I could use just a little guidance

[BABAR]

Congrats. Hope your results are that great when you get to your max motors and three stages!

 

[JimJarvis50]

Congrats. Hope your results are that great when you get to your max motors and three stages!

I don't think it would make any sense to put this on the top of a three stager. It wouldn't work that well, and there would be a weight penalty for the entire flight.

Just in case you missed it, though, I did use the system on the three stage flight at Balls in 2018. It was located between the first and second stages, and then was separated after the first-stage coast (before lighting the second stage). It actually worked quite well. With respect to the distance of the recovery, if you consider the angle that the rocket would have been at after the coast versus the actual angle, it was the difference between a recovery 20 miles out versus the actual 3 miles.

Jim

 

[BABAR]

Thanks for reposting, very cool. Especially was interesting how the rocket started tumbling but still ascending from 140k to 175k Apogee, I guess because it there was so little atmosphere at that point that the only effective forces were kinetic energy (momentum) vs gravity, minimal atmosphere so minimal drag, so orientation of the rocket didn’t matter.

I guess your pitch/yaw/roll control surfaces have the same issue, they become less effective per square inch of surface area the higher the altitude.

 

[RGClark]

I was able to do the flight today (the M3100 to M745 test rocket flight). It went quite well. I did the boost with roll control only (since yaw/pitch wouldn't do much with the longer stack anyway), separated the stages at 3 seconds after a 2 second burn, and then turned on yaw/pitch at 4 seconds. You can clearly see the yaw/pitch control come on in the video...
Jim

PS - The spin can got a thorough workout on this flight. Lots of wind shears on the way up.

A very pretty rocket. Thanks for that.

BTW, can you explain again what is the advantage of a spin can rather than having just the entire stage spin?

Bob Clark

 

[Charles_McG]

I can answer this one. Jim is using active correction- if he spun the whole booster, the pitch and yaw axes would spin with it. That would complicate controlling those axes - to say the least. It might not be possible. So he controls roll, which keeps the pitch and yaw axes pointed mostly in constant directions. That means he can then control them when the time comes.

But experimentally, he discovered that the control vanes, when controlling, kinda torque the slipstream of air flowing around it so it doesn’t hit the aft fins symmetrically. That caused a reverse roll. So he made the aft fins spin freely in roll. They still function to keep the rocket stable in pitch and yaw, but are free in roll so it doesn’t cause the control reversal.

His unit looks a lot like the S19 sounding rocket boost guidance module. There has been a recent picture posted on YORF of an unidentifed Honest John-Nike rocket with, perhaps, an S19 test unit on top. That version appears to have strakes immediately aft of the control vanes, and I wondered if that was an alternative solution.

 

[RGClark]

I don't think it would make any sense to put this on the top of a three stager. It wouldn't work that well, and there would be a weight penalty for the entire flight.

Just in case you missed it, though, I did use the system on the three stage flight at Balls in 2018. It was located between the first and second stages, and then was separated after the first-stage coast (before lighting the second stage). It actually worked quite well. With respect to the distance of the recovery, if you consider the angle that the rocket would have been at after the coast versus the actual angle, it was the difference between a recovery 20 miles out versus the actual 3 miles.

Jim

Pretty impressive you were able to get ignition at high altitude. As you know, that has been a problem with other amateur teams.

An additional university team recently had a failure at a high altitude launch due to this:

https://m.facebook.com/story.php?story_fbid=2929589837068469&id=245686032125543

Anything you would want to discuss?

Bob Clark

 

[JimJarvis50]

A very pretty rocket. Thanks for that.

BTW, can you explain again what is the advantage of a spin can rather than having just the entire stage spin?

Bob Clark

These are different ways to solve the same problem. Spinning a rocket helps to reduce dispersion but is not used with active stabilization. Spinning introduces a new set of difficulties.

For active stabilization, the spin can reduces the effect of control reversal, as Charles mentioned. Here is a video that really showed what happens with control reversal. The rocket was spinning the opposite of the desired direction during the boost, but then roll-stabilized as soon as the lower booster fins were separated. I started building the spin can as soon as a realized what had happened on this flight.

Jim

 

[JimJarvis50]

Pretty impressive you were able to get ignition at high altitude. As you know, that has been a problem with other amateur teams.

An additional university team recently had a failure at a high altitude launch due to this:

https://m.facebook.com/story.php?story_fbid=2929589837068469&id=245686032125543

Anything you would want to discuss?

Bob Clark

I'm not familiar with this particular launch. I know TU Wien had a failure on the boost of their two-stage around this time?

