Quantcast

Stabilization and trajectory control

The Rocketry Forum

Help Support The Rocketry Forum:

Richard Dierking

Well-Known Member
TRF Supporter
Joined
Nov 25, 2018
Messages
320
Reaction score
131
Location
Temecula, CA
I believe the following applies to both model and high-power rockets.
I have the referenced sections of NFPA 1122, Model Rocketry that's in the California Fire Marshal handbook. The following is from the handbook. California codes for amateur rockets also stipulate mechanical methods to restrain horizontal movement until aerodynamically stable. I bet many other state codes have similar standards. From NFPA 1122:
4.10 Model Rocket Launchers
A model rocket shall be launched from a stable launch device that provides rigid guidance until it has reached a speed adequate to ensure a safe flight path.

In other words, aerodynamic stability.
This is a bit complicated, but I basically see two groups that are interested in controlled stability. There's the people interested in realistic launches of scale models, for example a Saturn V. Because they are interested in the launch and power phase, things such as the TVC method suit their objectives. Keep in mind that after the power phase, unless the rocket is spinning the rocket is not stable in any way. Then, there are others like me that are primarily interested in control from launch to apogee, and even through recovery. For me, I would like to ground launch rockets with no rod, rail, or tower.
So, do you see the problem; when does a model or HP rocket using something like TVC reach aerodynamic stability?
Do the regulations not apply?
If we can't test new stability methods, how can we prove they work and refine them?
I think there should be a way to allow research flights just like research motors.
I'm curious what others think. Say, if you were a RSO at a commercial or research launch and someone brought a rocket up to you with no fins and equipped with a TVC system or said they were going to spin it up on the launcher at takeoff. What would you do?
 

CoyoteNumber2

Original San Diego High Power Rocketry
Joined
Jan 30, 2009
Messages
607
Reaction score
93
I'm curious what others think. Say, if you were a RSO at a commercial or research launch and someone brought a rocket up to you with no fins and equipped with a TVC system or said they were going to spin it up on the launcher at takeoff. What would you do?
I would put them on the back row and say "Good luck!"
 

georgegassaway

Lifetime Supporter
TRF Lifetime Supporter
Joined
Jan 18, 2009
Messages
4,584
Reaction score
433
4.10 Model Rocket Launchers
A model rocket shall be launched from a stable launch device that provides rigid guidance until it has reached a speed adequate to ensure a safe flight path.

In other words, aerodynamic stability.
Well, it does not really SAY aerodynamic stability, even though that would be what the thinking was. Because when it was carved into stone tablets decades ago nobody considered model rockets with gimbaled engines or even guidance at all.

It says: "reached a speed adequate to ensure a safe flight path".

For a TVC rocket whose guidance and control system WORKS properly, then zero velocity is adequate to ensure a safe flight path. Just as a 10 foot long rail and leaving the rail at 500 mph does nothing to ensure a safe flight path if an unguided finned rocket is aerodynamically unstable.

I did not realize till a few months ago, that Joe Barnard's (BPS Space) TVC rockets use no rod or rail. They are tail-sitters, clamped to the pad until the ignition signal is sent and a servo mechanically releases the model. Very impressive, but you gotta nail that just right, and avoid launching in much wind (he often launches near sunset, often in low wind days).

If I was RSO, I'd want to check to see that it's not likely that the rocket could fall over during ignition, and would really prefer that it would have been successfully flown in private first, otherwise a LOT would depend on how knowledgeable the flier was about their system, to get some more trust in what they are doing, and that the motor is not too high in thrust (some get picky about thrust to weight ratio, which is applicably to an extend for ballistic rockets, but a high TWR for a guided rocket is silly and can be dangerous). Also would want to see it powered up and do a tilt test to see if the TVC was moving properly.

