Flow Bench / CFD analysis

[Paul Achard]

Hey everyone,

I'm currently in charge of setting up a flow bench for testing our team's throttle designs. I know this isn't supposed to be be rocket science (oops, i mean rocket engineering), but I didnt know squat about about airflow and whatnot before looking into this.

I've read that the basic architecture I should follow is
(testpiece)->(test gauge)->(flow control valve)->(mass flow sensor)->(shop vac)
We have a Bosch HFM5 air mass meter already on hand.

I was wondering what device to use as the air flow valve.
Also, is it viable to place the sensor anywhere within the system? Suppose I wanted to place it between the air filter and the throttle to help determine CFD boundary conditions...

Finally, speaking of CFD analysis, could someone clarify what exactly ARE boundary conditions and at what points they should be evaluated? (For example, if you're studying a throttle with integrated restrictor, taking into account air filter effects...)

Many thanks.

[Paul Achard]

Hey everyone,

I'm currently in charge of setting up a flow bench for testing our team's throttle designs. I know this isn't supposed to be be rocket science (oops, i mean rocket engineering), but I didnt know squat about about airflow and whatnot before looking into this.

I've read that the basic architecture I should follow is
(testpiece)->(test gauge)->(flow control valve)->(mass flow sensor)->(shop vac)
We have a Bosch HFM5 air mass meter already on hand.

I was wondering what device to use as the air flow valve.
Also, is it viable to place the sensor anywhere within the system? Suppose I wanted to place it between the air filter and the throttle to help determine CFD boundary conditions...

Finally, speaking of CFD analysis, could someone clarify what exactly ARE boundary conditions and at what points they should be evaluated? (For example, if you're studying a throttle with integrated restrictor, taking into account air filter effects...)

Many thanks.

[Xeilos]

Simply put boundary conditions are condition that occur at the boundary(s). I am not trying to be smart in this but literally that is what you are.

Work through the math of solving a differential equation (any differential equation will do). You have to integrate, which will lead you to a constant of integration at the end of your solution (if you did not get a constant, you need more help with calculus than I can provide here). Now ask yourself, how do I solve for this constant. You need something, where the solution is known. If you are solving an equation that is for velocity extending away from a wall (velocity is a function of a one-dimensional position to keep it simple), you need to know the velocity at some given point.

In fluids, a non-slip condition at the wall (velocity at 0-distance from the wall) is one type of boundary condition. Another is the pressure at the entrance, or the mass flow rate, or the temperature, or ...

Your numerical methods course coupled together with a fluids course would go a long way to teaching you this stuff. My lecture is just scratching a scratch on the surface (the idea still holds).

I would also question your leaders judgement in assigning you this project without fluids experience or even the necessary calculus experience to be able to complete this. I know (looking back of course) that after my first year of Uni, I knew virtually nothing. I still know virtually nothing and would be hesitant to give a CFD project to anyone before 3rd year (atleast at our school, I do not think the Ontario/Quebec system vary that much).

For all of that good luck.

[Paul Achard]

Thanks for the reply.

Just by the way, I'm not at all in charge of CFD analysis. For now I'm only working on getting a flow bench together, and senior members will help for calibration and analysis.

I was just probing the infinite chasm of wisdom of the FSAE forums to collect a little info on CFD, since it will ultimately depend on our flowbench data.

[Ben K]

Yeah...CFD is definitely a can of worms you don't want to open just yet. Our team is going to probably make a small scale flow bench in the winter. Definitely post some updates on how it goes for you!

Ben

[samphlett]

You also need to remember that your restrictor will not have steady-state boundary conditions. It may behave very differently when there's an engine downstream of it.