I think grades should have two parts - an absolute correctness depth of knowledge grade and a percentile of the year. Thus "A-62" is "A" for correctness, while the "60" says they were at the 62% of that year.
----------------
The article I referenced said that British students were regrouped by ability, so they were graded on a relative scale against near-peers. Was that your experience?
Mine, from the 50's and 60's, was that we were grouped by age rather than ability and stayed together throughout, unless one fell or was pushed into Advanced Placement, which made the first year at college merely a review.
Mixing the children of mill workers with those of elite prep school teachers was somewhat troublesome but it fits our notion of equality, and the absolute grading scale gave anyone with ambition a fair chance.
As the bridge was being built to an application Standard, EN1090, and that was the basis of contract. All parties therefore took the responsible sensible step and employed "Code-talkers" / "Standards-talkers" to defend their interests. Nothing more than what the Standard obligates the employing organisation to do. Problem - fill a room with "Code-talkers" and they "defend" each other to a stop. They only object - they don't suggest. Paperwork was being generated everywhere, all giving a rosy picture - yet the bridge was not progressing.
I wrote a memoir
formatting link
"Memoir - the 3rd Bosphorus Bridge project, Turkey, 2015" It says the best that can be said what I did there.
Having been in the US education system up till 1982 when I graduated from high school then moved back to the UK I think the comparison of the education standards is not easy as I don't recall the US having any basic standardised tests other than SAT. In the UK at the time there were O levels, now GCSE, and A levels, my US high school diploma was considered equivalent to O levels so//I was effectively 2 years behind where I would have been if educated in the UK. The fact I normally took the more higher level courses in maths, chemistry, physics etc made no difference as there was no way to show what was covered to compare. I know at least one of my US teachers graded on a bell curve and I wasn't always popular for getting high marks as it skewed others downwards.
----------------- The SAT system had specialized achievement tests in math, physics, language, English composition (on which my gf scored 800) and so on.
My BS in Chemistry meant nothing when I enrolled in night school to try for an EE degree; I had to start with Algebra. I found the math courses much easier to follow when taught by instructors with a productive day job who used it as a practical tool, instead of as an art form.
My first and most important task, almost sole task, was to show that the concept within the maths is useful and good. In some practical way.
--------------------
Unfortunately the application may not be evident at first. I saw no use for the "imaginary" math of the square root of -1 when I learned it. If it doesn't really exist, what good is it in our world??
Much later when I got into digital radio I found that it was the preferred tool to model AC power for multiphase motors and signals for radio modulation.
formatting link
"It turns out that any form of modulation can be performed simply by varying the amplitude—only the amplitude—of I and Q signals, and then adding them together."
The amplitude of a sine or cosine wave is very easy to control at high speed with a computer. Likewise it's easy to measure in the receiver. The tech benefited from the rapid advances in digital storage oscilloscopes which are similar.
This means that the hardware of a digital radio circuit can become anything the attached computer tells it to be. In principle the same circuit could function as an inverter welder, AC power generator, a stereo, a shortwave, AM or FM radio or a TV. Although they are meant for 1~2GHz, cell phone receivers can double as 100MHz FM radios by using the ear bud cable as the antenna.
The relevant math trick is to place the Q component in the imaginary realm so an equation can contain both I and Q, without them interacting until desired to. Graphically they are orthogonal, at a right angle to each other. To avoid confusion the square root of -1 is called 'j' in electronics. Your brilliant WW2 Huff-Duff U-boot location system used this scheme, as does the digital radio in cell phones. For motors I describes the real power that you pay for, Q the apparent power temporarily stored in capacitance and inductance.
It has to simply be seen to work. See that and apply them. You are going to start noticing things and it may draw you to the theory.
I think it's helpful to say you don't have to "understand" maths. You must start off simply "cranking" it - then you have a basis, a foundation, where you might study and get a bit knowledgable.
My head doesn't operate well that way. I'm pretty good at visualizing and animating things in 3 dimension but when I can't I have trouble relating and filing away their formulas. For example when I first learned calculus in college I could memorize some of the differential and integral rules and formulas and look up the rest. I could "see" the functions of powers and roots and how to set up the equations to solve the problem but I couldn't for trigonometry. I got the right answers for trig and log problems but didn't really understand why, so the lesson didn't all make it into my long term memory. I still have to pause to reconstruct in Pythagorean terms why SIN^2 X + COS^2 X = 1
The second time I took Calc in night school the instructor spent two weeks explaining the underlying limit process and everything finally made sense. At work I was solving calculus problems of inductance and capacitance in my head, I = C dV/dT and V = L dI/dT.
