Active Flight Control

The Rocketry Forum

Help Support The Rocketry Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
I thought about changing that, but I couldn't come up with something that sounded better. Flight control didn't really ring well with me, but might be the best other option. Suggestions would be welcome.
"Direction regulated"? I am sure the collective minds can come up with alternatives.
 
Suppose the control system fails and drives the control surfaces to some maximum unfavorable deflection? Does that have to be stable? That might be unreasonably constraining or at least hard to prove and favors TVC over aero control.
If the size of the surfaces is small enough, in the case of canards, the rocket can remain stable.

FYI my system has the fins at neutral until after burnout. This ensures the rocket is well up in the air before any potential control system issues kick in. It also stops the system being fooled by pegged gyros during boost.
 
I thought about changing that, but I couldn't come up with something that sounded better. Flight control didn't really ring well with me, but might be the best other option. Suggestions would be welcome.
How about 'active control'?

For TVC, as long as thrust/weight is greater than 1, it should be able to lift-off and fly. Maybe not high nor fast but at least lift off the pad and translate.
 
How about 'active control'?

For TVC, as long as thrust/weight is greater than 1, it should be able to lift-off and fly. Maybe not high nor fast but at least lift off the pad and translate.

I like active control.
 
I thought guided amateur rockets were illegal.
Guess not.
What about gyroscopes?
Many drones and RC planes have gyroscopes to make them easier to fly. They will hover or fly level, unless directed to do otherwise.
Might be easy to incorporate into a rocket. Gyroscopes for RC planes and drones are pretty cheap. Very small and light too.
 
What about gyroscopes?
Many drones and RC planes have gyroscopes to make them easier to fly. They will hover or fly level, unless directed to do otherwise.
Might be easy to incorporate into a rocket. Gyroscopes for RC planes and drones are pretty cheap. Very small and light too.
Not "gyroscopes" in the true sense of spinning wheels. But hi-tech sensors that provide the same effect.

The Eagle Tree Guardian "autopilot" (normally used for winged R/C planes) works for vertical guidance. Alyssa Stenberg did an R&D project using it with a canard model, and later also a gimbaled engine model (most times with fins, a few times finless).

https://www.eagletreesystems.com/index.php?route=product/product&product_id=50
Some $25-30 Multicopter controllers the size of a cracker can be made to work for vertical guidance too. But you have to "learns some stuff", where by contrast the Eagle Tree Guardian is mostly plug-and-play.
 
Yes, thank you Boatgeek. Also, I like the comments Mikec made regarding making rules for this.
Please keep in mind that this thread is called "Active Flight Control" and I intentionally avoided the term guidance or any specific method. My intention was to have something to point to in the future if concerns about my projects came up. You know, a reference I could send people to, so I could focus on the technical parts of my projects.

I suggest a new post when something is finalized. I think it would be very useful make it clear that flight control is allowed. This will assure people that they can pursue their project and ask questions about whether they can do something specific. The idea is to actually attract capable people of doing stuff like this.

I suggest the scope be focused on intensions, not specific methods. We should rely most on the normal process of review at any launch. If you mention specific methods there's going to be a ton of questions about methods (e.g. reaction wheels).

It should start out saying that flight control is OK to pursue and perhaps a reminder that rockets should be designed to be safe throughout the flight and stable during boost.

I would encourage people to reach out to their respective club and Prefect/Section Advisor about their project before launch day. Obtain a clear understanding about their project and what's going to be required. We will gain experience with this with no incidents.

So, I'm not a fan of the Guidance During Boost and Coast section except for the part about the RSO considering moving rockets with unproven flight control systems to an increased stand off distance. But, even this could be omitted because I have confidence that this would be done already. Also, I would suggest not having the Guidance During Recovery Section.

No rocket may be flown with the intention of causing harm or injury to any person or object on the ground or in flight. This is the clear intention of SLP F.1 and K.2. All rockets must follow the TSC and SLP, including planning for safe recovery.
I really like this a lot.

