0
$\begingroup$

So, currently my team and I are working on a glider project for our senior design. The design we've come up with causes XFLR to crash for some unknown reason (the issue seems to be at the root airfoil which is wedge shaped). What I would like to know is if there any viable alternatives to XFLR for airfoil and aircraft analysis that won't crash like XFLR does.

Edit: So the main goal of this project is develop an autonomous system to be used in hypersonic glide vehicles. While obviously we're not going to be making a hypersonic glider, my professor is insistent that it at least looks hypersonic. It needs to be able to glide for about 2000 feet after being dropped from a height of 300 feet.

$\endgroup$
  • $\begingroup$ Are you trying to characterize airfoils for use in hypersonic flight? $\endgroup$ – Marius Sep 12 '19 at 3:49
  • 1
    $\begingroup$ my professor is insistent that it at least looks hypersonic. Please tell me that this is a professor in the field of visual communication or marketing, not in the field of aerospace engineering. $\endgroup$ – DeltaLima Sep 12 '19 at 9:09
  • 2
    $\begingroup$ What is the criteria for "looks hypersonic"? That aside, this question is about resource location, which is off-topic here. $\endgroup$ – Ralph J Sep 12 '19 at 9:10
  • $\begingroup$ @RalphJ basically, similar in appearance to the x-20 or x-30 $\endgroup$ – ironduke97 Sep 12 '19 at 14:34
  • $\begingroup$ @DeltaLima let me rephrase. He wants it to appear as though it has the characteristics of a hypersonic vehicle. He is an aerospace engineering professor $\endgroup$ – ironduke97 Sep 12 '19 at 14:35
1
$\begingroup$

My initial response is that, if you have a wedge-shaped airfoil on a glider, what sort of spec are you designing to? Most of the time, gliders are low-subsonic aircraft with airfoils similar to Epplers, Wortmann FX, or 4-series NACAs, not wedges. I associate wedges with massive flow separation or supersonic flows. Can you provide a picture of what you're talking about, as well as a discussion of what you're designing to? Also, are you attempting to produce a 3D analysis of a wing that, at its wing box, has this wedge-shaped airfoil?

That said, my short answer is this: XFLR is a panel-method solver, which means that it cannot deal with flow separation save by approximation. As a result, shapes such as a wedge, which are quite prone to flow separation by virtue of their shape, are very hard, if not impossible, to analyze with these solvers. Unfortunately, lower-order solvers (i.e., something short of finite-element or finite-volume methods, such as Fluent, CFX, etc.) bank on using these panel methods or other lower-order fluid flow equations that cannot handle flow separation.

If you want to try to accurately model flow separation, you'll need a higher order CFD solver, which brings with it additional complexity, cost, and headache--and getting accurate drag information is a tricky prospect. There are some freeware solvers out there, but, by far, the most popular (and widely validated) is OpenFOAM. This is a command-line program at heart, but there are some GUIs developed for it that you may be able to obtain for free or at a reduced cost, such as SimScale. Some other packages are discussed here as well, if you'd like to have a look.

That said, what data do you need? If all you need is a sectional drag coefficient, I would highly recommend a look through Hoerner's Fluid Dynamic Drag (not to be confused with his other text, Fluid Dynamic Lift), a vital reference that contains data from a plethora of experiments done on a multitude of shapes. In it, he describes the drag force on objects from ellipsoids to cones to cylinders and more, all based on size, aspect, roughness, and Reynold's number. There is an updated copy available, but I haven't found it online as of yet.

|improve this answer|||||
$\endgroup$
  • $\begingroup$ So the main goal of this project is develop an autonomous system to be used in hypersonic glide vehicles. While obviously we're not going to be making a hypersonic glider, my professor is insistent that it at least looks hypersonic. It needs to be able to glide for about 2000 feet after being dropped from a height of 300 feet. $\endgroup$ – ironduke97 Sep 12 '19 at 3:20
  • $\begingroup$ @ironduke97 That's a pretty important bit you should probably mention in your OP...what flow regime are you running in the solver? $\endgroup$ – JZYL Sep 12 '19 at 3:23
  • $\begingroup$ @Jimmy so far, we've run analysis for Reynolds numbers from 100k to 900k in increments of 10k. The glider is only travelling at around 22 mph or so. $\endgroup$ – ironduke97 Sep 12 '19 at 3:27
  • $\begingroup$ This just got more interesting. Yes, XFLR is definitely not the tool you want. There is some evidence for hypersonics done well in OpenFOAM and variants (bit.ly/2maCEwl), but it is going to be much more difficult. Possible answers of the non-CFD variety: aviation.stackexchange.com/questions/31607/…, thesis.library.caltech.edu/5025/1/Wunch_ws_1948.pdf. $\endgroup$ – Marius Sep 12 '19 at 3:57
0
$\begingroup$

From what I've gathered, you are designing a small, low speed, unpowered aircraft with a glide ratio of 6.7 that must look like a hypersonic vehicle. In order to meet the latter requirement, you chose a wedge airfoil. I'm also going to assume that your aircraft has very low aspect ratio, highly swept, thin body that generates most of the lift.

I wouldn't worry too much about the airfoil. Just assume it's a flat plate. Try it out with XFLR's panel method. If it still fails for some reason, then just go with a VLM method like AVL. It should get you into the ball park for lift. Maybe assume a $C_{L_{max}}$ of 1 for conservatism.

For drag, just use the Blasius solution for skin coefficient; or you can use Hoerner's as suggested by the previous answer. Make sure you have the right Reynold's number. 900,000 as you mentioned in the comments seems way too high for what you have described.

|improve this answer|||||
$\endgroup$
  • $\begingroup$ You pretty much got it right. The only issue is a lot of the points are unconverged and we get a lot of "outside flight envelope" errors $\endgroup$ – ironduke97 Sep 14 '19 at 3:59
  • $\begingroup$ @ironduke97 Have you tried AVL? $\endgroup$ – JZYL Sep 14 '19 at 5:03
  • $\begingroup$ I've only heard of it $\endgroup$ – ironduke97 Sep 19 '19 at 19:41

Not the answer you're looking for? Browse other questions tagged or ask your own question.