Tonight's mechanical problem is to dream up, bend and attach a wire support leg for a power supply that folds either straight out or back flush. It's for this, and I neglected to consider that small stick-on feet wouldn't let enough air out the bottom vents.
formatting link
took a little while to realize that the pivot arms should rest at 45 degrees to the case bottom either way. I might be getting old.
You mentioned uncertainty about MIG arc voltage. The standard way to compensate for high current cable voltage drops is to run thin wires from the cable ends back to the source, called a "Kelvin" or 4-wire connection. It could be taped to the outside if it can't be fished through internally. At the power source housing where the wires can be tied down they go into a voltmeter whose power supply floats relative to the welding power, to allow for the different "ground" voltages. I use cheap/free obsolete 5V cell phone chargers to power floating meters.
material has the Elastic Modulus of steel at 210GPa. But that spectrum of Poisson's Ratios - just to see what happens. And what you see happens...
The physical object modelled is this
formatting link
"Tensile-test rig for beam-configuration fillet-weld samples"
formatting link
"Fillet welds tensile tested in beam test"
Why that flat area on the otherwise downsloping graph?
The range of deflections isn't that great, it needs to be said. So it's not "world-changing".
BTW that range of Poisson's Ratios where nothing changes from about
0.3 to 0.43 about matches the Poisson's Ratio of most engineering alloys...
?!?!?!
-------------------------
Could the flat be from a dip on the stress/strain curve between the linear and plastic regions? I haven't measured the dip but I can definitely feel it, as when straightening copper wire by stretching it.
Lacking a pressure gauge or easy way to add one, (the quick connects and tee I suggested previously is lab-only fragile) you could make an adapter to use a beam-type torque wrench as the pump handle, without other modification to someone else's pump. It should be at least a repeatable and recordable numerical indicator and could be calibrated. The adapter could be a crows-foot wrench or impact adapter welded to tubing or a turned stub of rod stock.
The adapter could be a crows-foot wrench or impact adapter welded to tubing or a turned stub of rod stock.
---------------------
The square hole to round peg adapter I proposed might be more adaptable with two slugs of round stock fish-mouthed to the socket on opposite sides, so it can be chucked and turned in a lathe to fit your now and future pumps.
That stunt of yours with the 3mm ball end mill reminded me of Richard Feynman dunking the Shuttle O ring in his drinking glass of ice water. So simple yet so effective.
Sorry - those results are in the parallel universe of numbers and arithmetic relationships. That's the Finite Element Analysis output. A linear-elastic FEA has not yield (and implicitly infinite strength).
You are thinking of the physical phenomenon which gives the Luders Bands on the bottom of a spray-can (for cheapness they don't "temper-roll" as they do for cars just before pressing so they don't have Luders Bands).
I think this is a case of "Texas sharpshooting". (apologies in advance to Texans). That is - you fire a magazine-full / a chamber-full of rounds at the barn wall and draw a ring around the tightest cluster. I perceived a pattern in the rendered "deflection in 'y'" output images, and went looking for it. Which means there's already a non-random contribution.
The adapter could be a crows-foot wrench or impact adapter welded to tubing or a turned stub of rod stock.
---------------------
The square hole to round peg adapter I proposed might be more adaptable with two slugs of round stock fish-mouthed to the socket on opposite sides, so it can be chucked and turned in a lathe to fit your now and future pumps.
-----------------
The beam torque wrench I was thinking of when I wrote that ends in a 1-1/4" cylinder parallel to and centered around the beam that would easily adapt in-line to the pump handle socket but I didn't see an image like it on Google. I think the usual type with the end cylinder at a right angle would work if the square socket it plugs into is offset on a bar welded across the end of the rod/tube that fits the pump socket, to center the beam so the handle doesn't twist.
I think this is a case of "Texas sharpshooting". (apologies in advance to Texans). That is - you fire a magazine-full / a chamber-full of rounds at the barn wall and draw a ring around the tightest cluster.
----------------------
That sighting-in method saves cease-fire interruptions and walks downrange to change targets. After you have minimized the group size in fresh areas of the target you can adjust the sights onto the bullseye.
Sorry - those results are in the parallel universe of numbers and arithmetic relationships. That's the Finite Element Analysis output. A linear-elastic FEA has not yield (and implicitly infinite strength).
You are thinking of the physical phenomenon which gives the Luders Bands on the bottom of a spray-can (for cheapness they don't "temper-roll" as they do for cars just before pressing so they don't have Luders Bands).
-----------------
Being a lab tech I tend to think of graphs as measurement results instead of simulations.
Jim - I lost the response where I posted a graph which was of output from the Finite Element program and you responded.
I plotted the "maximum deflection in 'y' (vertical)" against Poisson's Ratio.
formatting link
Mistaken for a physical data plot.
