May 25, 2013, 11:27 AM
Registered User
Switzerland
Joined Mar 2003
2,154 Posts
Mini-HowTo
FLZ Vortex

FLZ Vortex is a program that uses the vortex lattice method to calculate model aircraft. The software is a free download from here: http://www.flz-vortex.de/flz_vortex.html
And this is what Wikipedia says about the vortex lattice method: http://en.wikipedia.org/wiki/Vortex_lattice_method

I have used it with good success to optimize incidences, washout and CG on my scratch built scale planes.

It seems to be fairly unknown around here and can be a bit intimidating at first. So I thought I'd show here step by step how I use it.

The screen shot/appetizer shows my Hawker Tempest in very slow flight without flaps deployed. See the Tempest build thread here: http://www.rcgroups.com/forums/showthread.php?t=1471867

Hans

# Images

May 25, 2013, 11:39 AM
Registered User
Switzerland
Joined Mar 2003
2,154 Posts
this is how the program looks when opened.

I'll set up a low wing trainer with a single taper wing.

Hit the "new" button in lower third of the screen and fill in designer's name, plane's name and projected AUW.

Hans

# Images

May 25, 2013, 12:02 PM
Registered User
Switzerland
Joined Mar 2003
2,154 Posts
then go to tab "wing" and hit the "new" button to generate a new wing. FLZ calls all surfaces (including the rudder) wings.
Fill in the wing's name and set its mass to zero. I do this to keep the confusion down, wing mass would be added to plane mass we entered earlier.
FLZ now shows the plan view on the left (flight direction is towards the top) and 3-dimensional view on the right. The 3-dimensional view can be rotated in all directions by clicking the view and dragging it.

The next tab "pos" tab lets you change the wings position, we'll leave this at 0/0/0 for this wing (this will be used when placing the tail).

The next tab "airfoil / chord" lets you set those for the root. I'll use 0.3m and a NACA2415. The airfoil can be chosen from a long list.
You'll need the find the .dat file if you need one that's not in the list and copy it into the airfoil directory "profile_dat".

The next tab "panels" sets the number of lengthwise panels for the vortex lattice calcs. I usually leave this at the default setting of 4 and linear.

The next tab "angle" lets you set incidence of this wing relative to the fuselage. I'll just leave this at zero for now. We'll calculate the correct value later.
I don't know what root wash angle does; I never use it.

Hans

# Images

May 25, 2013, 12:24 PM
Registered User
Switzerland
Joined Mar 2003
2,154 Posts
the rest of the wing's geometry is entered in the lower right.

First tab "width": enter span of the selected panel (red in the left picture). I use 0.75m for a total span of 1.5m.

Next tab "airfoil / chord" lets you set the outboard airfoil and chord for the selected panel. I use 0.18m and NACA2415.

The wing got asymmetric as you can see. Fix this by ticking the box "always mirroring".

Next tab "dihedral" lets you set dihedral for this panel. I'll use 3 degrees. Dihedral is per side. You can see this in the 3-dimensional view on the right.

Next tab "sweep" gives you several options to enter sweep. I'll use variety 1 and enter 0.03m.

Next tab "geometric twist" lets you set washout. I'll leave this at zero for now. Angles need to be negative for washout, positive for wash in.
Exaggerate if you're not sure and enter something like 20 degrees and see the result in the 3-dimensional view.

Next tab "panels" sets the number of spanwise panels for the calculations. I usually leave this at linear and use the "panel automatic" button. More panels are better but take longer to calculate. It's also better to have more panels in areas of geometry changes (edges of flaps, taper changes, dihedral changes etc).

The next two tabs "flaps" and "split" we'll not use at this time.
Flap lets you define moving surfaces, split lets you split a panel into two ("flaps" will be used when we define the elevator).

Hans

# Images

Last edited by hul; May 25, 2013 at 12:35 PM.
 May 25, 2013, 12:33 PM Registered User Switzerland Joined Mar 2003 2,154 Posts forgot to mention. The free demo version of the software only lets you save "flight scenes" under one name "demo.flz". You can rename this in Windows, but next time you save it's demo.flz again. Make sure you don't overwrite things you still need. Hans
May 25, 2013, 01:47 PM
Registered User
Switzerland
Joined Mar 2003
2,154 Posts
we'll have to repeat the same steps to define the horizontal tail, see post#3.

