98mm Min Dia L3 Build

The Rocketry Forum

Help Support The Rocketry Forum:

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

A5tr0 An0n

Well-Known Member
Joined
Sep 29, 2012
Messages
521
Reaction score
10
Intro:
Well I have been thinking about this rocket for about a year now and I thought it was about time to announce my intentions to build a L3 rocket for my certification at BALLS 23. Now most of you do not know me, I tend to keep rather quite on here. I try to post but I am ever so busy with school, work, my internship, and blah blah I am a bad TRF member lol. Anyways I will keep this short and simple (or so I thought). My plan is to build and launch a 98mm minimum diameter rocket..... successfully. The details will come all in good time. First, however I feel that I must name the people who have greatly contributed to amateur rocketry, pushed the envelope in knowledge and who have inspired others (at least in my case). Nic Lottering, Jim Jarvis, Mike Passaretti, Derek Deville, and the AeroPac group. There are many others who have done great things in this hobby but I am trying to keep the list as relevant as possible to this build. The documents that these men have released have been of great value and help to me and the community. I have learned a lot from these documents and it is in fact because of those documents that I am currently ready to begin my L3 rocket. Without their hard work and innovation I would not of been able to progress so rapidly. I plan on documenting this build very closely so to therefore help others as I have been helped. Now lets move on to the rocket!

Design:
So my basic design is rather simple. A MD 98mm ALL composite rocket that will fly on a 6GXL Engine.

1. First and foremost is to have a safe and successful flight; meaning no failures and the rocket survives and is recovered.

2. I will try to keep the rocket as simple as I can, to therefore reduce the probability of failure.

3. Everything about this rocket will be tested and tested again. I will calculate, experiment and simulate as much as I can to ensure that the overall flight will be successful.

4. NC is a 6:1 FWFG w/ aluminum tip. It is ~3mm thick with a 10'' shoulder that is of the same thickness.The extended shoulder allows for a more sturdy connection with the airframe. This not only increases the coupled strength of the NC/AF but also reduces the volume in-between the forward closure and NC shoulder bulkhead. This makes it easier to vent and hopefully will increase the chances of separation at apogee. The NC has been made to handle temps up to 600F. I have played around in the sims with the standard 5:1, 6:1, and 7:1. I have chosen to go with the 6:1 for better performance over the 5:1 and shorter length over the 7:1. However the 7:1 did seem to offer 0.01 more calibers of stability and ~500ft more of altitude. I chose FWFG for its mixtures of strength, lightweight, and RF transparency.

5. Airframe is FWCF that has also been made for temps up to 600F. The airframe is 66'' in length X 4'' in diameter and is ~3mm thick. It is also a single continuous airframe. I have looked at failures that occurred during high mach flights and I believe some of them were due to the coupled airframes. So my plan is to eliminate that structural weakness in the airframe and hopefully increase the percentage of success. Not only have I designed a single airframe but the entire vehicle is reinforced. The thickness of the NC, NC shoulder, and the airframe is ~3mm thick. Furthermore because of the NC shoulder and the engine hardware, the entire rocket is almost completely "triple walled." The length of the entire vehicle is 90'' and the true OD is ~3.983''.

6. Engine retention will be either the Aeropack MD retainer or a "forward thrust plate retainer," piece that I had machined for MD flights. Both pieces are 6061-T6 aluminum and both sit in front of the engine. They are retained to the AF with epoxy and counter sunk screws.

7. The basic thought on the fins are that they will be as small as reasonable stability allows to reduce the aerodynamic loads they will experience. They will be ~3mm thick of G10 FG plate that is layered with 3k aerospace grade CF and also capable of 600F. Now the aerodynamic loads on this flight are a force to be reckoned with, therefore to prevent fin flutter (all types) I am doing full t2t with 3k aerospace grade CF. I plan on doing 3 layers but will be doing some experiments with static loads and heating to see if I require more layers. Prior to this I am doing epoxy, surface prep., and additive testing (link can be found in notes section). Another big concern is delimitation of the CF that is due to the aerodynamic heating. I have spent some time looking at ablatives and have opted out of using them. I feel that ablatives however effective are just too much of a pain. I would like the rocket to be reusable without having to re-coat it with an ablative and having any downtime. I also am concerned with the rate of which they burn off requiring more testing and calculations to ensure that I have the correct thickness to be burned. Also ablatives can leave part of the rocket uneven and I would like to keep the rocket as aerodynamic as possible throughout its entire flight. So I am going with Nic's idea that he used on Mad Max. That is, I am putting metal over the leading edges of the fins… I think this is brilliant! I mean we put a metal tip on the leading edge of our composite NCs, why not on the leading edges of our composite fins?! Thank you Nic for that.

