Puzzling problem with flight path stability

The Rocketry Forum

Help Support The Rocketry Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

Joharvey1

Well-Known Member
Joined
Mar 8, 2021
Messages
58
Reaction score
61
So I had a puzzling result at the range this past weekend, and after consulting with several experienced rocketeers, I am still at a bit of a loss as to what may have cause this issue. It was suggested I post the results here to see if the forum had any more ideas.

Lets start at the beginning. I purchased secondhand a PML Endeavour, which I repainted and added a third rail guide to, along with an Aeropak retainer and Aeropak Adapter down to 38mm from 54mm. Now, other than that, the rocket is as it was when I purchased it. No additional nose weight or significant modications have been done under my ownership.

Now, as I looked at the rocket prior to launch in Open Rocket, it simmed to around 3412 grams or approximately 7.522 lbs.

Individual component weights were: Nosecone (325 grams), Payload (881 grams), and Fin Can (2206 grams.) At this point, the rocket had a stability in Open Rocket of 2.93 calibers.

It was noted prior to launch that any motor below a J460T would result in a caliber rating above 2.0 calibers. A J420R registered about 2.1 calibers. Ultimately however, I was advised that certain motors can straighten out a rocket path well enough off the pad as long as the thrust to weight is something above 7:1. Closer to 10:1 was best as I was told. Taking this information with the goal of a lower altitude flight, ultimately we settled on an I500 single use. Which, as I was told, would "straighten anything out off the pad," and that stability really shouldn't matter much except when the rocket was "higher up" in this case.

I should also note there was very little wind that day, if any.

endeavourlaunch2.jpg

endeavourlaunch1.jpg

When the rocket came off the pad however, it almost immediately pitched several degrees north and altered the flight path. It came down some distance from the pad on a chute release, whereas most of the flights of the day had been coming directly down near the pad. A subsequent flight on the same pad did a similar pitch for no apparent reason.

It was suggested several times that I add more nose weight. However, the computer simulations suggest that adding more nose weight actually makes the stability problem worse.

Interestingly, the motors that simmed the best stability wise prior to launch were: J460T (1.82 calibers), J540T (1.65 calibers), and K535 (1.55 calibers)

I guess my question is: does anyone know what might cause an I500 to wobble straight off the pad in this way?

View attachment Endeavourflight1.mp4

















For now, I am puzzled.

Perhaps I should fly it again on a more powerful motor with a better computer stability rating to see if that improves it.
 
What you have focused on is the measure of static stability. This tells you only that the center of pressure is behind the center of mass.

When you launch, all sorts of dynamics enter the picture some of which are poorly modeled in simulation packages. Launch rod/rail interactions aren't modeled at all.
 
It looks like the smoke right at the point where your rocket began tilting started being blown the other way, indicating a layer of higher wind speed at that altitude. If that’s truly the case this would appear to be simple weathercocking, which is fairly common with high stability margins.
I would try it with a lower thrust motors see if it is better or worse.
As David said above, lots of other dynamics come into play. A very high thrust motor will not always straighten out a flight. If construction resulted in a center of gravity that’s not along the centerline of the rocket you can end up with a moment which rotates the rocket around the Cg. Some rockets just do some weird things.
 
Last edited:
It looks to me like you have off axis thrust. That is, the thrust vector does not run through the CG. This creates a disturbing torque that will cause the rocket to pitch. However, since it is stable it will simply rotate to a trim angle of attack. This trim able of attack is inversely proportional to the static stability margin, so in that case a larger static stability margin will result in straighter flights. You should first check the alignment of the engine mount, then check for nozzle misalignment.
 
You say the rocket "simmed" to 3412 grams. Did you confirm the mass and CG with actual measurements?
Sorry, I may have used the wrong word there to represent my meaning. Yes, those were scale weights I put in as overrides. And the CG was finger balanced and later inspected by an RSO. Also correlated with Open Rocket values.
 
Last edited:
Did it tilt toward the third rail? If it did, there is your answer.

