>> You are given a bucket of water (fixed to the ground) and an open-ended
>> flexible pipe leading off into a "black box". All you know about the
>> black box is that it may allow some flow into it, at a variable,
>> unpredictable rate.
>
> I presume the "black box" is at the same level as the bucket?

More or less, compared to the pressures involved yes.

>> Design a system to take the water from the bucket, and supply it into
>> the pipe at a constant pressure (not a ridiculous pressure, just
>> something like the pressure you would have in your water pipes at home,
>> maybe 3 bar?).
>>
>> Maybe I just hang about too many engineers, but to me even a
>> non-engineer would be able to have at least a stab at some ideas for
>> solving it.
>
> My first stab would be to re-invent the water tower - though I guess
> you're after a more compact device. (Which is a pity, because it's the
> solution that places the lowest demands on the pumps.)

To be honest I'm pleased with any device! After one person couldn't 
think of anything, I asked how they thought the mains water pressure was 
created, then they came up with the water tower method.

It's also good if they probe to better clarify the requirements (ie does 
it need to be compact or not, any cost limits, availability of 
electricity/compressed air etc). So in one sentence you've done better 
than all the mechanical engineering graduates so far :-)

> My next stab would be to use a gear pump, and offload the tricky part to
> some other engineer whom I'd contract to build a constant-torque motor -
> though I guess you'd consider that cheating.

No not at all. I think this is maybe some of the problem, the candidates 
are thinking they need to get into the detail design or something novel 
from scratch. Where actually, saying you can use a pump and a safety 
valve (or measuring the pressure electronically and using that to 
control the speed of the pump motor) is a much better answer.

> Yet another stab: Use a pump that can provide the projected peak flow at
> a pressure of 3 bar, and connect a safety valve that opens at 3 bar.
> Obviously this wastes a lot of water, but we can fix that by feeding the
> spilled water back into the bucket.

One person did actually come up with that answer after some suggestions. 
It's how the fuel system works in a car (fuel is bled back to the fuel 
tank to maintain the correct pressure at the injectors).

Lots of candidates though started off by talking about using a secondary 
holding container, pistons, springs, then motors and safety valves etc. 
Maybe exactly because they have been studying mechanical engineering in 
detail for the last 4 years they can't look at the bigger picture?

Post a reply to this message