8. Recovery is DD out of a single airframe. This takes out a weakness in the design that could of otherwise been fatal. Most likely the parachutes will be a small 18'' drogue and a 72'' main deployed via "high altitude capable," BP. The parachutes Cd is 1.6 and 2.2, respectfully. This creates a descent rate of ~80ft/sec and 28ft/sec, respectfully. This is all assuming a total rocket weight of 15lbs. I came up with a way to go about DD from a single airframe without the need of something like an AARD, cable cutters, or a tender descender. It is a very simple design and some what of a traditional manner. The biggest points of this system are that it moves the main into the NC and the drogue in-between the NC shoulder and the engine retainer/thrust plate. The only new items that are introduced, are another bulk plate and thus it is more like a traditional setup v. something that requires your main parachute to float out in the atmosphere during descent from apogee. I hope that this creates a more simple and reliable setup. I will go into slightly more detail with some CAD drawings and explanations later on in the thread. I have two other rockets that are setup for this but I have yet to fly them. The sole point of one of these rockets is test out the new DD setup to ensure it is reliable; it will also be tested in a vacuum chamber. I will update you all with the results once all that has been accomplished.

9. Electronics are most likely going to be TeleMega x2, BRB 70cm GPS x1, stratologger x2. That gives me 2 primary flight computers, 2 redundant flight computers, and 3 GPS/telemetry systems. All of which will be operating on different frequencies to avoid interference. This system will also be vigorously ground tested. I have some Ravens laying around and might use those, but as of now that is the setup I am thinking of. Every board will have its own power source and the 2 primary flight computers will each use 2 batteries. 1 for the pyros and 1 for everything else; the power sources will be LiPos. All flight computers/GPS systems will be mounted in the NC. I will have on board cameras on this flight and the cameras will be GoPro H3 x2. They will be housed in the shoulder of the NC. I would like to play with mirrors to get some cool angles for flight video but I do not want to increase drag. If there are any ideas on how to avoid this then please share!

10. I will paint this rocket with high temperature paint (1500F). My goal with the paint is not so much for form but more so function. I am hoping the paint will act as both a insulator and an ablative. This will hopefully eat up some dangerous energy and hopefully aid in preventing delamination and structural weakening. I will do at least 3-5 layers of paint to make it more effective. As far as the colors and style of the paint job... idk. I will think about it once I get there.










Mostly designed using OpenRocket. You cannot really tell from the picture but the NC shoulder goes back 10-12" (TBD), which would explain the position of the aft bulkhead. The forward mass object is the flight electronics and the smaller mass object is the cameras. Then comes the main parachute and lastly the drogue. The other parts are bulkheads and a engine retainer.










This graph is pretty self explanatory; altitude vs mach vs stability. Instead of showing the acceleration (peak of 40G) I thought it was more interesting looking at the before mentioned. On a side note, the vehicle experiences max Q ~3.25 seconds into the flight at an altitude of ~7kft and traveling faster than the speed of sound 3.5 fold. This is interesting sense most rockets go through max Q at transonic speeds. At max Q the rocket experiences ~1,300N of force. Note, this is note a as-built sim.







13940745773_5a3e19403f_o.jpg



The stability of the rocket during the course of its flight never drops below ~1.5 calibers. This should allow enough stability for any error of margin and to minimally help protect stability drops due to changes caused by weather, AOA, etc. One thing that I would like to point out is that the CP moves in higher velocity flights down as normal, then rapidly up towards the CG, then eventually down again. The up movement towards the CG can be quite dangerous due to the fact that as the CP and CG near, the stability of the rocket drops. In my case the CP drops from about 74'' to 61'' in 2 seconds and approaches my CG with a distance of about 6". This is just one more rocket killer that must be dealt with to survive a higher velocity flight. Note, this is not a as-built sim.











So here is my solution to DD out of a single airframe. I have not been able to successfully draw up a pretty CAD of this yet, so please bear with me. So the way this work is pretty basic and can be explained in two major parts. 1) The drogue parachute is housed in the AF in-between the NC shoulder bulkhead and the engine retainer. It is connected via 1500# Kevlar shock cord to standard forged eyebolts. The drogue is deployed via “high altitude capable,” BP at apogee. The pyro canisters house the BP on the aft NC shoulder bulkhead; this bulkhead has a “lip,” that prevents it from travelling into the NC shoulder due to the explosive event. Furthermore to prevent the bulkhead from travelling outside of the NC shoulder (due to the energetic pull from the shock cord during separation), a couple pins will bolt it to the NC shoulder until the main event. 2) The main parachute is housed inside the extend NC shoulder. There is a bulkhead forward and aft of the main parachute. The main parachute is also attached via 1500# Kevlar shock cord to a forged eyebolt. The forward bulkhead is bolted to the NC shoulder via counter sunk screws and houses two pyro canisters. These canisters house the BP that will shear the pins holding the aft bulkhead and thus eject the main parachute. The aft bulkhead has a whole in the origin, where the shock cord runs through. This prevents the bulkhead from ejecting in the atmosphere and also keeps the NC and AF together.