As others have said, if the thrust vector is not exactly through the CG, the rocket will experience an overturning moment which will result in an angle of attack vs. the air stream. I see this in bullets all the time. The rocket may still fly a good trajectory, but if the angle of attack gets too large, the overturning moment will cause it to tumble.
 
Did it tilt toward the third rail? If it did, there is your answer.

As others have said, if the thrust vector is not exactly through the CG, the rocket will experience an overturning moment which will result in an angle of attack vs. the air stream. I see this in bullets all the time. The rocket may still fly a good trajectory, but if the angle of attack gets too large, the overturning moment will cause it to tumble.

If by third rail you mean the one on the far right in the photo above, then yes. it tilted in that general direction. There may have been a slight displacement from a z-axis perspective. I said due north, but the landing zone may have been slightly more north west. I presumed due to the strange flight path. But thrust vectoring is a new explanation I will have to look more into. Such as the alignment of nozzle and motor mount as mentioned above.
 
I'm not sure if this applies to all sim programs, but Openrocket definitely does not simulate split fins correctly, so the actual CP is almost certainly higher than simulated.
 
It looks to me like you have off axis thrust. That is, the thrust vector does not run through the CG. This creates a disturbing torque that will cause the rocket to pitch. However, since it is stable it will simply rotate to a trim angle of attack. This trim able of attack is inversely proportional to the static stability margin, so in that case a larger static stability margin will result in straighter flights. You should first check the alignment of the engine mount, then check for nozzle misalignment.

This is a good explanation in theory, and it appears in handbooks, like Stine. In practice, though, how do you diagnose off-axis thrust with any certainty? Unless the rocket was built by a 3 year old with a hammer, the MMT will be aligned in the airframe within the tolerances of the parts (which is pretty good, these days). I've never assembled an HPR motor that had a noticeably crooked nozzle. Unless you are launching some kind of crazy side-burning motor or have a seriously lopsided and heavy payload, the axis of thrust and CG of the rocket will be misaligned only by fractions of a degree, IMO.

How do you measure the off-axis CG of the rocket? Trying to balance it with any accuracy while it is standing upright seems impossible. Simulation would be the best way, if the RS or OR software allows for off-axis CG overrides (I don't remember, off hand). You would need to enter the x,y,z CG overrides piece by piece, which would still be challenging.

I would chalk up the OP's flight to more basic CP/CG issues, weathercocking, or simply "rockets do weird things" rather than trying to find a fraction of a degree axis misalignment.
 
If by third rail you mean the one on the far right in the photo above, then yes. it tilted in that general direction. There may have been a slight displacement from a z-axis perspective. I said due north, but the landing zone may have been slightly more north west. I presumed due to the strange flight path. But thrust vectoring is a new explanation I will have to look more into. Such as the alignment of nozzle and motor mount as mentioned above.
"Third rail" is the nomenclature you used in your original post.
 
"Third rail" is the nomenclature you used in your original post.

Oh I see I misunderstood what you were asking. I was wrong in either case. Thanks for pointing that out. I thought you were asking about flight path direction. haha.
 
No, he added an additional rail guide to the two he already had. He said he added a third “rail guide”, not a “third rail” guide. 😀
Sorry. Misunderstood.

Nevertheless, adding mass on one side of the rocket without balancing it will move the CG off the centerline. If you apply thrust to the centerline, you will get an overturning moment. This may be offset by the stability provided by the fins, or it may not be.

In a bullet, overturning moment is offset by spinning the bullet.
 
Sorry. Misunderstood.

Nevertheless, adding mass on one side of the rocket without balancing it will move the CG off the centerline. If you apply thrust to the centerline, you will get an overturning moment. This may be offset by the stability provided by the fins, or it may not be.