People say I get lucky that my upper stages light. Hmm. My interpretation is that I make arrangements for them to light, and then they do.

Jim

 

[Steve Shannon]

Pretty impressive you were able to get ignition at high altitude. As you know, that has been a problem with other amateur teams.

An additional university team recently had a failure at a high altitude launch due to this:

https://m.facebook.com/story.php?story_fbid=2929589837068469&id=245686032125543

Anything you would want to discuss?

Bob Clark

Bob,
In 2018, their failure to ignite the sustainer had nothing to do with high altitude. They had a wiring problem. That’s briefly mentioned in this report which discusses their 2019 failure.
https://spaceteam.at/2019/10/31/final-report-the-hound-2019/?lang=en

 

[Reinhard]

Bob,
In 2018, their failure to ignite the sustainer had nothing to do with high altitude. They had a wiring problem. That’s briefly mentioned in this report which discusses their 2019 failure.
https://spaceteam.at/2019/10/31/final-report-the-hound-2019/?lang=en

The statement Bob Clark is referring to, is also from 2019. It was written right after the launch without much data at hand, but already then "upper stage didn't ignite" was admittedly an euphemistic way to put it, considering we already knew that we lost telemetry on the sustainer.

Reinhard

 

[RGClark]

Jim may be the only amateur in the world able to do successful ignition at high altitude.

It’s an important accomplishment for amateurs to reach, for reasons I have discussed before.

Robert Clark

 

[JimJarvis50]

Jim may be the only amateur in the world able to do successful ignition at high altitude.

It’s an important accomplishment for amateurs to reach, for reasons I have discussed before.

Robert Clark

Bob, I do a few things to help get motors lit at high altitude. These include things like propellant selection, igniter strategy, etc., most of which I have posted here on this forum. I know I can get the motors to light, albeit with a few seconds of delay, which I account for in the flight profile. Perhaps the actual correlation is that I often have rockets at high altitude, in a position to be lit, with tested electronics to accomplish that?

Techniques to light motors properly at high altitude are not secret so far as I know. Any serious university team would have the manpower, time and resources to do the required testing for their motor style. They should do that and not worry about what I do.

Jim

 

[JimJarvis50]

I had a chance to fly the stabilization system again last weekend. It was an L-1050 to about 7,800 feet. We set up the rocket at a tilt of about 7°. I wanted to set it at 10°, but it was pointed into a pretty brisk wind, and setting a higher angle than we did just didn't pass the smile test (and at 10°, the pad probably would have fallen over). Looks like the rocket weathercocked to around 20° and then corrected to vertical after 5 seconds.

The flight does show some roll early on. I think that when the canard angle is high, it is easy to induce some roll if things aren't aligned perfectly. Then, there is relatively much less control action available for roll control, so it's difficult to completely eliminate roll. From that perspective, I think the system worked pretty well.

Here's a short flight video.

Jim

 

[TonyL]

Hi Jim,
Very nice result. Your initial roll looks like it could just be a typical transient response [saturated or not]. If it is saturating like you suggest, reducing the control travel limit for pitch a little would be expected to improve it and should get you closer to the sweet spot for overall control balance for this rocket. If that does not help then raising the roll gain a little [especially if you have recently reduced it] would be another candidate.

br/

Tony

 

[JimJarvis50]

Hi Jim,
Very nice result. Your initial roll looks like it could just be a typical transient response [saturated or not]. If it is saturating like you suggest, reducing the control travel limit for pitch a little would be expected to improve it and should get you closer to the sweet spot for overall control balance for this rocket. If that does not help then raising the roll gain a little [especially if you have recently reduced it] would be another candidate.

br/

Tony

Actually, there may be another variable at play. If you look at the video (and other recent videos), it appears that the spin can locks up under boost. You would expect it to be turning at some point during the burn, but it doesn't. Then, at burnout, the spin can starts to spin, even though the velocity is still high. I would speculate that the fin can is drag separating from the air frame such that the downward force is reduced to the point where the fin can can spin. At least I didn't spend many $'s on the spin can! So, it is possible that the spin can is just acting as a set of fixed fins during the boost, and that the roll at that point is caused or confounded by the action of the canards. I'm not sure how to sort this out, but I think I need to find some metal bearing surfaces.

Jim

 

[Chad]

Actually, there may be another variable at play. If you look at the video (and other recent videos), it appears that the spin can locks up under boost. You would expect it to be turning at some point during the burn, but it doesn't. Then, at burnout, the spin can starts to spin, even though the velocity is still high. I would speculate that the fin can is drag separating from the air frame such that the downward force is reduced to the point where the fin can can spin. At least I didn't spend many $'s on the spin can! So, it is possible that the spin can is just acting as a set of fixed fins during the boost, and that the roll at that point is caused or confounded by the action of the canards. I'm not sure how to sort this out, but I think I need to find some metal bearing surfaces.