Please keep in mind that rockets may have things like TCV alone, TVC at launch then aerodynamic stability. How about using TVC method for a sustainer? TVC rockets can have thrust to mass at liftoff just over 1:1. How about landing under thrust? Methods might be OK at some launches and a complete no-go at others.

Again, actually avoiding specific technical requirements. So, don't like the part about inactive aerodynamic control systems. Too specific in my opinion.

I greatly appreciate the time and effort everyone has made in this thread.
 
How about using TVC method for a sustainer?

Lots of issues. TVC is good for liftoff and slow boost. As it speeds up, aerodynamic forces increase exponentially. If the rocket is stable, the aerodynamic stability will fight the TVC, so the TVC will have less and less control authority. I've seen this occur, including with onboard video. OTOH, if it is a not stable rocket, the TVC is "fighting" to keep it straight, the faster it goes, the worse the battle becomes and the model gyrates or cones and would lose it altogether with enough speed.

It makes a LOT more sense for aerodynamic control for a sustainer (upper stage).

Also, TVC for liftoff is bad if the motor is a high thrust motor that makes the model fly fast.

TVC is most impressive, and most practical, done SLOW with a long burn. Not FAST. Aerodynamic control is best for fast.

I another message, someone mentioned that models with aerodynamic control need to be stable, in case the canard steering quit. Well, I agree aerodynamically controlled models ought to be stable, but simply to avoid massive problems with trying to keep the pointy end pointing the right way. You need incredibly fast servos, and perfect programming of all relevant parameters (including moment of inertia) to even attempt to keep an unstable model flying stable aerodynamically. Great to try that for a college project perhaps, after a lot of experience, but otherwise not a good idea.

But anyway, the problem with the aerodynamic steering going dead is.....not about whether the rocket is stable or not. Actually, an unstable rocket whose aerodynamic control quit, would quickly go unstable and flip around and around and around in the air, and probably burn out in the sky and fall to the ground. A STABLE model whose steering quit, would veer sharply into whatever direction the control surfaces froze at and crash fast, possibly still under thrust.

One time, with sunguidance, after delays, I finally launched the model, and it half-looped into the ground. The onboard battery had gone dead while I had waited for a cloud to go away (didn't NEED the sun to be visible to work, but it was most impressive to fly with the sun out). So, it pitched over at launch and did that aerodynamic half-loop. Lessons learned there were not to let the battery run too long without stopping to swap. AND to confirm the guidance was working before liftoff (I always did a pre-launch test of waving my hand over the sensors, but after the delay I did not repeat that test once the sun came out and it was time to launch it).

That is a lesson learned from R/C Rocket Boosted Gliders (RBG's), wiggle the transmitter sticks to see if the controls work properly. I will note that with other guidance methods, it is not too easy to do a last-minute test before liftoff. But there is some testing that can be done during the loading process (is the TVC gimbal moving the correct direction when tilted?), voltage monitoring, and/or using fresh-charged batteries and knowing what the safe maximum "wait time" is, and not go much beyond half of that before swapping batteries. This is not the sort of thing to "codify" in detail, because various guidance systems are different. More practical to make people aware of things like this, then have some wording generically saying the flier should be sure the system is operating properly, and has sufficient battery power, before launch.
 
Last edited:
There has been a lot of good input here. I deliberately didn't include roll control in the initial writeup because I didn't see either (a) controversy about roll control since it doesn't change the rocket's direction of travel and (b) any particular operational problems that might arise if the roll control system failed. Since the guidance here is to try to reduce controversy and address potential safety concerns, I didn't include it. It could be added, at the cost of increasing the length of the document and maybe raising more questions than it resolves.
 
I really appreciate what you did and I don't want you to think otherwise boatgeek. The main reason I'm suggesting a short version with the main intension without any specific methods or requirements is the number of questions something like this would get. That is, if it was a single thread. You might answer two questions before bed and get up in the morning to find a half dozen more questions and opinions. As threads become longer and more complex people shy away from them.
I put some possibilities out there just for examples.
 
