With all the discussion of tailwheel shimmy recently, I decided to see if I could find formal studies on the problem. And lo, I did.
The first study doesn't apply as much to us, but is somewhat interesting. It is a 2014 study which discusses differences in trailing arm length and wheel height. Worth a glance if you're deeply into the topic, but otherwise not worth the time. That PDF is attached.
The next study is very interesting for us. It's a 1940 study by a German government group (translated into English by NACA) specifically looking into tailwheel shimmy, which I imagine can get pretty destructive in a bigger plane.
It's worth looking the whole thing over, but they basically conclude that caster arm, or trail (the distance between where the wheel contacts the ground and where the pivot axis, if extended, would contact the ground) is the major determining factor in shimmy, with damping being the most important factor for preventing shimmy. We can control our trail for a given tailwheel, within a small range, by changing the angle of the kingpin (with 'negative' or rearward tilted caster increasing trail).
There's discussion of the stiffness of the fuselage, stiffness of the tire (more inflation == more better, from what I understood) and trailing arm length and other stuff in there. It's a very large PDF file that the forum won't accept as an upload, so I've put it on my website:
http://obairlann.net/reaper/aviation/biplane/reference/german-shimmy-study.pdf
The final 1937 NACA study I found is considerably more math-filled, but (based on a quick skim without examining the math) it basically says that friction on the pivot is the way to prevent shimmy. It's also hosted on my website, though I downloaded it from nasa.gov.
http://obairlann.net/reaper/aviation/biplane/reference/naca-shimmy-study.pdf
The summary of what I learned from these three studies (given that I only glanced each of them over without any deep reading) is that, as far as we can affect things with commercially available mounting springs and tailwheels, the pivot angle in and of itself is not actually very important. On the other hand, pivot friction and caster arm (the distance from the wheel's contact patch to the pivot axis's landing point) are quite important. Because the design of some tailwheels allows the caster arm to be shortened noticeably by changing the caster pivot angle, that angle can be a practical factor in changing shimmy behavior.
I hope this is helpful for folks looking into tailwheels and tailwheel shimmy issues.
View attachment 173-M0031.pdf
The first study doesn't apply as much to us, but is somewhat interesting. It is a 2014 study which discusses differences in trailing arm length and wheel height. Worth a glance if you're deeply into the topic, but otherwise not worth the time. That PDF is attached.
The next study is very interesting for us. It's a 1940 study by a German government group (translated into English by NACA) specifically looking into tailwheel shimmy, which I imagine can get pretty destructive in a bigger plane.
It's worth looking the whole thing over, but they basically conclude that caster arm, or trail (the distance between where the wheel contacts the ground and where the pivot axis, if extended, would contact the ground) is the major determining factor in shimmy, with damping being the most important factor for preventing shimmy. We can control our trail for a given tailwheel, within a small range, by changing the angle of the kingpin (with 'negative' or rearward tilted caster increasing trail).
There's discussion of the stiffness of the fuselage, stiffness of the tire (more inflation == more better, from what I understood) and trailing arm length and other stuff in there. It's a very large PDF file that the forum won't accept as an upload, so I've put it on my website:
http://obairlann.net/reaper/aviation/biplane/reference/german-shimmy-study.pdf
The final 1937 NACA study I found is considerably more math-filled, but (based on a quick skim without examining the math) it basically says that friction on the pivot is the way to prevent shimmy. It's also hosted on my website, though I downloaded it from nasa.gov.
http://obairlann.net/reaper/aviation/biplane/reference/naca-shimmy-study.pdf
The summary of what I learned from these three studies (given that I only glanced each of them over without any deep reading) is that, as far as we can affect things with commercially available mounting springs and tailwheels, the pivot angle in and of itself is not actually very important. On the other hand, pivot friction and caster arm (the distance from the wheel's contact patch to the pivot axis's landing point) are quite important. Because the design of some tailwheels allows the caster arm to be shortened noticeably by changing the caster pivot angle, that angle can be a practical factor in changing shimmy behavior.
I hope this is helpful for folks looking into tailwheels and tailwheel shimmy issues.
View attachment 173-M0031.pdf