Jim

the R/C car hobby uses very good (and small) metal ball bearings if you want to incorporate those to your spin can to keep it from binding under acceleration. Edit: i just realized they probably won't help at all. The motor is where the bearings would be hah

 

[UhClem]

The servo control code in MatrixPilot is designed in a way that will induce unwanted motion. That is because it sends out the servo control pulses one at a time. That delay means that opposite fins are not always working together which will add roll when only pitch or yaw is commanded.

Roll corrections of course result in unwanted pitch and yaw motion.

It isn't that difficult to send the servo pulses out simultaneously.

 

[JimJarvis50]

the R/C car hobby uses very good (and small) metal ball bearings if you want to incorporate those to your spin can to keep it from binding under acceleration. Edit: i just realized they probably won't help at all. The motor is where the bearings would be hah

The spin can does have ball bearings, probably 150 or them or so (I use two sizes and don't remember specifically how many of them). However, they are between two fiberglass surfaces (the flattened ends of two fiberglass tubes). I suspect that this fiberglass deforms just enough under load to reduce the spinning freedom of the spin can. I know it still turns under the load, but the force required may be more than what is required during flight.

By the way, there are lots of pictures of the spin can design earlier in this thread.

Jim

 

[TonyL]

Certainly without something comparable to a thin section bearing under it, it is is almost assured to lockup until the axial forces on it are low enough. If you are controlling 3 axes, then the system would seem to work without the spin can given a bit of tuning...

 

[JimJarvis50]

Certainly without something comparable to a thin section bearing under it, it is is almost assured to lockup until the axial forces on it are low enough. If you are controlling 3 axes, then the system would seem to work without the spin can given a bit of tuning...

Right now, the upper and lower bearing surfaces are the ends of fiberglass tubing. I'm sure things would work better if I could replace that with metal tubing. Unfortunately, the standard 4" aluminum tubing doesn't quite fit.

Jim

 

[JimJarvis50]

I had a chance to fly my stabilization system last weekend. It was a pretty severe test of the system with the rocket launched at about 12° into a brisk wind. Thge rocket weathercocked initially but then went vertical. The on-board video is interesting because you can clearly see the rocket go vertical from the smoke trail. Everything came back fine, so I get to fly it again.

Jim

 

[ksaves2]

That is really cool Jim. I know the reason for the spin can but by mere chance I built a 4 inch rocket with plywood fins that apparently ended up mis-aligned “just right” and the rocket didn’t spin on the J motors I flew it on. Keychain camera video was was interesting. I couldn’t repeat this feat in a million years. Gotta go dig out the video. It’s stashed on a loose hard drive in the basement. Kurt Savegnago

 

[JimJarvis50]

That is really cool Jim. I know the reason for the spin can but by mere chance I built a 4 inch rocket with plywood fins that apparently ended up mis-aligned “just right” and the rocket didn’t spin on the J motors I flew it on. Keychain camera video was was interesting. I couldn’t repeat this feat in a million years. Gotta go dig out the video. It’s stashed on a loose hard drive in the basement. Kurt Savegnago

There's things I've done in rocketry, caught on keychain cameras, that I don't think could be repeated! I had a flight where stages collided, and I remember telling Stu "you're not gonna believe this"....

Jim

 

[BABAR]

Nice flight!

 

[JimJarvis50]

Nice flight!

Thanks! I wish I had a better ground video of this flight. It was very interesting how the rocket just did this graceful arc upwards to vertical. Unfortunately, when the button get's pushed, it's hard for me to focus on video.

For the next flight, I was thinking about doing a flight with roll control only. On many of my flights, it doesn't appear that the roll control part of the flight is working as well as it should. The problem is that vertical stabilization takes relatively large canard movements, and if things are not perfectly aligned with respect to roll, then it is hard for the relatively small canard roll movements to counteract that. I'm surprised it works as well as it does. So, I'm thinking of doing a flight with roll control only, just to show how it works when it doesn't have to fight the vertical stabilization.

Jim

 

[JimJarvis50]

So, I did the flight I mentioned above with roll control only (at AirFest). It wasn't a particularly good flight, as there was a steady-state roll through the flight. In addition, the flight showcased a limitation in the flight program, which is the accumulation of apparent tilt if there is rapid rotation of the rocket. So, we made some changes to the flight program. The cause of tilt accumulation was error resulting from the use of a small angle approximation when the roll rate was high. Adding the third term of the Taylor's expansion fixed the problem. Then, we also calibrated the roll gyro (i.e., put in a gain factor) to make overall roll angle estimates much more accurate.