Last edited:
I see the main problem with flying active flight control is that the safety code was never intended to accommodate it. All the established safe practices and hazard assessments are not directly applicable much beyond '...not intentionally cause harm'.

I would suggest that guided recovery is a different topic as it has very different hazards and risks associated with it, worthy of a separate consideration. Also seems fine unless it is 'winged' and autonomous.

On a rocket with active flight control, the safety fan is realistically a circle with a radius equal to the maximum range of the rocket [equivalent to about a 41 to 45 degree launch angle]. Any test range will hold you to that if you were to bring your HPR rocket, active control or not. That we are not subject to that restrictive a requirement for safe distance I would claim is tied to the assumptions in the safety code about how a rocket should be constructed and how it should be expected to behave [along with folks willingness to actually follow the safety code]. Good on us.

If we had a much worse safety record, it would be reasonable to suppose that the safety code would have been updated with new rules to attempt to manage whatever people had been doing that caused so many problems. We don't have such a bad safety record and so I think the safety code has a lot detail not included because there has not been a need for it.

As active controls become more accessible we can expect to see more flights and learn about which systems work well enough, and which are 'unfortunate'. Any designer would be well served by starting modest with a system that can't steer the rocket more than a small number of degrees until that system has a history of performance and reliability. That self-limitation makes the rocket more passively stable than a 'hail Mary' design with a lot of control authority that is a land shark waiting to happen at the first serious fault in the system. Anyone smart enough to design a practical active control system should be smart enough to estimate the size of the land shark it should be capable of doing, and set a 'safe distance' accordingly. For some that will mean that Blackrock is one of the few reasonable places to fly it.

I don't see much that can be effectively put into new rules beyond establishing some rationale for different safe distances and prohibiting some things that are identified as carrying excessive risks. Rules can't keep a one-off active system from failing or being designed inadequately. Restricting all active systems to passive stability under all fault conditions would be desirable but difficult to train every single RSO to be able to evaluate reasonably. Being smart about how much room to give it at the launch is achievable. Treating active control flights similarly to an L3 cert attempt is another candidate, but enough TAPs or volunteers with enough technical depth to evaluate an active control project would still be a gap to fill.

If we as a group can keep it together and be smart about active flights, we should be able to enjoy a thoughtfully developed set of rules and guides for these types of systems developed gradually over time as we learn things. If we can't, then new rules will surely find us quickly.

br/

Tony
 
Trajectory control when it works has the benefit of better being able to stay within the waiver radius at site. When it fails, a failure effect may be a gross violation of the waiver radius, much more so than your average garden variety passive rocket.

How would users and developers of such solutions propose to prevent this failure mode?
 
Trajectory control when it works has the benefit of better being able to stay within the waiver radius at site. When it fails, a failure effect may be a gross violation of the waiver radius, much more so than your average garden variety passive rocket.

How would users and developers of such solutions propose to prevent this failure mode?
Any non-guided rocket can veer off and exceed the waiver radius. How do you propose to prevent that failure mode for ballistic rockets?

How do you propose to prevent that failure mode for clustered rockets that can veer off and exceed the waiver radius if a side motor does not ignite?

As for system failure, can be said for any rocket using electronics for a safe flight. e.g. altimeter ejection instead of motor ejection. How do you propose to prevent that failure mode for rockets using electronic ejection?

How do you propose to prevent that failure mode for reloads, if the flier improperly assembles them (i.e. forgets the ejection charges, or an O-ring error?)

So, you're inventing a special straw horse that ignores how this is handled for other rockets.

There is a balance between strangling innovation, creativity, and new technology, and whether a specific person with a specific rocket has designed, built, and prepped it so it will fly safely..... or not. Regardless of the design or special features.

In the end, no RSO really knows if the rocket that is about to launch has been prepared or designed properly or not. I mean, I know of no launched with "prep monitors" who baby-sit every rocket being prepared, to make sure if everything is right, or not. And some kinds of rockets, I would not trust a flight sim worth a darn to be accurate.

