How to modify the trailing edge thickness

[gerritgroot]

Hi,

Does anyone know whether there's a standard procedure for modifying the trailing edge thickness of 2D airfoils?
(maintaining the airfoil's relative thickness)

Thanks,

Gerrit

 

[Sparweb]

Get out the electronic belt-sander!

Seriously, if you have a data file with point in it, then just edit the file and change the y-coordinate of the upper trailing edge.


Steven Fahey, CET

 

[gerritgroot]

I guess you don't know about aerodynamics

 

[serker]

The reality is that aerodynamicist always (99.99% of the time) use the NACA tabular form or otherwise give out a pointy TE airfoil and the design and/or manufacturing group always open up the trailing edge to something reasonable to avoid knife edged parts.

It helps keep the assembly mechanics and ground crews with all their limbs and away from the aid stations. On composite parts it also helps avoid chipped edges on the finished unit.

Most of the time when you tell the aero guys what you are going to do they say 'yeah yeah whatever' unless they are the analytical cubie dwellers who are completely disconnected from reality.

If your chord length is so small that your desired manufacturing TE thickness can't be obtained smoothly with just changing the last set of points you have to smear it out through two or three sets to get a smooth transition.

 

[gerritgroot]

Yes, the chords are too small partially.

I compared different types of "yeah yeah whatever" and the results are too different to leave it to the structural guys.

I'll increase the chord ratio for the modified part.

Gerrit

 

[btrueblood]

Gerrit,

Do you have access to the original NACA airfoil reports, and do those definitions state anything about the tolerance on the trailing edge? I tried to pull up a few on the NTRS server yesterday, but couldn't quite drill down far enough to find the original reports, only surveys done later.

A blunter research effort, searching only on NACA reports with the term "trailing edge radius" yielded:

http://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/19930086925_1993086925.pdf

The abstract of which refers to a "usual 0.15 percent chord" trailing radius for 65-series airfoils, and discusses the effect of blunting the trailing edge to 1 and 2 percent of chord.

this earlier report

http://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/19930081466_1993081466.pdf

talks about a systematic variation in trailing edge radius for propeller airfoil sections, and states that 0.5 percent t.e. radius yielded acceptable results.

FWIW, I have seen aero engineers make estimates of airfoil wake thicknesses, and estimate section drag coefficient changes based on increased profile drag for blunter te profiles...but I don't know that any conclusions were ever reached, just that they did manage to justify (as earlier posters allude) whatever radius the production guys could reliably offer.

 

[serker]

With respect to a percent number, that is probably spot on for 2D sections and for parts where the chord doesn't change much.

When it comes to other applications it may be as likely a set number. In other words your wing TE may be somewhere between a tenth and a quarter of an inch and possibly the same number from root to tip even though the chord varies from several hundred inches down to two or three feet. (And that isn't even getting into multiple sections as is often required. Not to mention the compromises involved with including trailing edge devices.)

Essentially it is the same issue as the tabular data. When you scale .15% of 3' you end up with 54 thou but the same on 20' and you are up to 360 thou. Maybe your application is good with those numbers or maybe your production methods and other considerations limit you to say 150 thou minimum.

On other parts, propellers or impeller blades and the like your minimums may be a lot smaller and the TE sharper.

In the best of worlds it is something that gets worked out and agreed to up front in the design without too much argument. In the worst everyone is convinced they have to have their way so the engineering gets muddled because the design gets driven by one or the other technology area and the manufacturing ends up changing it on the fly to make the parts. If you are lucky it not only can be produced but still meets the design goals when they are done and the changes get incorporated so you know not only what you designed but what you built when you need to modify it or put out a replacement part for something 20 years down the line.

Much better to hash it out and get it close to right before you start bending metal or make molds.

 

[gerritgroot]

Thanks for the answers, I'll have a look at the provided links!

Gerrit