Rollerons in HPR

Hello,
I just watched a video on the rollerons on the AIM-9 Sidewinder, and was wondering, has anyone here ever built them on a rocket? They seem like a great way to stabilize a rocket in flight and kinda want to build some on a rocket. I think it could be cool.

-Matthew

Ive wanted to do it for years. They need to be fairly heavy to function and they would probably need a gas system (compressed air or CO2) vented over them prior to launch to spin them up because I think it would take too long foe them to spin up after launch.

Cameron Anderson said:

Ive wanted to do it for years. They need to be fairly heavy to function and they would probably need a gas system (compressed air or CO2) vented over them prior to launch to spin them up because I think it would take too long foe them to spin up after launch.

Click to expand...

The real missile traveled at mach 2.5. I think if you have rocket over mach 1, with really high thrust to weight ratio it should be okay, especially of you use a higher surface area on the front of the discs, but the discs themselves have to have mass for the gyroscopic forces to take effect. That is something for a future personal project I think.

You could 3d print the parts and add weight to the disc, or machine the disc and 3d print the rest of it. I might have to try some stuff out and experiment with it. Get a prototype designed. No way to know without testing it, right?

-Matthew

3D printing the surface would be easy, I wonder how it would hold up to the heat and aerodynamic stresses. And the wheel wouldn't even need to be that heavy because of how light the rest of the airframe would be.
I was looking at scrap fins online from military surplus dealers and the costs were insane. I even thought about using thick cerf circular saw blades.

Cameron Anderson said:

3D printing the surface would be easy, I wonder how it would hold up to the heat and aerodynamic stresses. And the wheel wouldn't even need to be that heavy because of how light the rest of the airframe would be.
I was looking at scrap fins online from military surplus dealers and the costs were insane. I even thought about using thick cerf circular saw blades.

Click to expand...

The blades would be bloody deadly if they came loose! Just imagine a circular saw blade spinning at thousands of RPM flying off the rocket. That is a funny image to me.

Was thinking you could mill those discs out of aluminium and 3d print the surface and mount to the fins, and use some small ball bearings for the axel of the disc, and just bolt the modules to the fins, like the real thing. I don't have a 3d printer, but I know someone who does. When I have some free time I will design something. I'm no mechanical engineer, but I will do my best.

-Matthew

Definitely post your results.

I'd love a simple way to stabilize my two-stagers.

Yeah, i will attach the design file. Anyone who wants to print one will be able to. I'm thinking I will design it for a 54mm rocket with 38mm mount. I feel like that is a good mix of size, and power to get over mach.

I was thinking 3d print a mold ; use a proper tiny bearing ; route a thin core from FG or CF sheet ; and cast the remainder from tungsten-loaded epoxy.

I would almost think you'd just need mass along the circumference and the middle could be relatively light. FG plate would be an easy way to start with a circle.

An IREC team used them back when I competed. They were steel, about 1.5" diameter and ~3/16 thick. Spun very nice on display and had neat control surface housings.

Shame it crashed into a rock face, the flight data would have been nice to see

I saw a team at IREC last year that used them. Worked very well. For less work I'd look into grid fins. I've seen three flights with grid fins and each time they rotated and arced less than traditional planar fins for the conditions.

Edward

AlphaHybrids said:

I saw a team at IREC last year that used them. Worked very well. For less work I'd look into grid fins. I've seen three flights with grid fins and each time they rotated and arced less than traditional planar fins for the conditions.

Edward

Click to expand...

And a general reminder to everyone that 7-around-1 tubefins are just lazy gridfins as long as you keep them short (e.g. 2/3-4/5 of their dia.); and they share some of the not-overstable attributes of same.

FalconRocketry said:

The real missile traveled at mach 2.5. I think if you have rocket over mach 1, with really high thrust to weight ratio it should be okay, especially of you use a higher surface area on the front of the discs, but the discs themselves have to have mass for the gyroscopic forces to take effect. That is something for a future personal project I think.

You could 3d print the parts and add weight to the disc, or machine the disc and 3d print the rest of it. I might have to try some stuff out and experiment with it. Get a prototype designed. No way to know without testing it, right?

-Matthew

Click to expand...

The air launched Sidewinder used four rollerons, while the ground launched variant, Chaparral, only used two. Can anyone cite tech references on why only two rollerons were deemed sufficient on the ground launched missile?

AlphaHybrids said:

I saw a team at IREC last year that used them. Worked very well. For less work I'd look into grid fins. I've seen three flights with grid fins and each time they rotated and arced less than traditional planar fins for the conditions.

Edward

Click to expand...

They flew off of my pad. We did spin them up with compressed air just before liftoff.