Why would you suddenly apply a blanket rule to guided rockets that might have something go wrong, that is not applied to everything else?

What is tricky is that if a person is not very good at this, then they could come up with a pretty dangerous model. But when I say "this", i'm not specifically talking about guided rockets. I mean clustering, and reloads, and staging, and altimeter electronics, and staging electronics, and air-start electronics, or actual rocket designs such as flying say full sized outhouses and snowmobiles (to mention a couple of disasters by people who usually know how to fly safely). It's not necessarily the KIND of technology or rocket, but whether the builder/flier seems competent to pull it off, or not.

And no I'm not suggesting a "certification" for guidance, unless you are all prepared to have certification for all the other categories I listed above, no grandfathering, then go and certify if you want to fly a cluster, or a stage, air-start, or use altimeter ejection, or use a reload, or fly your own non-kit design, and so on.
 
The thing that's going to prevent a disaster is not going to be a code or safety practice, it's going to be someone saying "no." Someone is going to propose doing something like a loop using a TVC system and someone like cranky Steve is going to have to step forward and say that's not a good idea. BTW, I actually know of a L3 certified person that mentioned doing this. And, I have seen students place a HP (38 mm) motor in an untested TVC rocket with no fins and launch it. And, it really was a L O N G burn motor, like seemed to last forever. I didn't have the call on that one, but I was in a bunker.

John, people need to be told to develop their project slowly with incremental and well thought out steps, test along the way, and be patient. There are people, actually experienced L3 flyers, that want to develop a flight control system and "test" it with a M.

Again, I suggest any new safety practice document be general in nature and not even mention specific methods. It could be one paragraph and be very clear.

If you want, I can post some info on what I'm doing. I'm doing it the right way.
Well, one of the right ways instead of many of the wrong ways I've seen.
 
I saw George‘s post in #24 and always wondered how the powers that be feel about a guided descent.
To me it sounds good to put a rocket where I want it to go, that is to remain onsite when at altitude and then come back to me a safe but close recovery distance away. No harm done to anyone.

When I flew R/C sailplanes, I got to the point where I could get the sailplane to fly right at me and do a full stall landing right into my left hand. Took me awhile to learn that trick. Many times if the ship was too hot, I’d have to step away and let it fly past.

My favorite ship was a Goldberg Gentle lady that I modified to a straight Schumann wing with ailerons, elevator and rudder. That thing flew like it was on rails and once I got the knack of catching it (a rocketry no-no I know) it was a blast. The thing with the glider was if it was too hot, I had plenty of time to change the direction (after all it is R/C) or step aside and control it to a nominal landing just farther away from me. Kurt. Savegnago
 
Any non-guided rocket can veer off and exceed the waiver radius. How do you propose to prevent that failure mode for ballistic rockets?

How do you propose to prevent that failure mode for clustered rockets that can veer off and exceed the waiver radius if a side motor does not ignite?

As for system failure, can be said for any rocket using electronics for a safe flight. e.g. altimeter ejection instead of motor ejection. How do you propose to prevent that failure mode for rockets using electronic ejection?

How do you propose to prevent that failure mode for reloads, if the flier improperly assembles them (i.e. forgets the ejection charges, or an O-ring error?)

So, you're inventing a special straw horse that ignores how this is handled for other rockets.

There is a balance between strangling innovation, creativity, and new technology, and whether a specific person with a specific rocket has designed, built, and prepped it so it will fly safely..... or not. Regardless of the design or special features.

In the end, no RSO really knows if the rocket that is about to launch has been prepared or designed properly or not. I mean, I know of no launched with "prep monitors" who baby-sit every rocket being prepared, to make sure if everything is right, or not. And some kinds of rockets, I would not trust a flight sim worth a darn to be accurate.

Why would you suddenly apply a blanket rule to guided rockets that might have something go wrong, that is not applied to everything else?

