Ideal radius for a bellmouth inlet

[Talmadge]

Surely this is basic, I want to calculate the ideal bellmouth radius (and lenght?) for a 2" opening where the flow rate is about 900 cfm of air (assumed 28" H2O presure drop).

Thanks in advance,

 

[PEW]

Talmadge,

I think the length of the intake tube depends on what you want the response of the engine to be, the longer you go the better the mid-range torque.

The leading edge radius is required to allow the airflow entrained from the sides of the air intake to turn smoothly down the intake without overshooting the turn or becoming turbulent, which can cause effective airflow "narrowing" in the first part of the inlet tube. I don't know if there is a method of calculating this but I would have thought that a radius of 1/4" would be sufficient provided that you use an edge that is fully turned back on itself. I have seen some so-called performance inlet trumpets that don't have this and have a perforated metal screen as a "filter". I think these are probably worse for airflow than an open intake would be, the only advantage would perhaps be to stop small children falling in.

David Vizard has written a good section on intake tubes/ rampipes in his book "Tuning the A series engine". It is primarily for that articular engine (Leyland Mini) but it is full of useful stuff for general consumption. David uses airflow benches to prove his ideas.

Hope this is of some use, I noticed that folks aren't exactly rushing to give advice on this one.

 

[Talmadge]

Thanks for the insight. The device is actually a flow meter and I plan to increase the bore size to get more flow. The flow meter is about 4" long and opens up into a plenum. I was thinking of a radius say about 1/2D.

Regards

 

[PEW]

I am sure that 1/2D is more than ample for the radius.

Have you considered using an elliptical edge rather than a round one? I understand that this form prevents the build up of resonance in the airflow, the last thing you want in a flowmeter.

 

[jimkden]

It's time for a little 200 series fluid mechanics.
If you have 28"w.g. to pull the flow, you won't make it because the fluid when set in motion required energy to do it (fluid is at rest initially because you want a bell mouth inlet). Mean velocity is approx 41,300 fpm and the velocity head (energy in the fluid flowing, not static head) is about 106"w.g. So you gotta have about over 106"w.g. + fittings and meter loss to get the 900cfm. And keep in mind sonic is about 1138fps at 80F (kgRT)^0.5

 

[Warpspeed]

You are trying to accelerate the air from stationary to a very high velocity with minimum pressure drop. The best way to do this is with constant acceleration of the air along the entry, which you will get with an exponential horn.

jimkden is right, you are not going to be able to reach your target flow and pressure drop.

 

Are there formulas for determining the proper exponential entry/exit shapes or is it necessary to use some kind of method of characteristics approach (like when dealing with sonic flow)?

 

[Tmoose]

relative flows of various inlet treatments appear here -
source unknown.
I figure unless I'm at full throttle >>I<< am the limit to airlflow and power.

http://www.spannerfodder.com/the_wo...e=images/the_works_escorts/pinto/ramflows.gif

 

Those are from David Vizard's book on the BMC A-series engine.