14075813786_3f9c25e5fb_o.png




Put together a quick possible failure analysis.​



Notes:
Here are some thoughts that I had and I will update this list as this thread progresses.

*As far as the sims go, I have been primarily using OpenRocket and RASaero. Although I do need to get a little more fluent with RASaero. RockSim is a nice piece of software but does not seem up to par for supersonic flights. I also have Solidworks, Autodesk and some ANSYS programs that I hope to use for CADs and flow analysis. I also need to get more fluent with them! I do not have finsim and unfortunately it is no longer being sold. If anybody would like to help me out with that then thank you very much ;)

*I would like to build on the knowledge of high mach flights in regards to temperature of various parts of the rocket due to aerodynamic heating. I would like to use some method of temperature indication on the NC, AF, and fins. I was thinking of using something like a welding marker or paint. Sensors may also be an option. I think it would be nice to have an approximation of how hot it is actually getting; this can help with construction techniques.

*I am providing some internal/external links to my personal website and the other threads on the forums. This may be redundant but will have more detailed information on the design, construction process, simulations, experiments, and post flight data. Please feel free to comment on it however you would like; any feedback is good feedback.
- coming soon
- https://www.rocketryforum.com/showthread.php?65106-Battle-of-the-Epoxies-A-Road-to-BALLS-Experiment (Epoxy experiment thread)
- https://www.ausrocketry.com/forum/viewtopic.php?f=10&t=4522

*This will naturally be tower launched. In the near future (as BALLS approaches) I will make a tower for this vehicle and will create a new thread showing that build.

Conclusion:
I am well aware that a flight of this magnitude has a high probability of failure and that a lot of things come into play on this scale. I am also well aware that some members may feel the need to post negative comments and I ask you to please refrain from that. I have not stated which engine I will use for this flight in hopes that I will not instigate any arguments or negative behavior in this thread. I do however encourage constructive criticism and will give it serious thought. I am also well aware that I could build a "low and slow," L3 rocket that would greatly increase the chances of success.... but that is not who I am. This flight is more about pushing my limits and knowledge and thus furthering my progression and the progression of the community; I feel this is more important than obtaining an L3 certification. I want to thank everyone for viewing this post and now lets go build a rocket.
 
Last edited:
Lofty (in more ways than one). Nice to see someone "pulling out all the stops" for their L3. Depends on who your TAP's are, but I don't think mine would have let me do that type of project for my L3.
 
If you could attach sensors within the airframe and on the nosecone to test for local heating, that would be awesome. Bryce (Bandman444) and I were going to do that for one of our rockets, but we couldn't find anything small, cheap, reliable, and simple enough to implement on a 98mm rocket.

Have you made a decision as to which motor you're going to use? Longer burn or higher thrust?

Remember that if you're breaking mach 3, aero heating goes past 800F and will glass your epoxy and tear your carbon reinforcement. It's happened on more than a few 98 6XL rockets. The nosecone will be pitted and messed up, but should remain okay if it's reinforced, though the bulkheads could possibly become jammed or knocked loose due to the G kick.
 
Interesting. I like that you're trying to do some numerical analysis. I take it you're an engineer.
 
N5800, huh? Daring.

Fins fins fins. If you have sufficient thickness, I don't think flutter and material strength is an issue anymore, but rather you need to focus on attachment.

As of right now, no minimum-diameter N5800 has successfully (or semi-successfully, either) flown that had all-epoxy fin attachment: Don't Debate This had a brazed fincan, Kari Pill and Bare Necessities had all-aluminum fincans (the latter of which survived corkscrewing at M4), and Mad Max had aluminum fin brackets bolted to the airframe from the inside.
Project60K in 2012 didn't use any tip-to-tip, but their fins' bonds exceeded the peeling strength of the FWCF airframe. To be honest, I would step away from the filament-wound stuff unless you can find a source of ribbon-wound tubing that is not ground on the outside.
I recall overhearing Tom Rouse at last year's LDRS saying that he ordered decorative ribbon-wound tubing (gold and black, really cool-looking) made for covering motorcycle mufflers, but he said that the minimum order was 50 feet.

Otherwise, I'd look into alternative fin attachment.
 
