Taepered plenum

[shivam popli]

We are makinng a cylindrical plenum in which inlet is from one end of it.To make the flow equal in each cylinder we need to taper it from one end to another.
This is due to the fact that taper would cause an increase in velocity of air as we go from first cyliner to fourth.This would make even flow in all cylinders.
I need to know how much taper it would be?
Are there any calculations regarding this?

[Gaanja]

Hey Shivam,

Try and get a CFD tool like ansys cfx. I am sure it must be available at your university. Run your model on the software to obtain optimum flow. Keep changing the design until you are satisfied with the flow. Also remember you can design an intake which looks great on CAD and gives the best results on CFD analysis. But you may not be able to manufacture it. Keep manufacturability in mind when you are designing.

cheers

[Hector]

To agree with Gaanja, this is definitely beyond the realm of classical fluids analysis. You're going to want a CFD program to do this analysis.

It is worth mentioning that air doesn't flow through an intake quite as you'd expect. The air sort of pulses through the intake based on pressure waves created by the intake valves opening and closing. These pressure waves are dependents on plenum size, runner lengths and geometry, intake and exhaust cams, and the efficiency of your restrictor. In order to accurately model this flow you're going to need some complex analysis. Ricardo VECTIS is a program designed to do exactly what I am describing.

Judges are not going to be impressed with rough hand calculations showing theoretical flow through your intake. They will also not be impressed with a steady-state CFD analysis. A true dynamic analysis is what you need.

[Demon Of Speed]

Additionally look at "The Scientific Design of Exhaust and Intake Systems" by Philip Smith. It is an old book, but the principles hold up.

[Jimmy01]

Judges are not going to be impressed with rough hand calculations showing theoretical flow through your intake. They will also not be impressed with a steady-state CFD analysis. A true dynamic analysis is what you need.

I disagree. Judges are more likely to be impressed with someone who has done the calculations and understands what is happening (and can explain it) rather than pretty CFD pictures. Sure, an in depth study with testing to validate your models and an excellent understanding will reap the most benefits and possibly points in design. However, unless you have heaps of spare time (you don't) or take it on as a big project the results will likely be average. Judegs want to see that you understand the systems on your car and why you have done them a particular way (in my experience anyway).

[Hector]

I guess I disagree with that statement because I don't think that hand calcs are a viable way to accurately analyze an intake. It would be like trying to do stress analysis of something with complex geometry (think a billet upright) with nothing but equations from your sophomore-level strength of materials class. Sure, you could make a valiant effort, but how good do you think it would be?

[Demon Of Speed]

Originally posted by Hector:
It would be like trying to do stress analysis of something with complex geometry (think a billet upright) with nothing but equations from your sophomore-level strength of materials class. Sure, you could make a valiant effort, but how good do you think it would be?

Considering all FEA does is is solve a LOT of the solid problems for each element(kind of), and when you use enough elements it is pretty accurate, and people love to trust FEA. Why couldn't some one do CFD with hand calculations and get pretty good results?

[Charlie]

RE: the original question, I am not convinced you need a tapered plenum to achieve equal cylinder flow. You should be, before deciding 'how much'. A plenum is not designed for smooth, constant velocity flow. It is not like designing an exhaust manifold.


Originally posted by Demon Of Speed:
<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Hector:
It would be like trying to do stress analysis of something with complex geometry (think a billet upright) with nothing but equations from your sophomore-level strength of materials class. Sure, you could make a valiant effort, but how good do you think it would be?

Considering all FEA does is is solve a LOT of the solid problems for each element(kind of), and when you use enough elements it is pretty accurate, and people love to trust FEA. Why couldn't some one do CFD with hand calculations and get pretty good results? </div></BLOCKQUOTE>

Not sure if you are tonque-in-cheek or not, but doing FEA by hand would be a monumental task for most things that utilize FEA. It can be done but major simplifications would be required.

Now for an intake, the entire problem becomes vastly more complex.

All I can say, is if you want to do FD without the C, give it a go and let me know how far you get!

[Demon Of Speed]

Originally posted by Charlie:
RE: the original question, I am not convinced you need a tapered plenum to achieve equal cylinder flow. You should be, before deciding 'how much'. A plenum is not designed for smooth, constant velocity flow. It is not like designing an exhaust manifold.

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Demon Of Speed:
<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Hector:
It would be like trying to do stress analysis of something with complex geometry (think a billet upright) with nothing but equations from your sophomore-level strength of materials class. Sure, you could make a valiant effort, but how good do you think it would be?

Considering all FEA does is is solve a LOT of the solid problems for each element(kind of), and when you use enough elements it is pretty accurate, and people love to trust FEA. Why couldn't some one do CFD with hand calculations and get pretty good results? </div></BLOCKQUOTE>

Not sure if you are tonque-in-cheek or not, but doing FEA by hand would be a monumental task for most things that utilize FEA. It can be done but major simplifications would be required.

Now for an intake, the entire problem becomes vastly more complex.

All I can say, is if you want to do FD without the C, give it a go and let me know how far you get! </div></BLOCKQUOTE>

The point isn't that you would do FEA or FD of an CFD by hand, it is that you can get a simplified idea of what is happening. Then Compare it to CFD and the real thing. Additionally, you have to know how to properly use CFD (like FEA) to get good results.

[jpusb]

Well for me it all depends on your goals on that event. Maybe you didn't have time-resources and you have to do a steady state analysis, or whatever. Maybe you are in a 8 people team which not only has to do the car but also get resources or whatever. In FSAE it all depends on your particular situation an the situation of the team.

Of course if your design event result will decide if you win or not the competition (as seen many years) you better not go with hand calcs...