With these improvements, we added two other features. First, we added a heading-hold capability borrowed from RC helicopters. Basically, from an initial roll position, the roll rate is integrated and used to generate a second term for calculating roll feedback that drives the rocket back to the initial roll position. It's a very interesting approach. Second, we added a capability that we tested a few years ago that allows the orientation of the rocket (tilt and direction) to be specified prior to the flight. Basically, it is possible to specify the tilt angle (or two angles if desired) along with the heading. The capability keeps the rocket pointed in a certain direction, although the actual trajectory of the rocket is still affected by gravity, limitations of the control surfaces, and wind. But it is fair to say that the net result of all of these changes is to generally allow the rocket to be flown at a specified angle in a specified direction with no roll. Sweet! One problem with vertical stabilization is that the rocket tends to drift downwind during the ascent, even if it's vertical. Then, it drifts further downwind during recovery. The ability to fly into the wind in a controlled manner will allow recoveries that are much closer to the pad.

I did a flight this last weekend to demonstrate the above capabilities. The flight went quite well, and I suspect that we'll try another flight or two to fully demonstrate these features (and because it's a lot of fun to fly this rocket). Here's a link to the flight video.

Jim

 

[o1d_dude]

Wow!

The fin can “absorbed” the roll while the fuselage and camera kept the orientation solid.

 

[JimJarvis50]

Wow!

The fin can “absorbed” the roll while the fuselage and camera kept the orientation solid.

I think the spin can helps to avoid control reversal, but it actually makes roll control much more difficult because the moment of inertia of the air frame is lower without the fins being attached. If the fins were attached, and if control reversal could be avoided, roll would be more damped and more aesthetic for side-looking cameras.

The heading hold capability is interesting. There are two terms contributing to the position of the canards for roll. The first term, which we had previously, generates a canard response in proportion to the roll rate. So, if the rocket starts to roll, the canards resist that roll, but this term doesn't try to hold a specific roll position. The second term generates a canard response that increases as the difference from the initial position increases. For example, at the initial roll position on the pad, this second term is zero. If the rocket turns from zero to 180 degrees, the canard response goes from zero to the max allowed, in an effort to return to the initial roll position. If the roll ever exceeds 180 degrees, the maximum canard deflection is retained until the rocket starts to rotate in the other direction. At that point, there is a new heading established that is 180 degrees away from the point where the roll direction reversed. But as long as the rocket doesn't turn more than 180 degrees, the original heading is retained. In the flight, the roll angle never exceeded 70 degrees. It is also interesting that the first term of the roll feedback dampens the second term. That is, if the heading hold response is high, due to a roll position that is well away from the initial point, the canard angle generated from this would be offset by the first term if the return roll rate is high. This helps to keep the roll rate from getting too high. Bench testing all of this is pretty weird.

Jim

 

[Joshua Smith]

I think the spin can helps to avoid control reversal, but it actually makes roll control much more difficult because the moment of inertia of the air frame is lower without the fins being attached. If the fins were attached, and if control reversal could be avoided, roll would be more damped and more aesthetic for side-looking cameras.

The heading hold capability is interesting. There are two terms contributing to the position of the canards for roll. The first term, which we had previously, generates a canard response in proportion to the roll rate. So, if the rocket starts to roll, the canards resist that roll, but this term doesn't try to hold a specific roll position. The second term generates a canard response that increases as the difference from the initial position increases. For example, at the initial roll position on the pad, this second term is zero. If the rocket turns from zero to 180 degrees, the canard response goes from zero to the max allowed, in an effort to return to the initial roll position. If the roll ever exceeds 180 degrees, the maximum canard deflection is retained until the rocket starts to rotate in the other direction. At that point, there is a new heading established that is 180 degrees away from the point where the roll direction reversed. But as long as the rocket doesn't turn more than 180 degrees, the original heading is retained. In the flight, the roll angle never exceeded 70 degrees. It is also interesting that the first term of the roll feedback dampens the second term. That is, if the heading hold response is high, due to a roll position that is well away from the initial point, the canard angle generated from this would be offset by the first term if the return roll rate is high. This helps to keep the roll rate from getting too high. Bench testing all of this is pretty weird.

Jim

What, you don't have a transonic wind tunnel laying around?

 

[TonyL]

Hi Jim,
looks like you are getting your system pretty well dialed in.

br/

Tony