3
$\begingroup$

Quadrilateral $ABCD$ , $\angle ABD = 17^{\circ}, \angle DBC = 34^{\circ}, \angle ACB = 43^{\circ}, \angle ADB = 13^{\circ}$, Find $\angle BDC$.enter image description here

  • 1
    Please add your effort, and what you've tried.2017-02-28
  • 2
    Please upload a figure or explain how you labeled the vertices.2017-02-28
  • 0
    $\angle ACB = 43°$ ?2017-02-28
  • 0
    @S.C.B, I still have no idea on this problem.2017-02-28
  • 1
    @ Saksham, using Geogebra, the answer is $39^{\circ}$.2017-02-28
  • 0
    Only two angles were left when I tried this using elementary geometry...Nonetheless this can be easily done if you use a calculator and Law Of sines...The correct answer is definitely **39 $^{\circ}$**2017-02-28
  • 0
    @user35508, all calculating equipments are not allowed.2017-02-28
  • 2
    Well..If you could find the following expression without your calculator , you will have your answer.. $$\cot^{-1}\left(\frac{2\tan(64^{\circ})}{2\cos(26^{\circ})+1}-\tan13^{\circ}\right)$$2017-02-28

3 Answers 3

7

Nice problem!

Let $X$ be a point symmetric to $A$ with respect to $BD$. Let $Y$ be a point symmetric to $A$ with respect to $BX$. Let $Z$ be a point symmetric to $A$ with respect to $DX$.

Then $DA=DX=DZ$, $BA=BX=BY$, and $AX=XY=XZ$. Angle chasing gives $$\angle ZXY=360^\circ - \angle AXZ - \angle YXA = 360^\circ - 2\angle AXD - 2\angle BXA = \angle XDA + \angle ABX = 2\angle BDA + 2 \angle ABD = 2\cdot 13^\circ+2\cdot 17^\circ = 60^\circ$$ which along with $XZ=XY$ implies that $XYZ$ is equilateral. Thus $$\angle BZA = \angle BZX + \angle XZA = 30^\circ + 13^\circ = 43^\circ.$$ We also have $$\angle DBZ = \angle DBX + \angle XBZ = 2\angle ABD = 34^\circ.$$ This means that $Z=C$. Therefore $$\angle CDB = \angle ZDB = 3\angle BDA = 39^\circ.$$


Below is a trigonometric solution. Let $\angle CDB = x$. We use Snellius' theorem thrice: \begin{align} \frac{AC}{DC} & = \frac{\sin(13^\circ + x)}{\sin 64^\circ}, \\ \frac{DC}{CB} & = \frac{\sin 34^\circ}{\sin x}, \\ \frac{CB}{AC} & = \frac{\sin 86^\circ}{\sin 51^\circ}. \end{align} Multiplying yields $$1=\frac{\sin(13^\circ + x)\sin 34^\circ \sin 86^\circ}{\sin 64^\circ \sin x \sin 51^\circ}$$ so $$\sin(13^\circ + x)\sin 34^\circ \sin 86^\circ = \sin 64^\circ \sin x \sin 51^\circ.$$ Using $2\sin A \sin B = \cos(A-B) - \cos(A+B)$ twice we get $$\sin(13^\circ + x)\left(\cos 52^\circ - \cos 120^\circ \right) = \sin x \left(\cos 13^\circ + \cos 65^\circ\right).$$

Since $\cos 120^\circ = -\frac 12$, we have $$\sin(13^\circ + x) \cos 52^\circ + \frac 12 \sin(13^\circ + x) = \sin x \cos 13^\circ + \sin x \cos 65^\circ.$$

Multiplying by two and using $2\sin A \cos B = \sin(A+B) + \sin(A-B)$ we infer

$$\sin(65^\circ + x) + \sin(x-39^\circ) + \sin(13^\circ + x) = \sin(x+13^\circ) + \sin(x - 13^\circ) + \sin(x+65^\circ) + \sin(x-65^\circ).$$

Therefore $$\sin(x - 39^\circ) = \sin(x-13^\circ) + \sin(x-65^\circ).$$

We use now $\sin A + \sin B = 2 \sin \frac{A+B}2 \cos\frac{A-B}2$:

$$\sin(x-39^\circ) = 2\sin(x-39^\circ)\cos 26^\circ$$

or

$$\sin(x-39^\circ)(1-2\cos 26^\circ)=0.$$

Since $\cos 26^\circ \neq \frac 12$, we have $\sin(x-39^\circ)=0$ and so $x=39^\circ$.