Doug

Alan15578 said:

The air launched Sidewinder used four rollerons, while the ground launched variant, Chaparral, only used two. Can anyone cite tech references on why only two rollerons were deemed sufficient on the ground launched missile?

Click to expand...

This is all speculation so take it with a grain of salt but a wing mounted missile would probably not have the hardware to spin up the disks before launch, reducing the effectiveness at launch. The ground launched version could have a gas tank and nozzles for spinning the disks to high rpms before launch making two of them enough. Also they would be draggy and heavy and the ground launch version might need help in the range department.

Also any tech documents related to that decision are probably classified because it is military hardware. There might be something buried in the AIAA database somewhere that discusses them conceptually.

ArchitectOfSeven said:

This is all speculation so take it with a grain of salt but a wing mounted missile would probably not have the hardware to spin up the disks before launch, reducing the effectiveness at launch. The ground launched version could have a gas tank and nozzles for spinning the disks to high rpms before launch making two of them enough. Also they would be draggy and heavy and the ground launch version might need help in the range department.

Also any tech documents related to that decision are probably classified because it is military hardware. There might be something buried in the AIAA database somewhere that discusses them conceptually.

Click to expand...

Yes, sidewinders do not spin them up prior to aircraft flying. They are spinning as the airplane is flying prior to the missile launching. And yes for ground based rockets they should be spun up and ground support is required. Here is a shot of my set up about 8 years ago.



But it does not need to be heavy or draggy. Since the flight time is perhaps 30 sec - 1 min the rollerons will stay spun up without being in the air stream.

Also, here are some of the reference documents you mentioned.

Doug

Eyesinthesky said:

Yes, sidewinders do not spin them up prior to aircraft flying. They are spinning as the airplane is flying prior to the missile launching. And yes for ground based rockets they should be spun up and ground support is required. Here is a shot of my set up about 8 years ago.

But it does not need to be heavy or draggy. Since the flight time is perhaps 30 sec - 1 min the rollerons will stay spun up without being in the air stream.

Also, here are some of the reference documents you mentioned.

Doug

Click to expand...

How did those work for you? I had not thought of spinning them up that way beforehand. I have designed the rocket as 4 fins, and clipped delta to resemble the Sidewinder. With the rollerons on the fin tips. 3" Diameter rocket. So far I just have a very crude design. I am working on it little by little with the time I have. I will share it on here, and see if Lab Rat Rocketry can print them for me.

What diameter rocket is that and what motor did you use?
Lovely piece of launch equipment there.

-Matthew

As someone who worked on the AIM-9, I'd love to build up a couple of full sized missiles... With working rollerons. As has been mentioned before, as the missile is attached to the wing of the aircraft, the rollerons start spinning up once airflow past them begins, thus there is no need to mechanically start them. I don't know about ground based AIM-9s, but I don't think it wouldn't take long for them to get spun up even if they weren't given a blast of compressed air.

Had I the funds, I would have purchased a set of fins (complete with rollerons) that were up for sale on Ebay recently, but they were a tad out of my range at the time.

I'm curious about RPM and mass of the spinners...that math is beyond me but I bet a quick rule of thumb exists.

Cameron Anderson said:

I'm curious about RPM and mass of the spinners...that math is beyond me but I bet a quick rule of thumb exists.

Click to expand...

It’s not just mass and rpm, but how the mass is distributed. Google “moment of angular inertia”.

FalconRocketry said:

How did those work for you? I had not thought of spinning them up that way beforehand. I have designed the rocket as 4 fins, and clipped delta to resemble the Sidewinder. With the rollerons on the fin tips. 3" Diameter rocket. So far I just have a very crude design. I am working on it little by little with the time I have. I will share it on here, and see if Lab Rat Rocketry can print them for me.

What diameter rocket is that and what motor did you use?
Lovely piece of launch equipment there.

-Matthew

Click to expand...

I suppose it depends on your definition of 'well' on how well did it work. The rocket was 6" diameter and a M2400 motor. The rolleron were about 2" dia brass gears about 1/4" thick. They did technically stop the roll but it oscillated back and forth like a washing machine. In fact it was so violent that I'm surprised the booster didn't drag separate. If you read the documentation one of them describes that the rollerons must be dampened. Mine were not. And the math gets very difficult, very quickly. As previously mentioned there are a lot of factors including the moment of inertial of the rocket (which includes the mass and how its distributed), distance of the rollerons, their mass, rpms, etc. Finally, this dampening must be set for the exact rocket. Once these parameters change, the dampening might change.

