NivaarExam PrepOfficial exam papers ↗

22-Mec-B8 Engineering Materials · December 2017

Question 8 of 8: Three aluminium alloys for transport-aircraft floor beams

Nivaar worked solution (AI-drafted; not reviewed by a licensed engineer)

Notes on this paper

Paper format. National Exams, December 2017 — 16-Mec-B8 Engineering Materials. Three hours, open book; any non-communicating calculator permitted. Eight problems, all of equal value; any FIVE constitute a complete paper, so each problem is worth 20 marks. Candidates are urged to submit a clear statement of any assumptions made. All eight problems are solved below, because the set as a whole is the study resource.

Reference texts (22-Mec-B8 Engineering Materials).

  • Askeland & Wright, The Science and Engineering of Materials, 7th ed. — the primary syllabus text.
  • Callister & Rethwisch, Materials Science and Engineering: An Introduction, 10th ed.
  • Dieter, Mechanical Metallurgy, 3rd ed. — Considère's construction, plastic instability and volume constancy.
  • Shackelford, Introduction to Materials Science for Engineers, 8th ed. — ceramics, glasses and glass-ceramics.
  • Fontana, Corrosion Engineering, 3rd ed. — galvanic series, sacrificial protection and Faraday's law.
  • Ashby, Materials Selection in Mechanical Design, 5th ed. — selection criteria and material indices.
  • Polmear, Light Alloys, 5th ed. — aluminium tempers, Al–Li alloys and maraging steels.
  • Groover, Fundamentals of Modern Manufacturing, 7th ed. — shaping and consolidation of polymer-matrix composites.

Note on this sitting. Two of them are worth flagging because the fresh data changes the answer, not merely the arithmetic. Question 8 raises the copper content of the incumbent alloy, which makes it denser, and as a result both candidate alloys now meet the customer’s weight request so the selection turns on strength parity rather than on who clears the bar. Question 4 changes the aeroplane from a twelve-seat business jet to a hundred-seat commercial transport and the stretch from 25 to 30 per cent, which sharpens rather than softens the damage-tolerance objection. Every number below has been re-worked from this paper's own data.

Question 8: Three aluminium alloys for transport-aircraft floor beams (20 marks)

Question text not reproduced: the examination questions are © Engineers and Geoscientists BC. Open the official past paper (linked at the top of this page) to read the question, then follow the worked solution below.

Given. One set of floor beams, to be re-made in either of two candidate alloys to the same geometry:

Given data
QuantitySymbolValue
Alloy A (incumbent)—Al − 5.5 wt% Cu − 1.5 wt% Mg, 370 MPa
Alloy B (candidate)—Al − 4 wt% Li − 1 wt% Cu, 368 MPa
Alloy C (candidate)—Al − 3 wt% Li − 3 wt% Mg, 340 MPa
Weight of the existing floor beamsWA70 000 N
Weight reduction requestedΔWreq8000 N
Density of aluminiumρAl2700 kg/m3
Density of copperρCu8920 kg/m3
Density of magnesiumρMg1740 kg/m3
Density of lithiumρLi530 kg/m3

Find. (a) the density of each of the three alloys on the weighted-average rule the question prescribes, (b) the volume occupied by the floor beams, (c) the weight saving each candidate delivers at unchanged geometry and which one meets the 8000 N request, and (d) which of the three ranks highest on a strength-to-density selection index.

Check: the 8000 N reduction is read here as 8000 N to be taken out of the 70 000 N of floor beams, since the beams are the only weight the question states and the only structure being re-made. Gravity is taken as g = 9.81 m/s2 for part (b); as step 5 shows, g cancels out of the weight-saving answers, so nothing in (c) or (d) depends on it.

Approach. Compute each alloy’s density as the weight-fraction-weighted average the question prescribes; get the beam volume from the incumbent weight and density; note that re-making the same beams in a different alloy preserves the volume rather than the mass, so weight scales with the density ratio; then rank all three on the specific-strength index, which is a different question from the weight-saving one and does not have the same answer.