  • 0
    How did you see this sir?2017-03-01
  • 1
    @N.S.JOHN I played with geogebra for a while, noticed this crucial equilateral triangle and then proved that it is indeed equilateral which led me to the answer.2017-03-01
  • 0
    Nice solution ! In case that we cannot use Geogebra in the exam, how can we know what auxiliary lines will help. Do trial and error, right ?2017-03-01
  • 1
    In the exam I would suggest using trigonometry. Usually it's not so easy to find purely synthetic solution.2017-03-01
  • 0
    Thank you very much. I got an idea from your synthetic solution. Could you (or someone) please solve this problem using trigonometry with time limit, 20 minutes ?2017-03-02
  • 1
    @carat I have added a trigonometric solution.2017-03-02
  • 0
    I understand now. Thank you all.2017-03-02
1

We can consider in $\mathbb{C}$ : $A=0$, $B=1$

Denote $AB=a=1, \,AC=c,\, AD=d$. Then:

  • $\frac{a}{\sin 43^{\circ}}=\frac{c}{\sin 51^{\circ}}$, so $c=\frac{\sin 51^{\circ}}{\sin 43^{\circ}}$
  • $C=B\cdot c(\cos{86^{\circ}}+i\sin 86^{\circ}) = \cos{86^{\circ}}\frac{\sin 51^{\circ}}{\sin 43^{\circ}} + i \sin 86^{\circ}\frac{\sin 51^{\circ}}{\sin 43^{\circ}}$
  • $\frac{a}{\sin 13^{\circ}}=\frac{d}{\sin 17^{\circ}}$, so $d=\frac{\sin 17^{\circ}}{\sin 13^{\circ}}$
  • $D=(-1)\cdot d(\cos(-30^{\circ})+i \sin(-30^{\circ}))=-\frac{\sqrt{3}}{2} \frac{\sin 17^{\circ}}{\sin 13^{\circ}} +\frac{i}{2}\frac{\sin 17^{\circ}}{\sin 13^{\circ}}$

Now we know that $\angle DBC = \arg(C-D)+\arg(B-D)$. $\arg(B-D)=17^{\circ}$, so we need only to compute $\alpha = \arg(C-D)$:

$\tan \alpha = \frac{\mathfrak{Im }(C-D)}{\mathfrak{Re }(C-D)}=\frac{\sin 86^{\circ}\frac{\sin 51^{\circ}}{\sin 43^{\circ}}-\frac{1}{2}\frac{\sin 17^{\circ}}{\sin 13^{\circ}}}{\cos{86^{\circ}}\frac{\sin 51^{\circ}}{\sin 43^{\circ}}+\frac{\sqrt{3}}{2} \frac{\sin 17^{\circ}}{\sin 13^{\circ}}}$

Well, now that looks a little creepy, but it wil surely get less complicated, if we will use some trigonometrical formulas (sine and cosine of sum, sine and cosine of doubled angle... and so on). See, that:

  • $86^{\circ}=2\cdot 43^{\circ}$
  • $43^{\circ}=30^{\circ}+13^{\circ}$
  • $17^{\circ}=30^{\circ}-13^{\circ}$
  • $51^{\circ}=90^{\circ}-3\cdot 13^{\circ}$

Unfortunately I don't have now enough time to deal with this without calculator, but in some days I'll manage some time to do it.

Edit:

Wolfram Alpha [1] tells, that $\alpha=22^{\circ}$, so we obtain $\angle DBC = 39^{\circ}$.

[1] https://www.wolframalpha.com/input/?i=ArcTan((Sin(86Deg)Sin(51Deg)%2F(Sin(43+Deg))+-+1%2F2+Sin(17Deg)%2F(Sin(13+Deg)))%2F(Cos(86Deg)Sin(51Deg)%2F(Sin(43+Deg))+%2B+Sqrt(3)%2F2+Sin(17Deg)%2F(Sin(13+Deg))))

-1

See the image for a correct solution.

enter image description here

The original and, incorrect answer follows.

Don't read what follows, it's wrong.

The problem you presented has no unique solution. This is easy to prove.

First, only one angle of the triangle BDC is specified, therefore the angles BCD and CDB are free to change. In other words, this is a triangle with 2 unknown angles. Knowing one angle is not sufficient to determine the value of the other two, additional information is needed.

The other triangles are totally superfluous. The location of point A is irrelevant as it does not affect the shape of the triangle BDC. The two angles in the BDC triangle remain unknown ** and unaffected ** by any movement of point A.

Dynamic geometry software, such as GeoGebra can tell you the value of the angle because it is internally using a lot more information than the one presented in your figure. For instance, it knows the equations of every line, from that, calculating any angle is trivial.