But as for rule 4.10 I would not have a big concern as long as it can't simply fall over when it ignites. Because either the guidance system works or it doesn't. So that's why whether it has flown successfully before would be a big thing to me. If Joe Barnard wanted to fly his and I was RSO, no problem, he's proven it works. If I wanted to do it.... I'd do my first public tests using some type of rod or rail (or tower) first, and not try a tail-sitter until i had it working well and had worked out a good way for it not to fall over. And I'd try a tail-sitter in private first (and later post video so f it worked I'd have proof it's safe for public flying).

I have seen a few cases of TVC models where the gimbaled engine caused some sidethrust to cause the model to want to roll on a rod, or even to to bind on a rail. One fix was to go to two thick rods on each side of the model, with upper and lower lugs larger than normal for those rods to slide easily (like 3/8" lugs, over 1/4" rods). So, it could not roll or bind as it slides up the rods.

As for spinning rockets, I did that a long time ago. A Frisbee with a motor in the center, and a motorized platform to spin it up before launch (tricky part was the electrical contacts for ignition). The motor simply plugged about one inch into the top of a tube, friction fit. At ignition the motor popped off the top and the spinning frisbee flew.

Since you jump from mentioning model, and HPR, I will say this. If it's meant for HPR, test it with a small model first. You can learn more quickly at less expense, and have many other places and times you can test from. With less consequences than from a "first flight" of a TVC HPR rocket that has major flight control problems. Also, as I said, high thrust engines are not good for TVC. You need a long-burn engine, boosting sorta slow, for good flights using TVC. Such as the F10 or G12, 8 second burns for models weighing around 1.5 pounds or so. There's no really good endburning HPR engine suited for TVC beyond H power.
 
Last edited:

jadebox

Roger Smith
TRF Sponsor
Joined
Jan 18, 2009
Messages
5,711
Reaction score
203
In the case of a TVC rocket, "a speed adequate to ensure a safe flight path" is essentially zero.
 

Richard Dierking

Well-Known Member
TRF Supporter
Joined
Nov 25, 2018
Messages
320
Reaction score
131
Location
Temecula, CA
So, George got a "like" from me because he mentioned that the NFPA code I mentioned was archaic, maybe obsolete. The believe that's the problem. Of course, that's easy to think now. My fault for specifically mentioning aerodynamic stability and that's what people focused on. But the also says "stable launch device" and "rigid guidance."
It doesn't say send them to the back row, but I admit, I've done that with some rockets people have presented to me. Anyway, if you read the code, it's pretty clear what is required. It just doesn't say rod, rail, or tower. But, yeah, it's pretty obvious.
Maybe there should be some kind of distance requirement for research flight rockets.

California regulations are even more specific about restricting horizontal movement during launch and include HP rockets but I decided to use a national standard that would apply to everyone. If you ever want to see this code right in front of you, have some kind of significant incident with something like the TVC method. So, let's absolutely prevent that.

I understand that most TVC method rockets are model sized rockets and long-burn motors, but they could be larger. And, how about a rocket that uses control surfaces like movable canards? Even if the rocket has a traditionally proper CP/CG relationship and the control surfaces don't work correctly, say because the system is not tuned correctly, this could also be a problem.

Codes need to keep up to date and be appropriately revised. People are creative and can make important advancements if they allowed to do so.

I'm even thinking that there should be a minimum certification level for research flights.

I think it's important for people to know I'm not a TVC hater btw. I have one of BPS Space's kits and building it. I'm trying to think of ways to validate it.
I also understand the reasons for the realistic/low velocity flights. In the past, I've made a lot of effort to make some HP rockets look realistic with clear fins. So, I understand. It's just not what I'm personally interested in now. So, I support the research but not necessarily the business model.
 

georgegassaway

Lifetime Supporter
TRF Lifetime Supporter
Joined
Jan 18, 2009
Messages
4,584
Reaction score
433
I'm even thinking that there should be a minimum certification level for research flights.
What do you mean? One would have to be level 1 or level 2 to even try a model rocket with guidance? No freaking way. And that's bass-ackwards, one should really learn guidance with smaller models first before going to anything big.

