The volume flow rate is kept constant and the specific gravity is doubled.
I should have mentioned this before, my bad
Thanks, Cramster.
That clarifies the conditions. May be it has been already answered above, I didn't check. Too much words, so I will share my view on this paradox.
1. This question is not simple. Since it includes some internal paradox which was not cleanly mentioned in standard hydrodynamics textbooks.
Let first simplify the question. Let pipe be straight, with constant radius and at the same level. Then Bernouli said:
P+rho*g*h+0.5*rho*v^2=CONST.
What it tells us? if delta h = 0, cross-section does not change and volume per second = const, then we mast conclude that
P=const. at any pipe point.
It looks right, because if the speed is const, then sum of external forces should be zero.
Here is two consequences:
a) The rho does not affect pressure; Since I can apply any additional external pressure to both ends of the pipe, but movement will not change.
b) We need to avoid a mistake, to equate two bernoulli equations. For different rho we don't know if the CONST is the same. (it is not)
All in all for simple pipe Pressure does not depend on density. One more very important point that I will be using later: No work is done. It is just inertia.
2. Now, the interesting part. Let say our pipe has two segment: one with cross-section S, another S'.
Should we perform work in order to keep speed steady in this system? Surprise: We should. Why? Because we changing the Kinetic energy for the same small mass of liquid. Let me show that,
Let's apply force F to the entrance of the pipe using piston. Since we already proved that P=const in the pipe of the same radius, then we will apply force equal P*S for some distance X and perform the work P*S*X.
Are we done with work. Not yet: We need to remove the work which was done from another side of the pipe, or P'S'X', Total work o this segment of liquid is : W=(P-P')*Volume. (Volume=S*X=S'*X')
Where this work is going? To change the kinetic energy of the volume that we pushed to the second segment. Speed there is faster. Change in kinetic energy will be:
0.5*M*V^2=0.5*rho*Volume* (delta V)^2.
Work is equal to the change in energy for ideal liquid and we get:
(P-P')*Volume = 0.5*rho*Volume*(delta V)^2 or
P-P' = 0.5 * rho * (delta V)^2.
3. Last step.
Since by the problem conditions Volume per second = const, then (delta V)^2 is the same for any liquid. Then(collecting the constants)
P = P' + k*rho
Linear equation.... P grows with rho. The EK's authors may assume that P'=0 and get direct proportionality as in the answer.
4. Conclusions:
a) To keep constant speed in the pipe with changing diameter, we must do the work: Not against the friction or viscosity, but to speed up or slow down the mass that we pushing throw. This point is missed in basic textbooks.
b) If we are not keeping up with this work, the speed will drop or we will get not stationary flow.
c) We can use calculus to generalize this model for pipe of any shape.
Sorry, if it gets too long, but I hope this will clear the misunderstanding. I think it is not very fair from EK, to ask this question and expect a solution in 90 seconds. Best of luck.