If you wish to convince yourself, in your figure delete the segments AB and AD (don't delete the segment AC, keep it for visualization purposes.) After doing that you can move AC whereever you want, which will change the angle measure of 43 degrees but, the problem remains identical, one triangle with 2 unknowns. Knowing a portion (43 in your figure) of one unknown is useless.

If more information was given, such as, DA = DC then that would be sufficient to determine the value of the angles.

The problem you've presented is equivalent to presenting one triangle with one known angle, no other information and asking for the value of the other 2 angles. The answer is: there is no unique solution. It is undetermined.

Hope that helps.

  • 0
    Your reasoning is wrong. To show this - lest fix points $A$ and $B$. Using the rules formed in question, you can construct two triangles - $ABC$ and $ABD$ and this construction is unique (with a precision to the symmetry). The angle $BCD$ is then unique.2017-03-01
  • 0
    You are correct. My original reasoning is wrong. The fact that angle ACB is 43 degrees forces the triangle BDC to be rigid. It also forces the triangle DAC to be isoceles, that is, segment DA = DC. because of that the exterior angle at point A, can easily be calculated to be 116 degrees. The exterior angle at point C must also equal that and it is also equal angle BDC + 77 degrees, yielding BDC = 39 degrees. Angle ACD can initially be calculated as 180 - (77 + 39) giving ACD = 64 degrees. The angles thus calculated can be confirmed by showing that their sum is 103 degrees as it should be.2017-03-01
  • 0
    Your reasoning in the attached picture is still incorrect. You assumed without proof that $DA=DC$.2017-03-01
  • 0
    In the image, the last step, showing that the sum of angle X + angle Y is 103 degrees confirms that the triangle is isoceles as assumed. If it weren't, the sum would not yield the value determined by the exterior angle. Note that the value of 103 degrees is obtained without making any assumption. It is in your answer where you claim DA = DZ (which is the same as saying DA = DC, since you assume Z = C upfront) and do not provide any confirmation of that fact by alternate means not dependent on the implicit assumption your answer initially makes. Your result is correct but left without proof.2017-03-01
  • 0
    You are wrong. In the last step you only verify that if $DA=DC$ then the sum $X + Y$ is 103 degrees. But this was already known without the assumption $DA=DC$! It does not prove that $DA=DC$.2017-03-01
  • 0
    No, the last step verifies that Euclid's Book I, proposition 32 (an exterior angle equals the sum of the two interior and opposite angles) is not violated by the assumption that the triangle is isoceles. The value of 103 degrees is obtained _without_ making any assumptions as to the type of triangle and shown to be honored by the assumption that the triangle is isoceles, thereby proving that it is (if it weren't the sum would not equal 103). Your solution fails to prove your assumption that DA = DZ by other means. Instead of hiding the assumption, make it explicit and verify it.2017-03-01
  • 0
    I'm afraid you have problems with elementary logic. Let me phrase it another way. The assumption that $DA=DC$ can be rephrased as $\angle DAC = \angle ACD$. So, you assumed that $\angle DAC = \angle ACD$ and then you calculated the values of $\angle X$ and $\angle Y$ under this assumption. At this point the only thing you have to show is the equality $\angle DAC = \angle ACD$. Instead of proving this, you only checked that this does not violate the equality $\angle X + \angle Y = 103^\circ$.2017-03-01
  • 0
    By this logic I can prove that $\angle X = 74^\circ$. We proceed as follows. We calculate that $\angle PAD = 64^\circ$. Then we assume that $\angle PAD = 2 \angle PCD$, this will be verified later. Then $\angle PCD = \frac 12 \cdot 64^\circ = 32^\circ$. Since $103^\circ = \angle BPC = \angle X + \angle Y = \angle X + 32^\circ$, we have $\angle X = 71^\circ$. Now, the assumption that $\angle PAD = 2 \angle PCD$ does not violate the equality $\angle X + \angle Y = 103^\circ$. Indeed, $\angle X + \angle Y = 71^\circ + 32^\circ = 103^\circ$. Therefore $\angle PAD = 2 \angle PCD$ must be true.2017-03-01
  • 0
    That's nonsense. PAD is determined to be 64 degrees. If you assume that PAD = 2 PCD then PCD =32 which would clearly violate quite a few theorems and the exterior angle (the one from the fact that the triangle is isoceles, that is, 116 degrees) would no longer be equal to the sum of the opposite remote angles.2017-03-01
  • 0
    Please point out precisely where I would get a contradiction by assuming that $\angle PAD = 2 \angle PCD$.2017-03-01
  • 0
    Here is an obvious one (which you should have caught): angle JCA + angle Y would no longer equal 180 degrees as required. Instead, you'd end up with 116 + 32 (your value of angle Y) = 148 instead of 180. In your answer you didn't verify your assumption that DA = DZ. Since the assumption is true, your results are correct but, assumptions must be proved by showing that their results are equal to those of known theorems. Your answer fails to do that.2017-03-01