BTW - we have had self-guided models rockets since 1988 (sun-seeking). And piloted model rockets (R/C Rocket Gliders) long before then. It is certainly easier to do guided rockets using nose fins, than to do gimbaled engines. And for fast-flying models, nose fins (or aerodynamic control surfaces) make a hell of a lot more sense (of course the CG has to be ahead of the CP).

 
Last edited:

Richard Dierking

Well-Known Member
TRF Supporter
Joined
Nov 25, 2018
Messages
320
Reaction score
131
Location
Temecula, CA
OK, yeah, requiring a minimum cert level is probably not a good idea. Just thinking it through a bit with the help of my RF friends and throwing some ideas out there. Don't mind at all if someone says "no-way."
Wondering when the next revision of NFPA 1122 and 1127 is planned. Maybe Steve or another TRA director could provide some input on this subject?
What do you think about having requirements for research flights that use experimental types of stabilization just like there are research motor codes?

And, George, wow, you do Sun-seeking flights!!! I want to do Sun-based roll control. So, the two photocells are located looking out not up and intended for morning or afternoon flights. Two small control surfaces between the fins on opposite sides. Do you have info for your set-ups on RF or somewhere else? Are you a programmer? Sorry going OT here, but I can't help myself. It's so cool meeting someone that does this.
 

jadebox

Roger Smith
TRF Sponsor
Joined
Jan 18, 2009
Messages
5,711
Reaction score
203
Why are you proposing new rules? As George points out, the current rules have worked fine for years.
 

Richard Dierking

Well-Known Member
TRF Supporter
Joined
Nov 25, 2018
Messages
320
Reaction score
131
Location
Temecula, CA
If everyone including TRA and NAR are cool with the present regs and flying TVC and other active stabilization rockets at sanctioned launches it's fine with me. I would like to hear we are covered, but if there is silence that's fine as well. I'll go with the status quo with no questions. I'll be in a bunker when these flights go, but no questions. :)

Thanks for posting the video George, you brightened my day.

(Now, watch how fast I move on...)

The thing about researching and doing active stabilization flights is that the people that seem to know how to do it don't provide their programing. I understand why someone that develops this as a business and spends a lot of time in development would not open up the application (the so-called black box). But, some of us that are very interested in this type of thing and actually quite good at building rockets are not programmers.
There are lots of Youtube videos on how to do various things and include the programming, but when it comes to using gyro/accelerometers to stabilize rockets there's nothing available. Someone that has this knowledge must be out there that's willing to open source the code and help other people do this.
 

Joekeyo

Well-Known Member
TRF Supporter
Joined
Sep 12, 2019
Messages
348
Reaction score
243
Location
Sun City Center, FL
Or you could just drive across the state line to Nevada. Drive for 2 hour after you see the last house then do your thing. The only person could get hurt is you.
 

Richard Dierking

Well-Known Member
TRF Supporter
Joined
Nov 25, 2018
Messages
320
Reaction score
131
Location
Temecula, CA
I can't say I've ever done anything like that. :)

Saw your quote Jeokeyo. We could use a dose of Firesign Theater now. That, and maybe Pink Panther.
"Does your dog bite?"
 

georgegassaway

Lifetime Supporter
TRF Lifetime Supporter
Joined
Jan 18, 2009
Messages
4,584
Reaction score
433
And, George, wow, you do Sun-seeking flights!!! I want to do Sun-based roll control. So, the two photocells are located looking out not up and intended for morning or afternoon flights. Two small control surfaces between the fins on opposite sides. Do you have info for your set-ups on RF or somewhere else? Are you a programmer? Sorry going OT here, but I can't help myself. It's so cool meeting someone that does this.
The Sunseeker project was in 1988. No "code". Two photoresistors in series (per axis) to produce a varying voltage, fed into a simple 555 timer based servo controller circuit.