I do not know the rpm's that the rollerons spun up to but I would imagine at least several hundred RPM. The compressed air was a tank from harbor freight at over 100 psi and a solenoid activated air valve that was opened about three seconds before ignition. Lab testing showed they stayed spinning for >2 minutes which is why I chose to not have them on the outside of the rocket to prevent damage on landing.

Here is a video of the launch. I was still using the high speed film cameras at that time. The on board footage starts at about 1:00. Keep in mind that the on board is about 1/7 speed slow motion footage. The oscillations are somewhat amazing.



Doug

Wow, that flutter is intense. Great video though.

The team I saw used spring steel strips to dampen the flaps. Unsure if they calibrated them specifically for their motor/rocket combo or just chose a stiffness that would serve reasonably well for a range of motors.

That is intense, maybe a limited travel on them might help to lessen their effect? The spring steel is a good idea. Much testing is required, and a design is required for that. The gear takes a while in sketchup.

Nytrunner said:

The team I saw used spring steel strips to dampen the flaps. Unsure if they calibrated them specifically for their motor/rocket combo or just chose a stiffness that would serve reasonably well for a range of motors.

Click to expand...

Not sure how spring steel would damp the oscillation. You'd need something like a dashpot, wouldn't you? Something to dissipate energy.

Even though the rollerons are flapping back and forth, they're keeping the rocket remarkably stable in spin. So they're working at least. I feel like a properly tuned spring would slow them down, but (a) it's been a long time since I really did any dynamics or controls and (b) tuning it would be hard and risk getting into resonance. I like a dashpot or an elastomer spring, though fitting a dashpot into a fin would be tricky. Bungee cords, the poor man's dashpot!

boatgeek said:

Even though the rollerons are flapping back and forth, they're keeping the rocket remarkably stable in spin. So they're working at least. I feel like a properly tuned spring would slow them down, but (a) it's been a long time since I really did any dynamics or controls and (b) tuning it would be hard and risk getting into resonance. I like a dashpot or an elastomer spring, though fitting a dashpot into a fin would be tricky. Bungee cords, the poor man's dashpot!

Click to expand...

I think resonance is the issue here, yes? The oscillation of the rollerons is coupling to the oscillation of the airframe.

I wonder if giving adjacent rollerons different lengths of moment arm, so that they would have different fundamental frequencies as they swing back and forth, would mitigate this without having to kluge up some kind of viscoelastic damper?

jlabrasca said:

I think resonance is the issue here, yes? The oscillation of the rollerons is coupling to the oscillation of the airframe.

I wonder if giving adjacent rollerons different lengths of moment arm, so that they would have different fundamental frequencies as they swing back and forth, would mitigate this without having to kluge up some kind of viscoelastic damper?

Click to expand...

Hrm. I wouldn't call it resonance because roll amplitude wasn't increasing. On further reflection, maybe the issue is that the rollerons were too sensitive. With great bearings, a tiny roll in the rocket takes them to their stops in an instant, then the return takes them to the stops on the other side. They just don't seem to have a setting between "off" and "charging rhino". You could perhaps solve that by having a sleeve bearing instead of roller bearings. If it was a sleeve bearing already, lubing it with grease instead of oil might slow them down a little.

You could also probably do a little tuning by reducing the rotational moment of inertia since they seemed to have way more control power than needed.

boatgeek said:

Hrm. I wouldn't call it resonance because roll amplitude wasn't increasing. On further reflection, maybe the issue is that the rollerons were too sensitive. With great bearings, a tiny roll in the rocket takes them to their stops in an instant, then the return takes them to the stops on the other side. They just don't seem to have a setting between "off" and "charging rhino". You could perhaps solve that by having a sleeve bearing instead of roller bearings. If it was a sleeve bearing already, lubing it with grease instead of oil might slow them down a little.

You could also probably do a little tuning by reducing the rotational moment of inertia since they seemed to have way more control power than needed.

Click to expand...

There is certainly some over-controlling happening on that flight. I think if they have their movement limited it can relieve some of those issues. If the offset limit of the control surface is only a few degrees, they won't have as much control authority, so that would effectively dampen the magnitude of the effect, but there would still be the oscillation.

I am still working on the design. I'm thinking I will have it 3d printed and add mass to the wheel. The housing will be 2 parts epoxied together. I want to test:

Baseline performance
Limited travel
Resisted travel

This way I can see how I need to trim it, and what works or doesn't.

More on the design later

-Matthew

jlabrasca said:

Not sure how spring steel would damp the oscillation. You'd need something like a dashpot, wouldn't you? Something to dissipate energy.

Click to expand...

You get the idea. It introduces an extra resistive element to the system (just responding to distance/angle instead of rate like traditional "damping" element)