this point. But even Ph D programs often do not realize that potential. And yet, I do not like that romantic idea about education which is so favorite with certain kind of educated Indians and the New Age folks from USA: People of all different ages are supposed to gather in open nature and try to learn from a guru. Even Silicon Valley entrepreneurs dream about such ideas.... But it is grossly inefficient a system! In principle, and in practice--whether large-scale realization is practicable or not is not the main issue. The main issue is, the system itself is extremely inefficient. Consider this: The teacher isn't bound by any particular topic because he is supposed to be free to teach whatever he likes, whenever he likes; the student is not bound by the preparation he is supposed to have before he comes there. And then, there obviously will be that mix-up out of a whole lot of students of different levels trying to learn simultanesouly. Unlike the rosy picture of personal attention that is often painted, it is the individual student who stands to suffer the most... To ask for a more personal interaction between teachers and students in the present system we all follow is not the same as what the above system is really like! Take it or leave it, but the fact is, the education system we have got today evolved to its definitive form precisely in those same times that witnessed unprecedented growth in culture, ideas, Man's mastery of nature--i.e. post-Rennaissance, in the Age of Reason. The system sure could be improved. But not in the way people often talk about--e.g., as in above paragraph. And so, for all their limitations and for all their flaws in implementation, I do believe that the sort of examinations we have, as a system, are practical... That they are good to have. Provided you know how to conduct them, and how to use their results. Essentially, the guiding rule for conducting examinations ought to be: Look for what a student knows--not what he doesn't know. The guiding rule for evaluating a student ought to be: Look for consistency. The consistency of performance across years and across a range of topics or activities. Examinations, and their results, are the means to an end, not an end in itself. Lose the context or the purpose, and you can only have smart but bickering people left.... |
| Future Updates: I plan to expand this Web site with sections on course-notes, books, philosophic thoughts, etc.... At some time in future though. This site in this form itself has taken many hours and there is a lot that I must postpone for now... |
|
Rights: All intellectual property rights (copyrights, trade-marks, service-marks, etc.) of their respective owners, as well as the moral rights of the concerned authors (including those who do not assert any!) are acknowledged. All such rights for my own work (including those for this site) are asserted. The linked Web sites are responsible for their contents.
For any comments about this site: tonebrush[AT]vsnl[DOT]net |
1: As a colleague saw me (2005)
2: As a colleague saw me (2005)
3: As another colleague saw me (2005)
 |
 |
 |
 |
 |
 |
 |
 |
 |
 |
 |
| http://www.JadhavResearch.info. This site does not use scripts or store cookies on your computer. Created in Notepad. Online since March, 2006. Last updated (v. 1.02.03): August 16, 2008. Best viewed in MS IE 6 at 1024 X 768 resolution. Copyright (c) Ajit R. Jadhav. All rights reserved. Administrative contact: tonebrush[AT]vsnl[DOT]net |