An evening by a candle light

Yesterday, it rained in Pune. The newspaper reports today suggested a heavy downpour; 102 mm in 1.5 (or 3) hours. But in the parts of the city where I live, it couldn’t possibly have been 10 cm; by my rough estimates, it seemed like, may be, about 5–6 cm, at the most, about 7–8 cm.

But closer to the point at hand, as is usual in Pune after about 0.00001 cm of rainfall anywhere within a 100 km radius of the city any time within a preceding 1 hour, the electricity was gone.

It was evening, and the clouds had made it a bit darker than what it usually would be at this time of the evening—about 7:30 PM or so. And so, I first stumbled around a bit to find a candle, then realized that I had a cell phone, and so, using its light, I searched for and found the candle, and lit it—I mean the candle.

But as soon as I brought the lit candle to my room, the stronger winds in my room blew it out.

I then remembered a certain gift that my Idea Cellular subscription had generated for me on my birth-day this year (or on the day the last year). It was a scented candle jar. It’s a nice little jar of glass, say about 6–7 cm in diameter and height, with wax filled up to, say, half way through. The jar is thus a nice little thing, and the color of the wax is very beautiful: a faint yellow. It’s actually more or less the same yellow as what Idea Cellular’s brand carries. On a computer monitor (or on the plastic-coated printed paper they un-necessarily insert in the newspapers) it sometimes looks almost pierceingly bright. But a more or less the same hue, now used for the wax, had become, may be due to the dullness of the wax, a very pleasantly soft yellow shade to look at. And, the scent they used for the wax also is nice. (Being a male, both my sense of scent and vocabulary come to an end right at this level of resolution discrimination.) So, all in all, a pretty and neat gift, it was. And, precisely for that reason, I had not yet used it thus far. In the middle-class India, we naturally develop this habit of procrastinating when it comes to using up beautiful things like that.

However, in the faint light of my cell-phone, I now noticed that the level of the wax was such that if the candle were to be lit, its flame would rise up to a level that is only so slightly below the open top end of the jar. If a strong wind were to blow horizontally over its top, how would the flame behave? How fast would it flicker? Would it go out?

To preserve the battery of the cell-phone, I switched it off, and then decided to spend a little time trying hard to very carefully consider the decision: whether to use this candle in the jar right away or not. I could not succeed in it—I mean, in pursuing a nice, prolonged, fair, even-handed, two-sided, balanced, etc. kind of a vacillation about it. There were issues of engineering importance here, and about a minute had already been wasted by now. Thusly, I assured myself to my entire satisfaction that I had waited for a sufficiently long time in very carefully considering the decision. And so, I grabbed that candle-in-the-jar, lit it up, and brought it to my room.

The flame fluttered. But, no, not even the relatively strong winds would put it off.

The lights (I mean the electricity) came back a little while later. But, a couple of ideas for student projects had been born, in the meanwhile. I mean, projects at the master’s level; in particular, in mechanical engineering; and in more particular, in CFD. Here I am going to share these with you.

No, I no longer much care whether I divulge such ideas on the ‘net or not. If someone on the ‘net steals my ideas, he/his work would sure come up during my regular searches, and then I would make him feel ashamed. At least, I would bring out the theft to the notice of the research community. That would be enough for me.

Just one more point before we proceed. While reading the project descriptions, if you catch yourself thinking whether these are serious projects in mechanical engineering proper or not, do one exercise after you finish reading their descriptions: (i) note down the more advanced features of the equipment and experiments, and especially of the CFD simulations/software development, (ii) note down the ranges of the parameters involved in these experiments, perform some dimensional analysis, and then, (iii) think, do ‘net searches, discuss with other people, and thereby come up with at least three separate industrial applications for each.

If you cannot do the last bit, then the next course of action depends on who you are:

(A) If you are a student, then realize that you can always do a project under some other professor, but not with me.

(B) If you are a professor of mechanical engineering yourself, send your resume to the engineering colleges in the Savitribai Phule University of Pune, including COEP; they always very highly appreciate professors like you.

Ok. Let’s now move on to the project ideas themselves.

Project Idea 1:

(2–4 PG students)

Build a longish channel of rectangular cross section out of perspex. Place it horizontally on a table top. Divide the volume of the channel into a few (say 3–4) sections, by inserting vertical perspex walls mounted at the bottom/side-ways, and going up to some 80% of the height of the channel. Let water run through the channel, say from left to right, at various controllable speeds. Sprinkle some tracers in the water, shine some bright light on it, and make a video of the flow. Image process the individual video frames and thus experimentally determine the local flow velocities. Extend and refine the experiment a bit. For instance, think of using a converging channel, channels with uneven thicknesses and depths, inclined obstacle plates, water with suspended particles, or using precision sensors for local water flow, etc. Perform a CFD analysis of all such flows, starting with modeling just one compartment as the lid-driven (but not oscillatory) flow, and then build in the increasing complexity in a step-by-step manner. Some students may focus entirely on writing/adapting software, whether in C++ or Python. Compare the simulations with the experiments. For instance, can you accurately predict the pressure drops across the successive chambers? Can you accurately predict the amount of precipitation of the sediments in the various compartments of the channel? Their location profiles?