  • 0
    How would you know that $\angle JCA = 116^\circ$ (under the assumption $\angle PAD = 2 \angle PCD$)? If you have some questions about my solution above please comment the post with the solution. Here we are discussing why yours is incorrect.2017-03-01
  • 0
    I made no assumption that angle PAD = 2 PCD. PAD and all other angles are determined by the fact that the interior angles of a triangle sum to 180 degrees. GAC equals (DAF + PAB), that's a fact. Then assuming that the triangle is isoceles leads to GAC being assumed equal to JCA. Once the unknown angles are calculated, the assumption is validated by the fact that APD equals angle X + Y as it should. If the assumption had been invalid, the calculated values would have been incorrect and would have failed the test (for instance JCA + Y no longer adding up to 180 as in your example).2017-03-01
  • 0
    I know you do not assume that PAD = 2 PCD. My intention was to show you blatantly incorrect argument that is analogous to yours. This was supposed to help you find a logic flaw in your proof. You claimed in the last comment that "if the assumption had been invalid, the calculated values would have been incorrect and would have failed the test". This is not true. Wrong assumptions can lead to correct answers. That's why your argument is incorrect.2017-03-01
  • 0
    Yes, wrong assumptions can lead to correct answers. The assumption is correct and proven to be correct very much unlike in your answer where your results stemming from the same assumption is not validated by any theorem. Lastly, I'm done discussing this.2017-03-01
  • 1
    The assumption is correct, but you didn't prove that it is indeed correct. This is huge gap making your argument incomplete. If you have some questions about my answer please leave a comment under it, not here. I'll be more than happy to make it clearer.2017-03-01
  • 0
    @MathAddict - why do you assume, that $DAC$ is isosceles? This assumption seems to be groundless.2017-03-02
  • 0
    @Jaroslaw, there are two reasons for it. The first and most important is that it is actually isosceles (but that must be eventually proved.) The second one is that _if_ the triangle is isosceles then knowing the exterior angle (116) and the angle ACB (43) makes the calculation of X and Y very simple. After that, it is crucial to demonstrate that the assumption is true which is done by verifying the result using the exterior angle APD (103) which does not depend on any assumptions. There is nothing wrong with making an assumption to solve a problem as long as the assumption is proven true.2017-03-02
  • 1
    @MathAddict Note, that you've pointed in your solution "if $DAC$ is isosceles, then $X+Y=103$". What you need to show that assumption is true is not $X+Y=103$, but "If $X+Y\neq 103$, then $DAC$ is not isosceles".2017-03-02
  • 0
    No. What I pointed out in the solution is that if DAC is isosceles then JCA equals GAC. Making that assumption, the angles X and Y are calculated. The assumption is proven true because it leads to results that are the same as those that would be obtained without making the assumption. Specifically that the angles X + Y sum to 103 degrees which is a fact that does not depend in any way on the assumption. If the assumption had been false, the values of X and Y would *not* have added up to 103 as required by other theorems.2017-03-02
  • 1
    @MathAddict So you are saying, that "By assuming P, we've obtained Q and Q is true, so P also have to be true" ?2017-03-02
  • 0
    This is exactly what @MathAddict means. This guy has serious problems with elementary logic.2017-03-02
  • 0
    No, that logic does not work. P implies Q does not necessarily mean that Q implies P. Both implications must be independently proved and in the case of isosceles triangles, they have been. Isosceles implies two exterior angles being equal AND, the converse, two exterior angles being equal implies an isosceles triangle. If you prefer, in the problem presented, assume that the triangle is NOT isosceles and you will run into a contradiction, specifically, that the triangle ACB is not constructible with the angles shown. The angle at C could not be 43 degrees and the angle DBC could not be 34.2017-03-02
  • 0
    Could you please present **how** one can get a contradiction assuming that the triangle is not isosceles?2017-03-02
  • 0
    Very simple. Try to build the triangle presented in this question with it not being isosceles. Pick any angles you want for X and Y (that is, other than creating an isosceles triangle), and you'll find that you cannot construct a non isosceles triangle with the angles as specified in this question. That way, you can convince yourself. You use GeoGebra, try it with that.2017-03-02
  • 1
    Checking something in Geogebra is not a proof.2017-03-02
  • 0