In the video I posted, the first two of the Cineroc flights were roll control. It over-controlled badly on the first flight, literally spinning the payload section. Second flight with the control surfaces trimmed down went better, but the ballistic rocket ended up pitching over towards the sun, so it lost roll orientation.

Sunguidance actually will fly just about vertically on an overcast day.

More info on my website: http://georgesrockets.com/GRP/RandD/Sunguidance.htm

For a rocket that can fly vertically regardless of the sun, but more "plug and play" than programming, the best thing I know of is the Eagle Tree Guardian model plane autopilot. In 2D mode, in a plane, it will keep the wings level and the fuselage level. You can fly the plane as normal, but let go of the stick, to neutral, and it will automatically bering it to straight and level. And.... it works the same way to fly a rocket vertically.

Alyssa Stenberg did an R&D project using it 7 years ago. First for aerodynamically controlled D & E powered models, and then for a 4" diameter F10 powered gimbaled engine model. The Eagle Tree Guardian worked great for that. Video of her R&D presentation that includes videos of some of her flights.


I haven't done anything using my own coding. I HAVE had some experience using a multicopter type flight controller, in airplane mode, to get a model plane to go from 100% manual horizontal flight at high speed, to at the flip of a switch automatically pitch up and try to climb vertically, till it lost speed and stalled out (electric motor thrust less than mass of the model). I had intended to use it for a vertically boosted rocket but dropped the project. The attractiveness of it was the relatively low cost of the controller, about $25-30. It was programmed by a computer application where all the settings and parameters were selected. And it was using a model airplane autopilot option, in order to control servos for pitch and yaw. Because otherwise most controllers are set up for multicopters and only vary motor speed for 3 to 8 propellers for control of a multicopter, can't control servos for steering vertically in that mode. But I never quite felt comfortable with it, there is a lot to learn and there was one mysterious crash I had when the plane acted weird, never knew if it was the software or a R/C glitch or wiring having a problem. So, between the Guardian and multicopter-type controllers used in airplane mode, I'd suggest the Guardian as being easier to learn and get to work well. Though it's been a few years since I tried adapting a flight controller, maybe things have improved with the software (or maybe it was never the controller's fault, I just didn't have time to work it out and moved on to other things).

I certainly did learn a lot about using multicopter controllers for my own quadcopters. These keep themselves perfectly level when not steered by me to maneuver.
 
Last edited:

Richard Dierking

Well-Known Member
TRF Supporter
Joined
Nov 25, 2018
Messages
320
Reaction score
131
Location
Temecula, CA
Thank you for posting the video, I enjoyed watching it and remarkable that this was 7 years ago. Shows how creative young people can be when they are free to pursue stuff.

This was my main point; we shouldn't be locked into some old regulation. Authorized research flights should be possible. Maybe it's a matter of trust?

I haven't been much of a fan of the movable canards I've seen. I like the idea of control surfaces between fins like I'm creating because they are somewhat protected from landing damage, and actually add aerodynamic stability in the neutral position. My rocket is designed to actually be aerodynamically stable (at appropriate velocity ;-) with a low roll freq.
However, something like this would be even more difficult to create for a minimum diameter rocket. My rocket is 3" with a 54 mm mount. So, not enough room for a conventional servo that would work. I'm using micro-sized electric motors.
Here's a photo of my two control surfaces. Small, but keep in mind this is for a HP rocket using long-burn K-impulse motors.

IMG_2611-1.jpg

And, I have an electronics bay set-up with a two photo cell array. I could post a couple photos if anyone was interested. But, I'm still in the same boat (when it rolled over), I don't have the expertise to create a program that will work. I would actually prefer using an inexpensive 9 axis breakout. But, these are very difficult to integrate and create a stable platform that will not drift like crazy. I guess I could plug away at this even if it takes years.
I haven't searched for about 6 months so maybe something new is available?
Any suggestions?
 