  1. Part (a) — write out the weight fractions. The alloying additions are quoted in weight per cent and aluminium makes up the balance: $$\text{A:}\quad w_{Cu} = 0.055,\quad w_{Mg} = 0.015,\quad w_{Al} = 0.93$$ $$\text{B:}\quad w_{Li} = 0.04,\quad w_{Cu} = 0.01,\quad w_{Al} = 0.95$$ $$\text{C:}\quad w_{Li} = 0.03,\quad w_{Mg} = 0.03,\quad w_{Al} = 0.94$$ Each set sums to unity, which is the check to make before going any further.
  2. Compute the three densities. Taking the weighted average of density that the question prescribes, $\rho = \sum w_i\rho_i$, $$\rho_A = 0.93(2700) + 0.055(8920) + 0.015(1740) = 2511 + 490.6 + 26.1$$ $$\rho_B = 0.95(2700) + 0.04(530) + 0.01(8920) = 2565 + 21.2 + 89.2$$ $$\rho_C = 0.94(2700) + 0.03(530) + 0.03(1740) = 2538 + 15.9 + 52.2$$ so that $$\boxed{\ \rho_A = 3027.7,\quad \rho_B = 2675.4,\quad \rho_C = 2606.1\ \text{kg/m}^{3}\ }$$ Alloy A is the heaviest because copper, at more than three times the density of aluminium, dominates its additions. Both candidates are lighter, and lithium is why: at 530 kg/m3 it is the lightest metallic element, so every weight per cent of it displaces aluminium with something five times lighter.
  3. Part (b) — find the volume of the floor beams. The beams weigh 70 000 N in alloy A, so their mass is $$m_A = \frac{W_A}{g} = \frac{70\,000}{9.81} = 7135.6\ \text{kg}$$ and dividing by the density of alloy A, $$V = \frac{m_A}{\rho_A} = \frac{7135.6}{3027.7}$$ $$\boxed{\ V = 2.3568\ \text{m}^{3}\ }$$ Equivalently $V = W_A/(\rho_A g)$ in one step. That volume is the quantity carried forward: the beams are re-made to the same drawing, so their geometry does not change.
  4. Part (c) — weigh the beams in each candidate alloy. Filling the same volume with a lighter alloy, $$W_i = \rho_i V g = W_A\frac{\rho_i}{\rho_A}$$ so $$W_B = 70\,000\times\frac{2675.4}{3027.7} = 61\,855\ \text{N}, \qquad W_C = 70\,000\times\frac{2606.1}{3027.7} = 60\,253\ \text{N}$$ Note that g cancels in the ratio, so neither answer depends on the value taken for it. This is the pivot of the whole question: a candidate who scales weights directly by the weight fractions, rather than through the common volume, gets a meaningless answer.
  5. Evaluate the two savings against the request. Subtracting, $$\Delta W_B = 70\,000 - 61855 = \boxed{8145\ \text{N}}, \qquad \Delta W_C = 70\,000 - 60253 = \boxed{9747\ \text{N}}$$ that is 11.64 % and 13.92 % of the beam weight respectively. Against the 8000 N the customer asked for, both candidates qualify: alloy B clears the target by 145 N, or 1.8 % of the request, and alloy C clears it by 1747 N, or 21.8 %.
Weight removed from the floor beams (N) — savings, not beam weights02 0004 0006 0008 00010 00012 000Alloy B (4Li–1Cu)8 145 NAlloy C (3Li–3Mg)9 747 Nrequested8 000 NBoth candidates clear the 8000 N request; Alloy B clears it by only 145 N.
The saving each candidate delivers, set against the saving the customer asked for. Plotting the three savings rather than the three beam weights is what makes the 145 N margin on alloy B visible at all — on a scale of 70 000 N it would be two pixels wide.

Selection for part (c). Both candidates meet the customer requirement as stated, so the choice is made on what else changes. Alloy B is the one to select. It matches the incumbent’s strength to within 2 MPa — 368 against 370, a difference of 0.5 % and well inside the scatter of any real allowable — so the beams can be re-made to the existing drawing with no re-analysis, no re-substantiation of the joints and no change to the loads the surrounding structure sees, and it still removes 8145 N. Alloy C removes 1602 N more, but it buys that with an 8.1 % drop in strength, from 370 to 340 MPa, and that has to be paid for somewhere. The size of the payment settles the matter. If any part of the beam is strength-critical at its present section, restoring the strength means scaling the sections by 370/340, which raises the volume to 2.5647 m3 and the weight to 65 569 N and leaves a saving of only 4431 N — a clear failure of the target. The same correction applied to alloy B, needing only 370/368, still returns 7809 N, within 2.4 % of the request. Alloy B is therefore the robust choice and alloy C the fragile one, and the honest answer records the one reservation attaching to B: its margin over the request is 145 N, which is smaller than the scatter of a typical weight statement, so the saving should be confirmed by a weight audit of the finished beams rather than taken on the density arithmetic alone.