In a bullet, overturning moment is offset by spinning the bullet.

that is an interesting point. I may have to stick an opposing mass on the other side of the rocket. I felt the third "rail button" was necessary because some previous owner of the rocket had installed the rail guides low on the aft end of the chasse. I felt there was too much pressure being applied to the upper rail button, and I was concerned it might damage the rocket somehow. It felt more right with three distributing the weight. I was told that the rail button getting hung on the high power rail shouldn't have affected the path of the rocket. But just since we are discussing the rail buttons, I did notice that I felt resistance on the rail when I slid the rocket down onto the pad. I felt enough give that the rocket slid back and forth vertically on the rail, but it wasn't a perfect motion. I wonder if one of the guides was misaligned: either the one I installed or one of the original guides.
 
Last edited:
that is an interesting point. I may have to stick an opposing mass on the other side of the rocket. I felt the third "rail guide" was necessary because some previous owner of the rocket had installed the rail guides low on the aft end of the chasse. I felt there was too much pressure being applied to the upper rail guide, and I was concerned it might damage the rocket somehow. It felt more right with three distributing the weight. I was told that the rail guides getting hung on the high power rail shouldn't have affected the path of the rocket. But just since we are discussing the guides, I did notice that I felt resistance on the rail when I slid the rocket down onto the pad. I felt enough give that the rocket slid back and forth vertically on the rail, but it wasn't a perfect motion. I wonder if one of the guides was misaligned: either the one I installed or one of the original guides.
I never use three fixed buttons on a single piece of tube. Two points define a straight line. I have used a third button on a separate section that I can rotate to align. And I really don’t like rail guides on the bottom of the rocket. I’ve seen too many twisted off while loading the rocket on the rail. Buttons are much more forgiving.
 
I never use three fixed buttons on a single piece of tube. Two points define a straight line. I have used a third button on a separate section that I can rotate to align. And I really don’t like rail guides on the bottom of the rocket. I’ve seen too many twisted off while loading the rocket on the rail. Buttons are much more forgiving.

yes they were rail buttons not guides. I was using the wrong terminology again, sorry. To me, guide was a generic term referring to rockets going off the rail with something other than a launch lug. I guess I've never used a true 'rail guide' then. Interesting apparatus. I see that one wrong word changed a lot of the implied meaning, but you added an explanation in anyway which covered my misstatement.
 
Last edited:
This is a good explanation in theory, and it appears in handbooks, like Stine. In practice, though, how do you diagnose off-axis thrust with any certainty? Unless the rocket was built by a 3 year old with a hammer, the MMT will be aligned in the airframe within the tolerances of the parts (which is pretty good, these days). I've never assembled an HPR motor that had a noticeably crooked nozzle. Unless you are launching some kind of crazy side-burning motor or have a seriously lopsided and heavy payload, the axis of thrust and CG of the rocket will be misaligned only by fractions of a degree, IMO.

How do you measure the off-axis CG of the rocket? Trying to balance it with any accuracy while it is standing upright seems impossible. Simulation would be the best way, if the RS or OR software allows for off-axis CG overrides (I don't remember, off hand). You would need to enter the x,y,z CG overrides piece by piece, which would still be challenging.

I would chalk up the OP's flight to more basic CP/CG issues, weathercocking, or simply "rockets do weird things" rather than trying to find a fraction of a degree axis misalignment.
I diagnosed off axis thrust from the photographs that were posted. Joharvey1 bought the rocket second hand, so the alignment may be more suspect than a rocket that you built yourself. I did not want to lay the blame for an experience that I did not witness on the motor manufacturer. I have never used a reloadable motor, but those would seem to imply that the user has more control over nozzle misalignment and measurement. I have had AT SU motors with nozzles that were visually misaligned, but they flew OK, even to supersonic speeds. I am encouraged that you have never noticed off axis thrust. Even when there is nothing measurable prior to launch, some nozzles in use have blown out a chunk of the nozzle, or eroded unevenly.

I have seen many rockets with camera attached to one side of the rocket. Yet I have never seen wind tunnel force and moment measurements of such a configuration. Obviously the CG of the rocket is significantly altered as well. As you suggest, sport rocketeers take a lackadaisical approach to mass balancing. Symmetry is a wonderful thing and I like the use of two cameras, one pointing forward, and one aft, on opposite sides of the airframe. And of course many boost gliders fly well enough without radial symmetry.