Alan15578

Well-Known Member
Joined
Feb 11, 2018
Messages
385
Reaction score
97
If everyone including TRA and NAR are cool with the present regs and flying TVC and other active stabilization rockets at sanctioned launches it's fine with me. I would like to hear we are covered, but if there is silence that's fine as well. I'll go with the status quo with no questions. I'll be in a bunker when these flights go, but no questions. :)
Someone else can address the legal insurance questions. Unproven or experimental flights should first be flown in seclusion, e.g. out behind the barn, not at a sanctioned launch. If i was RSO, I would allow the BPS guy to fly at my launch because I know his stuff has flown flown successfully. I would not allow an unknown rocketeer with an unproven active stabilization system to fly.

The thing about researching and doing active stabilization flights is that the people that seem to know how to do it don't provide their programing. I understand why someone that develops this as a business and spends a lot of time in development would not open up the application (the so-called black box). But, some of us that are very interested in this type of thing and actually quite good at building rockets are not programmers.
There are lots of Youtube videos on how to do various things and include the programming, but when it comes to using gyro/accelerometers to stabilize rockets there's nothing available. Someone that has this knowledge must be out there that's willing to open source the code and help other people do this.
Hey Jim, what's the poke to turn this C64 into a WMD? Doing things, and even programming, is not as simple as many people like to think. I don't think you have the skill set required to do what you want to do, and you may be too old to acquire that skill set. Yukon has a similar interest, youth and much potential, but has not shown any serious interest in acquiring the necessary skill set. The BPS guy has done very well also lacking the complete skill set, but a big part of his skill set is showmanship, the ability to attract funding, and getting help from industry experts.

I'm surprised that no one has reverse engineered the Guardian yet. I suspect that most "code" is kept proprietary because because making everything available would reveal how unsophisticated the product is. No YouTube how-to? Great, figure it out and post your own.

Me? I have a Masters Degree in Aerospace Engineering and years of experience in missile G&C. I'm not inclined to let the genie out of the bottle, but I might offer some encouragement or a clue here and there. I'm more interested in seeing what a hobbyist, or even a terrorist, can do without expert help. And there is something to said for the "Just do it!" approach. You could get lucky. I'm also old and my skill set is lacking. I'm somewhat familiar with TRGs and RLGs, but I know nothing about the mems sensors, or quadcopter control, and my limited electronics skills are antiquated.
 

georgegassaway

Lifetime Supporter
TRF Lifetime Supporter
Joined
Jan 18, 2009
Messages
4,584
Reaction score
433
There's been active guidance models flown at NAR and TRA launches as far back as 1988. The Sunguidance project won R&D at NARAM that year. The project report was printed in TRA's HPR magazine in 1992 , IIRC, and the NAR webstore (NARTS) sold a copy of it for many years, but I no longer see it. In 1989 I used sunguidance for a gimbaled engine project, which also won NARAM R&D (had a number of crashes due in part to overcontrol and in part from Sunguidance not being very suitable for the launch phase - best flight it ever had was on an overcast day).

Dozens of people have done guided models thru the years, and even more so in the last 4 or 5 years with Arduino and other type of programmable flight computers. On of the most impressive early ones, about 1990 to 1992, was Dave Gianokos' team who developed a "real" spinning gyroscope based system, using a magnetically levitated piezo disc spun at high RPM and IR sensors to detect the tilt angle of the disk as the rocket tilted off vertical. He gimbaled the engine. IIRC, that was the first successful gimbaled model to fly vertically.

After that, then people began using solid state sensors to detect tilt (Cellphone and Wii type spinoffs), and custom made programmed flight controllers. Some of these flown at NAR and TRA events

When Alyssa did her Eagle Tree Guardian R&D in 2013, she did a Flight demo at NARAM to show how well it worked.