What is tricky is that if a person is not very good at this, then they could come up with a pretty dangerous model. But when I say "this", i'm not specifically talking about guided rockets. I mean clustering, and reloads, and staging, and altimeter electronics, and staging electronics, and air-start electronics, or actual rocket designs such as flying say full sized outhouses and snowmobiles (to mention a couple of disasters by people who usually know how to fly safely). It's not necessarily the KIND of technology or rocket, but whether the builder/flier seems competent to pull it off, or not.

And no I'm not suggesting a "certification" for guidance, unless you are all prepared to have certification for all the other categories I listed above, no grandfathering, then go and certify if you want to fly a cluster, or a stage, air-start, or use altimeter ejection, or use a reload, or fly your own non-kit design, and so on.
Chill dude.
 
George is right on target.
I wasn't proposing any new rules or strangulation of innovation of the sort. (if you or anyone read my post that was quoted). I was just suggesting developers consider the new and possible more severe failure modes active systems will have. George's response was a pompous over-reaction, in my humble opinion.
 
Someone is going to propose doing something like a loop using a TVC system and someone like cranky Steve is going to have to step forward and say that's not a good idea. BTW, I actually know of a L3 certified person that mentioned doing this.
Did someone actually suggest looping a rocket? :eek: What were they thinking? ^What could possibly go wrong with that?^
 
Did someone actually suggest looping a rocket? :eek: What were they thinking? ^What could possibly go wrong with that?^
If the math works out looping a rocket under active control is no more or less hazardous than flying straight up.
 
Correct, but the chances of novices getting it wrong (unknown unknowns) or off-nominal flight (a lot of people don't design for this) are significantly higher when attempting such cleverness. Control authorities are required to be higher also, increasing risk.
 
Correct, but the chances of novices getting it wrong (unknown unknowns) or off-nominal flight (a lot of people don't design for this) are significantly higher when attempting such cleverness. Control authorities are required to be higher also, increasing risk.
Any decent altimeter will blow the apogee charge at the top of the loop anyway... :)

Which BTW is a method to somewhat mitigate failed active control system......
 
Any decent altimeter will blow the apogee charge at the top of the loop anyway... :)

Which BTW is a method to somewhat mitigate failed active control system......
I was thinking a self destruction package (actually just a flight termination system), under the control of the RSO, might be required. But in some instances this "decent" altimeter may be sufficient.
 
I guess we are kind of standing by for Steve's response.

John's post #105 made a good point about recognizing the unique potential failures of active control. And, yeah, maybe George's response seemed a bit harsh. I think it might be due in part to the way we perceive questions in written form. George did take the time to respond and provide his opinion. If we didn't have opinions we probably wouldn't be on post #119.

Why would you suddenly apply a blanket rule to guided rockets that might have something go wrong, that is not applied to everything else?

I don't think John was actually trying to do that George. Flyers should test new systems and endeavor to address special failure modes. Just saying, George makes one of the only systems that helps with a major failure mode for multistage flights. So, he's helping us!
Yes, someone did suggest that they could do a loop. And, maybe they could. I don't think it's a good idea, but they are probably capable of this complexity. If they are capable of doing the loop, they are probably capable enough of not deploying on the first apogee. But, LOL if they didn't. So, what are the failure modes of doing a loop? How about not completing the loop, like doing about a 1/2 loop? And, you would still be under power. Yeah, that would not be good.

Also, Over the Top, I got a chuckle that you asked about this and you have personalized the This Way Up sign.

I'm not a fan of canards. Well, on rockets, because on airplanes they can help with improving stability. But, what does even a fixed canard do on a rocket? When I was first thinking about doing active flight control myself, I considered movable canards. I understand the benefits of using forward aerodynamic control surfaces. They are easy to create because you have the room for stuff like electronics and servos and can be placed in a nice compartment. Probably if a commercial system for HP rockets is developed, it would be a canard system. People could just get the right diameter, tune the system, and away they go. But. I don't like them because they are forward surfaces and could go wonky and that would be a problem.