Many sounding rockets are spin balanced to align the principal inertial axis to the vehicle. I have suggested several times that someone should design a spin balancing rig for sport rocketeers, but AFIK, no one has taken up the challenge. For large vehicles to big for the spin table, they just put all the known weights and measurement into the computer and hope for the best; of course those vehicles typically have stabilizing guidance systems.
 
Last edited:
I have owned THREE 4" split fin rockets, all with perfect fin alignment. I usually go with 10:1 or greater thrust to weight. EVERY split fin rocket I have owned has done this. Straight as a laser off the rail....then decides to tilt over.

I know in science, correlation is not causation....but I no longer own any HP split fin rockets and my flights are nice and straight. Just my two cents.

Andrew
 
Why would a single stage rocket have split fins? I'll have to Google the model to see what they look like, but it seems to me that unless they interfere with one another somehow, a proper stability calc could be made, per Barrowman. But I'm just getting into all that so I may be wrong. Interesting problem, interesting thread!
 
Off center CG is usually only a problem with high rotation rates which is why sounding rocket payloads pay a visit to the dynamic balancing machine. It would have to be really off center to have much impact to the flight path of a non-rotating rocket.

Two buttons are sufficient and should be at or near the CP and CG for best results.
 
Why would a single stage rocket have split fins? I'll have to Google the model to see what they look like, but it seems to me that unless they interfere with one another somehow, a proper stability calc could be made, per Barrowman. But I'm just getting into all that so I may be wrong. Interesting problem, interesting thread!

I went to Youtube... and looked at the top results for "PML Endeavour". Mind you I just did this, not before I launched my version. And j I see some similarities in the flight tilt in the first video.


Also, here is a larger version of the same rocket in the L2 range.



And a much larger two stage...

 
Last edited:
I have one as well, or at least it started out with that kit. Here it is in an intermediate configuration. I think this is from when I flew it on a K550. It survived but every time I think about those poor fins now I shudder.

EndeavouronPad1.jpg
 
Off center CG is usually only a problem with high rotation rates which is why sounding rocket payloads pay a visit to the dynamic balancing machine. It would have to be really off center to have much impact to the flight path of a non-rotating rocket.
That is true. I did not mean to imply that all all our rockets should be spin balanced. I was simply responding to Buckeye's question about measuring the off axis mass imbalance. I find it odd the Buckeye is dismissive of thrust misalignment, but concerned with off axis mass.

Rolling or spinning the vehicle is the best way to mitigate the effects of thrust misalignment and other asymmetries, and that brings us back to spin balancing and dynamic analysis.
 
One way to help balance your rocket axially is to mount the batteries on the opposite side of the sled from the electronics. Mounting them all on the same side, especially in a lightweight rocket, can cause the rocket to tilt.
 
This is a good explanation in theory, and it appears in handbooks, like Stine. In practice, though, how do you diagnose off-axis thrust with any certainty? Unless the rocket was built by a 3 year old with a hammer, the MMT will be aligned in the airframe within the tolerances of the parts (which is pretty good, these days). I've never assembled an HPR motor that had a noticeably crooked nozzle. Unless you are launching some kind of crazy side-burning motor or have a seriously lopsided and heavy payload, the axis of thrust and CG of the rocket will be misaligned only by fractions of a degree, IMO.

How do you measure the off-axis CG of the rocket? Trying to balance it with any accuracy while it is standing upright seems impossible. Simulation would be the best way, if the RS or OR software allows for off-axis CG overrides (I don't remember, off hand). You would need to enter the x,y,z CG overrides piece by piece, which would still be challenging.

I would chalk up the OP's flight to more basic CP/CG issues, weathercocking, or simply "rockets do weird things" rather than trying to find a fraction of a degree axis misalignment.
Maybe the original builder put a gob of epoxy on the MM then let it dry on one side. Plausible but stupid…
 
Back
Top