Thing is with all computer-numerical solutions : they don't reveal proportionality. With say beam calculations - you can look at the answer so far and work-out about what you have to do to get a value you need - a load-bearing capacity or something like that. You can do a lot of fine-tuning. I realised that say you could be making a gantry for hoisting and you could first-and-foremost make sure it would bend all the way to the floor without buckling - never have a drop in load-bearing capacity all the way to a cartoon-like "Ooops!" end. A "graceful failure". How you got that with a smaller cross-section square-hollow beam with thicker wall then meant that while still good, you'd see overload as a visible *elastic" bend of the top beam. Etc., etc., etc.
You don't get that with a computer-numerical solution.
So with a computer-numerical solution for a system of underlying equations, you have to run the solution multiple times and do as I did
- plot the trend.
However - the computer-numerical solution - in this case Finite Element Analysis modelling - can calculate the answer for "impossibly" complicated shapes for which there is zero possibility of an "analytical" solution - an "on-paper arithmetic solution".
So you do horses-for-courses. With structures, I think you might design overall with the long beams with (Euler-Bernoulli) beam - the arithmetic calculations with proportionalities all in-view. Then use FEA to design the end connections *knowing already* what force and moments (twists) they must take.
Hope this is suitable return for effort where I have sometimes not been clear and mislead effort.
... I realised that say you could be making a gantry for hoisting and you could first-and-foremost make sure it would bend all the way to the floor without buckling - never have a drop in load-bearing capacity all the way to a cartoon-like "Ooops!" end. A "graceful failure".
---------------------
I assembled the gantry with what I calculated was more than adequate center splice strength (unknown scrap steel quality) and then proof-tested it with a crane scale, to see if the deflection matched calculations and the parts I made to very close tolerances would bind when disassembled. As hoped, torsional rigidity of the 4" channels was the weakest link. They were somewhat bowed from previous service and the kinks were too difficult to straighten completely with my limited equipment.
It was marginally adequate with a 2000 lb central load and only end supports
16' apart which is as expected. I had planned for a central support and worked out a procedure to lower the load at the center and walk the two diagonal legs over it (a 12' long log) which kept the beam stresses well below the proof test.
... I realised that say you could be making a gantry for hoisting and you could first-and-foremost make sure it would bend all the way to the floor without buckling - never have a drop in load-bearing capacity all the way to a cartoon-like "Ooops!" end. A "graceful failure".
---------------------
I assembled the gantry with what I calculated was more than adequate center splice strength (unknown scrap steel quality) and then proof-tested it with a crane scale, to see if the deflection matched calculations and the parts I made to very close tolerances would bind when disassembled. As hoped, torsional rigidity of the 4" channels was the weakest link. They were somewhat bowed from previous service and the kinks were too difficult to straighten completely with my limited equipment.
It was marginally adequate with a 2000 lb central load and only end supports
16' apart which is as expected. I had planned for a central support and worked out a procedure to lower the load at the center and walk the two diagonal legs over it (a 12' long log) which kept the beam stresses well below the proof test.
------------------
The design process went backwards; I bought likely-looking used steel as I found it and then tried to make the best use of it. When designing custom equipment I found that the right materials to do it one way usually sufficed to make it another when the customer revised the spec. I could order material based on early guesses pretty safely. That way our machine was ready to test the module when it went into production, and we kept winning contracts.
One example that I thought was funny was having to add a smoke detector and emergency shut-down to the "burn-in" station for Chrysler Lean-Burn engine controllers. Let's say the design was a bit deficient under certain circumstances, from being on too tight of a timetable. I also burned up a prototype GM fuel injection controller by stressing it to its specified limits, which their lab equipment couldn't reach.
The sensors added to enable closed-loop combustion control were ones I was familiar with from Chemistry.
There is a red-blooded can-do attitude in America, as I have met it. That is the image of America too.
Be a bit forgiving of these academics. No-one talks with them seriously. The working world tends to be something of a miasma of uselessness anyway compared to what the potential could be, so for anything good to seep in for budding academics to tune into isn't that probable anyway.
I finished my Doctorate having solved not one but two previous unknowns about why the newer High-Strength Low-Alloy Thermo-Mechanically Controlled-Processed steels, then only from Germany and Japan, have properties so advantageous beyond comprehension compared to "classic" C-Mn steels
high weldability - most of the time weld with zero precautions (preheat, etc.)
highly resistant to "sour" (acidic with hydrogen sulphide) crude oils such at a pipeline can carry quite sour oils without problem
and titled my Doctoral thesis "Hydrogen distribution and redistribution in the weld zone of constructional steels" when the correct/reasonable/informative title would have been "Hydrogen distribution and redistribution in the weld zone of *structural* steels"
-or "structural and pipeline steels"
because no-one was talking with me.
Most would respect that what they know in detail is very little in the big picture and would respect you a lot...
PolyTech Forum website is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.