Lower left, "wing" tab: hit "new" button to generate a new wing. I'll call this one tail and set the weight to zero again.

Tab "pos": set the position of the tail. x is to the back, y is to the right and z is up.
The wing's root leading edge was 0/0/0. We'll make the tail 0.8/0/0.07 which is 0.8 behind the wing and 0.07m up (measured wing root leading edge to tail root leading edge). The 3-dimensional view lets you check if things moved the right way.

I won't show the rest of the settings again. I'll use NACA0010 with a 0.17m root and 0.13m tip, 0.04m sweep, no dihedral, 0.3m panel width, 0 incidence, 0 washout.

We still need to define the moving surface for the tail.

Tab "flaps" lets you set it in % of chord or in meters. We'll use that and use 0.05m left and right. Select flap group 1. This colors the surface in, but only in one panel. Select the other panel by clicking it with the mouse in the left picture (or by using the arrows under segment number) and select flap group 1 as well. Now both halves of the elevator are in the same group and can be moved with the same control.

Hans

# Images

May 25, 2013, 02:05 PM
Registered User
Switzerland
Joined Mar 2003
2,154 Posts
Now we're pretty much ready to fly it. But first we want to decide on a design speed. This depends on the mission of course, we'll just pick 30m/s (around 65mph) for now.

We really have no idea if our incidences are correct and where CG needs to go.

Tab "design computations":
select radio button "stabilty meas" and set it to 20%. Then hit the button "start computation" at the lower left.
We're lucky and the thing actually flies, too slow though at only about 23m/s. CG comes to 0.104m.

For the 2nd round we freeze CG by selecting that radio button, set CG to an even number (0.104m, not strictly necessary, but I like even numbers...) and to go faster we need to trim down. Don't want to deflect the elevator, therefore have to change tail incidence. Go to tab "wing", select the tail with the arrows, then go to tab "angle" and set incidence a bit more positive.
Repeat design computations and keep adjusting tail incidence until flight speed is the desired 30m/s (roughly). At +0.5degrees, flight speed is 31.31m/s which is close enough.

Hans

# Images

Last edited by hul; May 25, 2013 at 02:25 PM.
May 25, 2013, 02:21 PM
Registered User
Switzerland
Joined Mar 2003
2,154 Posts
we could now walk away and start checking stall behaviour but I usually try to get the fuselage to fly level and also try to improve accuracy a little.

Tab "design computations: tick the boxes "wake turbulence correction single alignment" and "wake turbulence correction complete alignment" (not sure what the difference is). And display the streamlines by clicking the button under the 3-dimensional view (because they're neat...).
Now re-run the calcs. Flight speed is now 31.66m/s at an angle of attack of -0.359degrees (this angle is between inflow and fuselage centerline, not wing chord). This is very close to level and we could just leave it, but to show the method we'll try to get closer to zero.

Negative angle of attack means that our incidences are too great. We'll therefore reduce wing and tail incidence by the same amount to get the angle of attack to zero. And re-run the calcs.
Setting wing at -0.3 (was 0) and tail at +0.2degrees (was 0.5) result in an angle of attack of -0.04degrees at 31.49m/s.

The current "flight scene" is attached if you want to play with it.

Hans

# Files

Last edited by hul; May 25, 2013 at 02:30 PM.
 May 26, 2013, 03:18 AM Registered User Estonia, Harju, Tallinn Joined May 2012 210 Posts hul, hank you for the tutorial. I tried to use FLZ Vortex a while back, but I really didn't know what I was doing. The interface seemed a bit confusing. So I dropped it and started using XFLR5. At the moment I'm using XFLR5 and AVL. I mainly use XFLR5 to define the plane's geometry and export it to AVL. The reason why I prefer to use AVL is that it's constraint driven. I can set CL=0.7 and CM=0 and get the corresponding alpha and control surface deflections as results. Also, finding the neutral point is a one step process; in XFLR5 you have to run multiple analysis to zero in on the NP. I've probably only scratched the surface of AVL's possibilities so far. Have you used XFLR5 and/or AVL? Would like to know what you think of their features in comparison to FLZ Vortex's features.
May 26, 2013, 03:34 AM
Registered User
Switzerland
Joined Mar 2003
2,154 Posts
I have tried XFLR5 but gave up quickly. Because I find it hard to find the motivation to learn something new when I already know something that seems to do the same thing.... I have the same problem with CAD programs.
Have not tried AVL.