  1. Part (d) — rank the three on strength to density. The selection index for a tension member of prescribed length and load, minimising mass, is the specific strength $\sigma/\rho$: $$\frac{\sigma}{\rho}\Big|_A = \frac{370}{3027.7} = 0.12220, \qquad \frac{\sigma}{\rho}\Big|_B = \frac{368}{2675.4} = 0.13755$$ $$\frac{\sigma}{\rho}\Big|_C = \frac{340}{2606.1} = 0.13046$$ in MPa per kg/m3, that is $$\boxed{\ 122.2,\ 137.5\ \text{and}\ 130.5\ \text{kN}\cdot\text{m/kg for A, B and C}\ }$$ so Alloy B is the best material on this criterion, 12.6 % above the incumbent against alloy C’s 6.8 %.
Strength-to-density index (kN·m/kg)115120125130135140Alloy A Al–5.5Cu–1.5Mg 370 MPa122.2Alloy B Al–4Li–1Cu 368 MPa137.6Alloy C Al–3Li–3Mg 340 MPa130.5Alloy B wins on specific strength, and it is also the alloy selected in part (c).Axis truncated at 115 so the 15.4 kN·m/kg spread between the three alloys is legible.
Strength-to-density ranking of the three alloys. Alloy B wins because it buys most of alloy C’s density reduction while giving up almost none of alloy A’s strength.

It is worth being explicit about why parts (c) and (d) agree here, because they need not. Part (c) is a fixed-geometry substitution, decided by density alone, and on that criterion alloy C is the lighter material and removes the most weight. Part (d) is a re-design question, decided by the index, and there strength and density trade against each other. The two answers coincide only because alloy C’s extra 2.6 % of density reduction over alloy B is bought with an 8.1 % loss of strength, whereas alloy B gives up 0.5 % of the strength for 11.6 % of the density — so C wins the raw weight contest but loses the efficiency one, and since B already satisfies the customer the raw contest no longer decides anything. Had the request been 9000 N instead of 8000 N, only alloy C would have qualified and the two parts would have returned different alloys. Recognising which regime the question sits in is the substance of the marks.

Results for the three-alloy comparison
QuantityAlloy AAlloy BAlloy C
Composition (balance Al)5.5Cu–1.5Mg4Li–1Cu3Li–3Mg
Strength σ (MPa)370368340
(a) Density ρ (kg/m3)3027.72675.42606.1
Weight of the beams (N)70 00061 85560 253
(c) Weight saving (N)—8 1459 747
Saving as % of the 8000 N request—101.8 %121.8 %
Saving after resizing to restore 370 MPa (N)—7 8094 431
(d) Specific strength σ/ρ (kN·m/kg)122.2137.5130.5
Verdict—Selected for (c) and best on σ/ρ; clears the request by 145 N at unchanged strengthClears the request by 1747 N, but at 8.1 % less strength
(b) Volume of the floor beams2.3568 m3 (unchanged by the substitution)

Check: the exam directs that the alloy density be taken as a simple weighted average of its constituents, ρ = Σwiρi, and that prescription is followed above. The rigorous volumetric mixture rule, 1/ρ = Σwi/ρi, gives ρA = 2783.7, ρB = 2334.0 and ρC = 2369.7 kg/m3 and hence savings of 11 308 N and 10 411 N. This sensitivity has to be checked on every paper of this type rather than assumed, because the two conventions can straddle the target. Here they do not: both candidates clear the 8000 N request under either rule, so the part (c) verdict is robust, and so is the part (d) ranking, alloy B leading on specific strength under either convention. What the volumetric rule does change is the ordering of the two candidates — it makes alloy B the lighter of the pair rather than alloy C, which only strengthens the recommendation above. The exam prescribes the weighted average, so that is the answer given. Two engineering caveats belong in any real report as well: the calculation assumes the beams are re-made to identical geometry, and it assumes the whole 8000 N is to be taken out of the floor beams alone.

Back to the paper →