Project Idea 2:

(1–2 PG students)

Take a glass jar of roughly the same dimensions as the one mentioned above (some 8–10 cm dia, and roughly the same height). Take some white candle wax, melt it, pour the wax into the jar, introduce some die in it, and let the melt solidify. The solidified wax should continue to show some die streaks. (If the die thingie doesn’t work, sprinkle some bright shiny particles such as the one they use at the time of the Ganapati decorations.) Take a thin copper rod (say 5–10 mm dia) and using the ordinary funnel-holding stand they use in the XI–XII chemistry labs, hold the rod vertically such that its bottom tip pierces the top wax surface to a measured depth (say 5 mm). Now, use a resistance heater (e.g. a circuit similar to that used in the ordinary soldering iron) to heat the copper rod at some distance below the holder. Attach a few thermocouples over the length of the rod between the heater and the wax; also insert a few at various places in the wax, and a few on the outer surface of the jar. Switch on the power supply and continuously record the temperatures at various places, using data-logging cards on a computer. Also, make a video of the wax undergoing melting. During image-processing, the streaks (or the embedded shiny particles) are expected to help locate the molten front. Make careful measurements, and model this process using CFD techniques.

One of the students may focus just on writing a custom-built software to simulate the process (or adapting existing Open Source software). In case you think this is a very easy problem, the answer is both yes and no. Yes, because simulations of melting and flow were done in the computer graphics field more than a decade ago; include in your ‘net search strings such as: “melting” + “flowing” + “Stanford bunny.” The answer is also a no, because as a mechanical engineering software, accuracy is of primary importance to us; mere attractive or realistic looking graphics with fluid dynamical approximations wouldn’t be enough. And, the problem is, in a way, challenging: multiple phases, transient heat transfer with phase change, and a moving interface.

Project Idea 3:

(3–4 PG students)

Design and build a working machine to experimentally study how a toothpaste fills a mould cavity. The mould cavity should be made in some transparent material. In real experiments, for reasons of costs, it would be just a simple chalk-paste, not a bought out tooth-paste. Also design and build apparatus for, or otherwise conduct, suitable experiments to accurately characterize variables/parameters such as: viscosity, wall friction, surface tension, etc. The mould-filling machine itself should allow for variations in the geometry of the inlet, the ram velocity, etc. Make videos of the filling process. Model the process using CFD software and compare with experiment. Once again, the CFD simulation here is complicated: multi-phase flow, moving boundaries, and, more importantly, non-Newtonian fluid (and flow).

Additional possibility: use a shear-thinning material such as the tomato ketchup. Pursuade some one in the food-processing industry that this is worth-while project, and obtain a large supply of the ketchup for free. As to pursuading the faculty of engineering that this is not a research in agricultural engineering alone, leave that worry alone. As with the other projects here, if our results are good, we will publish them only in the mechanical engineering journal(s).

Some of the students may focus just on writing a custom-built software. The problem again is challenging because there is a multi-phase flow, with complex boundary conditions (wall-friction is variable and important, and surface tension is important), and also, there are both moving and coalescing boundaries.

Compare the simulation with the experiment.

* * * * *  * * * * *   * * * * *

Note: Each of the above projects could also use 2–4 undergraduate students. The mix is approximate; the depth of research and the intricacy and accuracy of experiments is always a variable: sometimes, conducting even simplest-looking experiments, or building a working software, can take students a long time to complete. It all depends, on a lot of factors—including the ability of the student and the amount of hard work he puts in.

* * * * *  * * * * *   * * * * *

A Song I Like:

[Yeah, sure, Mumbaikars, go ahead and say it: “angrez gaye to zaroor, lekin jaane se pahele, apani aulaad yahaan pe chhoD gaye.” Yes, the language is not a limitation at such times; Hindi, too, is perfectly acceptable at such times, even though, the same thing can of course be expressed also in “assal” Marathi: “ingraz ithun gele tar khare, paN (or better still, pan), jaataanna aapli aulaad maatr ithech soDoon gele.” Also laugh and/or heavily clap thereafter, or exchange friendly slaps with each other, in your typical style(s).]

(Western Classical)
Composer: Mozart
Work: The Piano Concerto no. 21 in C-major (k. 467)

[Esp. andante, esp. its opening part, e.g. as in here [^]. I have given this link because it appeared on the first page in a Google search. However, I now remember, I have also listened to some other recordings that I had liked even better (though I had not noted down their particulars back then). And, yes, I have in the past listened to the full concerto, too; it has… and how do I put it… an amazing variety of “texture” at different points. … Incidentally, though popular culture makes this tune all too familiar in the Western countries, and in fact these days also in India, the fact of the matter is, the first time I heard this piece of music was when I was in twenties (about three decades ago), and at that time, it had not yet appeared everywhere. In any case, I had not heard it in a bare-bones format any time earlier, until I listened to it on a cassette. And, I remember, it was a tune I had liked the very first time I had heard it. When it comes to the Western Classical music, that’s very rare for me. In fact, speaking off-hand, this tune is one of about only two tunes of Western Classical music that I had immediately liked right on my first listening. As far as the other tune is concerned, I don’t remember its name, nor do I any longer have the cassette where I had heard it, nor have I found it during my infrequent random searches of the Western classical music on the ‘net. Some day, I will make a recording of my humming of it, and put it on the ‘net just to be able to locate it. Anyway, enjoy this piece, and bye for now; guess I won’t bother much with this post any more.]


One thought on “An evening by a candle light

  1. Pingback: More on the project ideas. Also, a new CFD software benchmark-cum-shop-floor test. | Ajit Jadhav's Weblog

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.