Here's a video from the Apogee Components youtube channel. D12-0 to E9-6.
The image of the NAR flag when it lands is the ultimate proof of NAR approval. :)
It did not begin to pull up to vertical till after staging, when it got enough velocity for the nose fins to be effective enough. That 45 degree launch was the ultimate test of how well it worked. She had it vertical on the pad when turning the Guardian on, because unlike more modern systems, it had to "learn vertical" every time it is turned on. Then she titled it for launch. I will note that the NAR has an R/C Rocket Boosted Glider code, that allows for models that can be guided on boost, to be launched at 45 degrees, so that's why she was allowed to launch less than 60 degrees. And the RSO had no objections (other than pointing in an uninhabited area just in case), since she had already proven in her presentation that it worked, including tests at that launch angle. My most fun way of doing sunguidance flights was 2-staged, angled oppsosite from the sun, so te rocekt had to come up to vertical, then past vertical, then lock onto the sun. As seen in some of the fligjhts in that earleir video.

Joe Barnard (BPS Space) has made presentations about his guided rockets at NARCON for the last 3 years, IIRC. Here's this years. Check out his channel for his 2019 and 2018 presentations.

So, doing guided models are not a problem with the NAR or TRA. Long as they are designed to fly "up", and not to try to hit an airborne or ground target. Beyond that, the only remaining issue that comes to mind is that anyone who tries flight path guidance needs to learn a lot in how to make it work successfully. and there may be crashes along the way, Joe Barnard had a few crashes as he began to learn, but he's also done nearly all (maybe all) of his testing in private. I was fortunate that sunguidance never had a crash except for stupidity (the backwards engine in a 2-stage flight in the sunguidance video, and allowing a partially used battery ot go dead before liftoff after too many launch delays). But sure had several crashes when I tried gimbaled engines as suinguidance would cause it to veer towards the sun at liftoff (as i said, best flight was on an overcast day when it flew near vertically).

One beauty of joe Barnard's BPS space guidance system is how he has designed an App for it, to allow the modeler to tune the guidance sensitivity and reaction to the specific model. Using a very easy way to determine the moment of inertia (https://en.wikipedia.org/wiki/Moment_of_inertia) without doing any math and enter the data into the App. Stuff I could only dream of in 1988, where Sunguidance had a lot of forgiveness for aerodynamic control. But bit me in the butt in 1989 for the gimbaled flights and I began to realize how nice it would be to have an active computer in the loop to determine hoe much servo deflection was needed, to avoid the overcontrol issues I had. Though if I'd done more testing I'd have dialed down the response a lot more than I did. But Joe's requires no trial-and-error to at least get enough into the ballpark to not crash (and more likely to be pretty close the first try).

So the big overlooked feature of Joe's system is not what is in the rocket, but his cellphone based APP for configuring the onboard flight controller to begin with. He's spent a LOT of time and money developing his system, so nobody should expect him to make all his coding available for free. He has literally learned EVERYTHING by himself. He's a music major. No classes in engineering or electronics or coding. Incredible.

But some others who have dabbled in rocket guidance may have made theirs available. Or might if asked. But I don't know anyone else, just have seen various Arduino-type TVC rockets on Youtube. I'll suspect some have used a lot of lines of code from Ardupilot type open-source software and made revisions. Too complex for me, on a scale of 1 to 10 in Arduino, I was at best a 3 (maybe a 1 now, since it has been so long since I was working on some simpler projects. A lot to re-learn from around 6-7 years ago)
 
Last edited:

Richard Dierking

Well-Known Member
TRF Supporter
Joined
Nov 25, 2018
Messages
320
Reaction score
131
Location
Temecula, CA
Thank you Alan and George, you have given me a lot to consider. I think I may be going in too many directions at once ;-) and need to step back a bit and start taking this simple to complex. I'm going to focus on the model rocket using the BPS Space system first. I've watched all the presentations that Joe Barnard has given at NARCON and follow BPS Space on FB. I'm going to post under another subject because I have some specific questions about using the BPS Space system. Hopefully, I'll receive some answers and help others that have similar projects.

And, this thread still may be valuable. We'll see.
 
Top