    That much is true but, just maybe it would help see what you are not seeing. I am done discussing this.2017-03-02
  • 0
    I still fail to derive a contradiction assuming that $DA \neq DC$. Please show us how to get a contradiction. If you do that, I will admit that your proof is correct and the discussion will be finished.2017-03-02
  • 0
    You'll see it very easily in GeoGebra if you attempt to construct the example using any non isosceles triangle. Try with angles Y = 65 and X = 38, just 1 degree off what they should be. You'll find, that while you can force some angles to have the values they should have, inevitably you run into contradictions. That way, you get to see the problems. That construction *depends* on the triangle ADC being isosceles. I suggest you build it starting at point D and working your way towards point B. When you get to B, you'll find it impossible to finish the construction with the given angles.2017-03-02
  • 0
    I'll repeat: checking something in Geogebra is not a proof.2017-03-02
  • 0
    Apparently you haven't noticed, I never claimed it was. I simply claimed that your doing the work might help you see what you are not seeing. It's up to you. You can have the last word if that makes you happy, I am done discussing this.2017-03-02
  • 2
    @MathAddict - so You're saing, that I need to check how this figure behave for $X\neq 39^{\circ}$? That's a lot of angles to check. Even GeoGebra won't be able to do it, because we need to check the whole intervals $(0^{\circ}, 39^{\circ})$ and $(39^{\circ}, 103^{\circ})$. Also it is not clear how assumption $DA\neq DC$ leads to the contradiction. Maybe instead of taking offense You improve your reasoning, for example by showing, that this construction is unique?2017-03-03
  • 0
    @Matlak: I'm not taking offense but, it's "peculiar" to have someone (timon92) who makes a veiled assumption and then simply proclaims it to be true (Z = C) without a shred of proof, talk about anyone's logic. Anyway, in what I presented GAC = ADC + ACD because GAC is an exterior angle. If, and only if, the triangle is isosceles, GAC also equals ADC + DAC (since ACD = DAC). Additionally, *if and only if*, the triangle is isosceles can GAC = DPC + BDC = JCA. If GAC = DPC + BDC and JCA = GAC + BDC then GAC = JCA therefore the triangle MUST be isosceles. I have GeoGebra files to visualize this.2017-03-03
  • 0
    In addition to the above, here are a couple links to GeoGebra files. The first one [link](https://www.geogebra.org/m/BX2TKVN5) shows that the verification is correct (Euclid figured it out 2300 years ago). The second example [link](https://www.geogebra.org/m/PmUzPzWD) shows that if someone wants to "fudge" the angles by adding to one what was subtracted from the other, this results in a non constructible shape. IOW, there *will* be a contradiction somewhere in spite of any ill-conceived efforts to conceal it (as timon92 suggested in one of his previous comments.)2017-03-03
  • 1
    @MathAddict If you have a question concerning my solution above you should leave a comment below it, not here. It seems you simply misunderstood what is going on in my solution. Anyway, here's a brief explanation: I defined point $Z$ and then proved that it satisfies exactly the same angle equalities defining $C$, thus $Z$ and $C$ coincide.2017-03-04
  • 0
    Just for the record, it takes 50 reputation points to comment. Therefore I cannot place a comment on your "solution". Additionally, even if I had the necessary points, I still probably wouldn't - it just isn't worth it. I can and, could vote your solution down as it deserves to be but I didn't. Lastly, you didn't prove anything, you simply declared at the end, the veiled assumption you made at the beginning that Z = C. The fact is, you, in a veiled way, assumed the triangle is isoceles and did not provide any proof that it is. Since the assumption is correct, as I proved, then it worked.2017-03-04
  • 0
    @MathAddict - You point it that way: triangle is isosceles iff angle is 39. Great, but you stil don't answer the question - WHY it is isosceles? You've pointed one thing not dependent of assumption - GAC = ADC + ACD, but it doesn't lead to isoscelesness of triangle (even, if you've used this in your reasoning). You have to show that this construction is unique by showing, that it doesn't work if triangle is not isosceles or angle is not 39 (for ALL angles) or uniqueness of construction of APBCD - otherwise your reasoning is incomplete.2017-03-07
  • 0
    I am afraid I cannot find another way of explaining the reason. It is evident that the triangle is isosceles _iff_ GAC = ADC + DAC and GAC = DPC + BDC = JCA, assuming either one in a non-isosceles triangle inevitably leads to triangles whose interior angles don't add to 180 degrees. GAC = JCA can only be true iff the triangle is isosceles additionally, the converse is true as well (in this particular case, P implies Q and Q implies P). If GAC was not equal to JCA the angles calculated making that assumption would violate the sum of angles in a triangle.2017-03-07