Forum: Technology

Roscoe, this one is for you. Is it possible to imitate Supersonic Airflow past a subsonic wind tunnel model in a wind tunnel by manipulating the denominator in the equation: RN = V*Dist/CoAbv/density by lowering the temperature within the density to produce a lower Vslo, thus lowering the sonic velocity IAW scale effect? Question

V, velocity; ft/s
Dist, feet at a point down wind on an airfoil
CoAbV, coefficient of Absolute Viscosity, derived from the ISA
Density, pressure/temp
Vslo, Velocity, sonic, local

F-16.net Sponsor

Hi Allen,

Could you rephrase so that the question is clear to all of us?

First off, RN has nothing to do with supersonic flow. Think of Reynolds just like Mach. A non-dimensional parameter to create similarity (or explain differences). If you want to truly scale, you have to match both. VERY hard technically and VERY $$ to do.

Mach is a measure of compressibility. For lack of a clearer description, RN is a measure of "energy" in the flow as it progresses through a medium or along a surface. For example as the flow runs along the body, surface friction builds up a slower moving boundary layer...eventually the BL will trip from laminar (smooth) flow to turbulent (tumbling) flow. Where that trip point occurs is defined by the RN. The BL starts thickening faster here and the drag goes up.

Matching RN on a scaled model is critical to accurately reproduce the transition from laminar to turbulent flow (important for drag calculation mostly). This is why scaling RN is so hard (and why really big wind tunnels were built to minimize the impact). On a big airplane the flow might go turbulent near the front of the aircraft. Since the vast bulk airflow now is turbulent flow, the overall drag is high. A scale model is so short that in the same airstream the Rn may never reach a value to trip the flow. Thus the entire flow is laminar and the drag is much lower and therefore not representative of the full scale version. Some tricks can be done...adding surface roughness or trip wires to the model to force turbulent flow, but you have to know where it occurs first which you probably don't and tricks can mess up other measurements.

As far as simulating supersonic flow...I'm pretty sure you can't do that with RN.

Fun time: Here is a way to see the effects of RN in your own home.

Open your sink faucet to a trickle. If you can, adjust it until you can get a smooth laminar flow all the way until it impacts the bowl. As it (1) passes through the air and (2) picks up speed, the RN is building. So long as it stays laminar, the RN has not reached the trip point (which varies by the way with different fluids). Now this part might be hard (some faucets create turbulence themselves so this may not work). Slowly open the faucet (faster flow) until it starts to go turbulent somewhere in the stream (it should start at the bottom and move up as the faucet is opened further). The water flow has too much internal "energy" to stay laminar...that is where the flow reached the magic RN trip point threshold.

In a wind tunnel, we have an aircraft model for testing in a subsonic condition. We would like to imitate a supersonic condition by lowering the temperature of the mass airflow to give a lower "Reynolds Number". A dimensionless number, usually in the hundreds of thousands and going as high as a million, depending on the temperature.

The viscosity of gases is a function of temperature and an increase in temperature increases the the viscosity, or better known as (kinematic viscosity, nu). It is acquired from the ISA.

What I am trying to figure out is this, can I imitate a supersonic condition in a subsonic flow? The equation for RN is remains the same as is mentioned in my first post.

If you lower the temperature, the denomiminator of the equation gets higher making the Reynolds number smaller, meaning a higher RN and subsequently a higher velocity.

The wind tunnel would need a temperature control, if this is possible, and the converging throat raises the temperature, can we still imitate a supersonic condition all things considered?

Thank you Roscoe, you've answered my question.

Quote:

Some tricks can be done...adding surface roughness or trip wires to the model to force turbulent flow, but you have to know where it occurs first which you probably don't and tricks can mess up other measurements.

Some small wind tunnels at Ames (all?) operate at higher pressures to account for differences in flow (scaling effect in RNs). The giant wind tunnel has not been used in a couple of years and IIRC, it will be dismantled. I imagine this means scaling techniques have gotten really good.

Hello everyone. I have researched and queried several wind tunnel companies including Ames, where they have tested the F-22 and JSF prototypes. They have all mentioned you cannot imitate supersonic flow in a subsonic wind tunnel throat. If you plan on experimentation in a supersonic flow, in the case of an atmospheric vehicle, you must place your model in a supersonic converging reservoir and then a constant atmospheric throat.

When temperatures reach in excess of 250 degrees F., I guess plastic models won't do in a supersonic wind tunnel, but rather must be constructed of aluminum or steel. Shocked

I do not have this technology at my house, but I guess I can do something on paper, (computer program), I have the equations. However I was looking forward to viewing the shock waves produced on some hypothetical designs as well as some already designed aircraft via the Schlieren and Shadow optical graph systems

RN's are a very superfluous and require many redundant computations, unless of course you're concerned with only a speciific area of an airfoil. NASA Ames requires extensive administration detail to get into but it is possible. The process of registration requires about a year. I brought up this post just to see what was out in the forum as far anybody knowing anything about wind tunnel testing, which I plan on doing sometime. Thanks Roscoe and Pat. Very Happy

Pat1 wrote:

Quote:

Some tricks can be done...adding surface roughness or trip wires to the model to force turbulent flow, but you have to know where it occurs first which you probably don't and tricks can mess up other measurements.

Some small wind tunnels at Ames (all?) operate at higher pressures to account for differences in flow (scaling effect in RNs). The giant wind tunnel has not been used in a couple of years and IIRC, it will be dismantled. I imagine this means scaling techniques have gotten really good.

Or there are too few customers to justify the expense of maintaining the tunnel...I've been in that tunnel. It is HUGE!

So I've been told Roscoe, they can fit an actual aircraft in it, but they did not specify the actual size or aircraft that have been inside the tunnel. Rolling Eyes

allenperos wrote:

So I've been told Roscoe, they can fit an actual aircraft in it, but they did not specify the actual size or aircraft that have been inside the tunnel. Rolling Eyes

A technician told me a 737 has been inside.

Roscoe wrote:

Or there are too few customers to justify the expense of maintaining the tunnel...I've been in that tunnel. It is HUGE!

That’s what I've heard too; due to improved simulations and scaling schemes.

Wow!! That's a big tunnel, thanks for the info, now I know what else they could have put inside of the tunnel to test gather data.

Forum: Technology

Reynolds Numbers and Scale Effect