Fokker D VIII Spiral Stability

[DLC]

I have a scratch built Fokker D VIII. I started with Earl Stahl plans for a 54" gas powered free flight, scaled them down to 36" span, took out the dihedral and added ailerons. It has a stong tendency to go into a spiral in turns. The only way I can make a slow turn is to bank it in the direction of the turn, then almost immediately reverse the ailerons and hold them that way through the turn. Just returning the ailerons to neutral wont do it. Any suggestions as to how I can improve the spiral stability without departing too much from scale? The D VIII has a big flat plate between the wheels which I have at zero incidence relative to the horizontal tail. Would adjustment of the incidence of the flat plate help? Despite the short nose, the C.G. came out O.K. I can vary it from about 25% chord to about 30% chord by moving the battery. Changing the C.G. through that range doesn't make any significant difference in the spiral stability.

[JetPlaneFlyer]

A few ideas:

Add dihedral (sure fire solution)
Replace the wheels with open spoked items (the big solid wheels so far below the CG will be de-stabilising)
Use the old free flight trick of filling in between the cabane struts with clear acetate

Assuming the model has a rudder; what is the roll response to rudder input? Does the model roll with the rudder induced yaw as would be expected in a stable model, or is the roll in the opposite direction, or perhaps rudder input produces no roll response at all?

If the response to rudder is a 'conventional' roll in the same direction as the rudder input (e.g. right rudder gives right bank) then reducung the size of the fin/rudder should help spiral stability.

Steve

[vintage1]

Odd. Lowering the CG might help.

There should be SOME dihedral..normally the wing TOP is flat but due to spar taper, the bottom surface of the top wing has a little dihedral.

Is the CG correct? too far forward might cause that.

[JetPlaneFlyer]

Good point re the too far forward CG...
I ran the Earl Stahl plan geometry through a CG calculator and your CG looks too far forward even at 30%. The Stahl design has a large horizontal stab (due to it's free flight origins) which leads to a rearward Neutral Point. The CG works out at 37% while still retaining a healthy 10% static margin.

The fin/rudder on the Stahl plan also looks like it's been increased quite a bit from scale which could be part of the problem ref my original reply.

Steve

[DLC]

Thanks for the help everybody. Vintage, you are right. The lower wing surface has a little bit of dihedral due to the spar taper. JetPlaneFlyer you are right about the vertical fin-rudder size. I hadn't noticed it before, but when I compared the vertical tail surface in another plan I have with that in the Stahl plan, the tail in the Stahl plan is significantly larger. I calculated the center of the side area (using ACCalculator) and plugged the result into a method in Lennon's R/C Model Aircraft Design for estimating spiral stability and the results say the plane is VERY spirally unstable. JetPlaneFlyer, based on your remark about using a clear plastic filler, I did another estimate assuming I filled in between the front wing mount struts and it didn't help much. However, I then tried filling in between the landing gear struts and the results say the plane will have good spiral stability. I realize that appears to contradict your suggestion of using spoked wheels, but it is such an easy fix I plan to try it while I decide if I want to go to the trouble of replacing the tail surfaces with scale-sized surfaces.