Finding NP is easy in FLZ. Run the analysis and hit the "computed values" button near the center bottom of the screen.

Hans

# Images

May 26, 2013, 04:21 AM
Registered User
Switzerland
Joined Mar 2003
2,154 Posts
we'll now deflect the elevator to fly slower and to see what the stall behaviour is like.

lower left, tab "wing": select the elevator with the arrows
lower right, tab "flap": either enter a "flap angle" or use the slider. Negative angle gives an up deflection, exaggerate to see this in the 3-dimensional view. I used -4 degrees.
Go back to tab "design computation" and re-run the calcs.

Speed is now 12.99m/s at an angle of attack (of the fuselage) of 7.8 degrees.

Navigate back to the wing: lower left tab "wing", select the wing with the arrows. Include cl in the graph (red arrow). You can now see the local cl at the selected location (red circle) and local cl along the wing in the graph (grey arrow) and maximum cl (green arrow). There is plenty of margin before the wing starts to stall (local cl is less than maximum cl).

Hans

# Images

May 26, 2013, 04:37 AM
Registered User
Switzerland
Joined Mar 2003
2,154 Posts
further increased elevator deflection to -5 degrees and got this.

This tells you the calculation didn't converge and give a result. Increasing the maximum number of iteration steps to 30 didn't help. It's trial and error to solve such issues.

I think in this case this is because the elevator is exactly in the wake of the wing (causing numerical issues). Could try to move it up or down or increase the number of panels. I did that for the elevator, went to 6 (from 5) and re-ran the calcs with better results.

Then further increased elevator deflection until the wing started stalling (at -7 degrees elevator deflection in this case). This is indicated in red where local cl exceeds maximum cl (red arrow) and with a warning (red circle). Angle of attack of fuselage) is 12.68 degrees, speed 10.63m/s.

But you see the stall is safe, the outer 2 panels aren't stalled. The plane should just drop the nose and not a wing. Washout shouldn't be necessary in this design.

Hans

# Images

May 26, 2013, 06:27 AM
Registered User
Switzerland
Joined Mar 2003
2,154 Posts
flaps can only be defined per wing segment. Our wing consists of only 2 segments (left and right). We'll have to split these segments or add new segments if we want to add flaps.

We want our flaps between 0.05 and 0.4 meters from the centerline of the fuselage.

This is how splitting works.
Select the wing, enter the y position you want to split the wing at (in the red circle). Then on the lower right select tab "split", select one of the two options "keep chord linear" or "keep trailing edge straight" and click button "split pos Y". Do the same for the split at 0.4m.
"keep chord linear" or "keep trailing edge straight" make a small difference on a wing with washout, where the chord in the center of the segment is a little bit shorter than the average of root and tip chord.
Go to tab "panels" and use panel automatic on all 3 new segments.

The airfoil at these new segments is blended (what's the right word for this?) between the previously entered root and tip airfoils, FLZ calls this "strak".

Go to tab "flaps" to define the flaps. We'll make these 20% chord both sides and add them to group 2 (group 1 was the elevator). Deflect them 30 degrees (positive is down) and check the 3-dimensional view to make sure everything is as it should. Don't forget to neutralize the elevator.

Run Design computations. Speed is now 9.43m/s, angle of attack 10.3 degrees and the wing is stalled between the flaps and just outboard of the flaps, the tips are ok as are the flaps.

Hans

# Images

Last edited by hul; May 26, 2013 at 06:34 AM.
May 26, 2013, 06:49 AM
Registered User
Switzerland
Joined Mar 2003
2,154 Posts

Select the left segment, click button "add L" to add a segment to the left of the selected one, repeat for the other new panel, then go and define chords, airfoils and sweep for all these segments.
Chords will have to be calculated manually or taken from a plan.

This is also the way to build more complicated wings, more than one taper or elliptical, dihedral breaks etc.
Washout is defined for each segment and is the angle between tip chord of the segment and root chord of the center segment. As always, exaggerate to see this in the 3-dimensional view.

That's pretty much it.
There are more functions, but if you got this far, you should be able to figure them out.

Hans

# Images

May 26, 2013, 07:25 AM
Registered User
Switzerland
Joined Mar 2003
2,154 Posts
an example of a plane with and without washout (flaps are not used in both cases).