For an example, here's some photos of one of the things I'm doing. It's a 3" rocket with a 54 mm motor mount (think about how much room I have for this). I also have a TVC rocket in the works but, it's for 29 mm motors. btw, I'm not planning on ever launching a rocket that doesn't have fins.

I'm also starting to not like TVC gimbals for solid motor rockets. IMG_2702.jpgIMG_2686.jpgIMG_2738-1.jpg
 
Trajectory control when it works has the benefit of better being able to stay within the waiver radius at site. When it fails, a failure effect may be a gross violation of the waiver radius, much more so than your average garden variety passive rocket.

How would users and developers of such solutions propose to prevent this failure mode?

Easy, The flier/developer goes out in the middle of nowhere and obtains several reliable launches of their system before going to a launch that might have more people around. I’m sure many folks have made “no launch“ decisions in the past. One I had that stuck in my mind is a single deploy “GPS tracker dog” rocket I was going to fly for the first time. Just so happens the prefect brought 40! 4H kids out to the launch for some modroc launching between the HPR stuff. I thought to myself, “NO way, I’ll wait for another day to launch this rocket.”. The kids were running all over the field to recover their modrocs. It was no problem the prefect had control but I didn’t want my first time flight to jeopardize anyone. I had other rockets SD and DD that are consistent performers I could fly so no love lost. Impress the crowd.

Turns out the kids were there for a given time and left. Was down to a few hardcore HPR fliers. Ahhhh, now I can fly this ship!! Oh, I have a picture of it below. So’s I start the nosecone tracker, setup for launch and the prefect pushes the button. I think it was an H or I with motor deploy. Rocket goes up, disappears and nothing. We wait and wait and wait and obviously the thing probably lawn darted. I heard a “ping“ from my GPS tracking/mapping program and there is this one dot that’s down range from the launch pads. I decided, “What the hay, I’ll try to navigate to that “spot” and see what I can see. Sure enough I get out there and the fincan is sticking out of the ground 3 inches from the leading edge of the fins. Long story short is I busted the NC getting it out of the ground and it was hotter than Hades from the vaporizing lithium battery used to power the GPS tracker. The tracker of course was toast and gave up its life for that one “ping” that brought me to the crash site. The rest of the rocket was fine. Got a new nosecone from Wildman and good gosh I love it as it’s transparent and I can see the tracker LED’s flashing on the inside. Moral of the story is: I’m danged glad I did the launch after the 40 kids left the area!!! Bring rockets one knows are reliable to a launch so they can always launch them and save some of the “iffy“ ones to a launch that may not be so well attended.

Oh, that ”black hole” just forward from the fincan is from a forward closure failure of a 29mm Mohave Green motor from another flight. I make danged sure I assemble motors like Aerotech says (except when I’m doing a research load;)).

I try to make sure I put the right O ring where it’s s’pose to go and stuff happens. In that case the prefect pushes the button and there were two green flames coming out of the rocket. One at the butt end and one up the side of the airframe where one can see that black “spot”. The chute was ejected but it was just too low to inflate. Thank heavens for the securely affixed fins. I was going to trash the rocket but I had a semi-rigid endoscope and stuffed it down in and the “failure jet” was 180 degrees from the harness. I shoved/epoxied down a coupler to plug the hole (after I cleaned the residue out with Dawn, a garden hose and a soft brush) and the rocket lives on. Flies fine but I get some funky looks at the RSO table.
Bottom line is if you think it’s not the right time to fly a particular rocket, don’t fly it!

BTW, the fillets on the fins are some ”leftover” Cotronics 4525B (or IP if that’s what they call it now) that I had lying around when I did the build. Really tough stuff and some would argue overkill but why not? Oh, I used the 4525 to make the primary bond to the motor tube too. The heat resistance is impressive. (Don’t want to start a “glue thread/flame”). Kurt Savegnago
 

Attachments

  • image.jpg
    image.jpg
    194 KB · Views: 17
Last edited:
Back
Top