From your drawing I'm having trouble understanding your DD setup-- if you have a single airframe and a single coupler tube, how do you get your main out for a second event? The only thing I can think of is some shear pins in the aft-most bulkhead on the NC assembly? The bulkhead would need to have some length to it, more like a coupler within a coupler, and I imagine the shear pins would need to be beefy enough to withstand the shock of the first event. You wouldn't want them shearing off when the drogue shock cord reached its full extension. More drawings and/or pics would be helpful.

Edit: ....or, after a bit more head-scratching, maybe you could break the NC coupler at the middle bulkhead, with a longer coupling "bulkhead" and shear pins at that point? More pictures please :)
 
Last edited:
Interesting. I like that you're trying to do some numerical analysis. I take it you're an engineer.

In the process of becoming on yes :)

N5800, huh? Daring.

Fins fins fins. If you have sufficient thickness, I don't think flutter and material strength is an issue anymore, but rather you need to focus on attachment.

As of right now, no minimum-diameter N5800 has successfully (or semi-successfully, either) flown that had all-epoxy fin attachment: Don't Debate This had a brazed fincan, Kari Pill and Bare Necessities had all-aluminum fincans (the latter of which survived corkscrewing at M4), and Mad Max had aluminum fin brackets bolted to the airframe from the inside.
Project60K in 2012 didn't use any tip-to-tip, but their fins' bonds exceeded the peeling strength of the FWCF airframe. To be honest, I would step away from the filament-wound stuff unless you can find a source of ribbon-wound tubing that is not ground on the outside.
I recall overhearing Tom Rouse at last year's LDRS saying that he ordered decorative ribbon-wound tubing (gold and black, really cool-looking) made for covering motorcycle mufflers, but he said that the minimum order was 50 feet.

Otherwise, I'd look into alternative fin attachment.

Yeah, one problem we found with the FWCF was that they wrap the tubes, then grind them to be smooth/the right diameter. What that causes is that the entire surface becomes frayed fibers, which are far more likely to tear than a single wrap. Honestly if they inverted the tube wrapping process, and wrapped almost parallel on the outtermost surface, and could make it smooth without grinding, it might work.

If I had to try again I would probably get a no overhang aft closure and glue the fins directly to the case, and use coupler tubes or aluminum tubes to make it nice and flush to the case wall, like Bare Necessities.

I think the biggest problem with a 98-6XL rocket is that you're working with the smallest margins of error possible. Extremely high thrust coupled with low mass and low altitude means stupid high acceleration and drag, which is what caused all the minimum mass rockets to fail. The ones that did succeed used heavy aluminum fincans to reduce speed to pull the loads down.
 
Lofty (in more ways than one). Nice to see someone "pulling out all the stops" for their L3. Depends on who your TAP's are, but I don't think mine would have let me do that type of project for my L3.

Haha thank you. I could see and understand that some TAPs would be uneasy about this. My two TAPs as of now are onboard with this project… both of my TAPs constantly push the limits of the hobby.

If you could attach sensors within the airframe and on the nosecone to test for local heating, that would be awesome. Bryce (Bandman444) and I were going to do that for one of our rockets, but we couldn't find anything small, cheap, reliable, and simple enough to implement on a 98mm rocket.

Have you made a decision as to which motor you're going to use? Longer burn or higher thrust?

Remember that if you're breaking mach 3, aero heating goes past 800F and will glass your epoxy and tear your carbon reinforcement. It's happened on more than a few 98 6XL rockets. The nosecone will be pitted and messed up, but should remain okay if it's reinforced, though the bulkheads could possibly become jammed or knocked loose due to the G kick.

I thought about adding sensors but I am not sure exactly how I would go about it ( I honestly have not looked into it that much). I was more or less leaning towards some type of temperature indicating paint, marker, crayon, etc. Would you happen to have a link to such needed sensors, if you have since found any? I am curious about their size and what reads/stores the data… maybe an Arduino?

Yes I know which motor this rocket will take. I actually designed it for only one specific motor.

Do you have any documents or data that shows the aerodynamic heating? There are so many variables that play a part in the temperature. Similiar vehicles have had parts that are rated for lower temperatures and have survived. The reason for this (I imagine) is that the amount of time at Max V is minimal and thus never reaches stagnation temperature or does so very a very short period of time. All of the high temp paint layers I plan on using should absorb some of the energy, increasing chances for survival.
 
Last edited:
I would like to build on the knowledge of high mach flights in regards to temperature of various parts of the rocket due to aerodynamic heating. I would like to use some method of temperature indication on the NC, AF, and fins. I was thinking of using something like a welding marker or paint. Sensors may also be an option. I think it would be nice to have an approximation of how hot it is actually getting; this can help with construction techniques.

If you don't already know about it, you should read up on this project over at Altus Metrum:
https://altusmetrum.org/TeleScience/

YikStik 2 and 3 were designed with exactly this goal in mind, mapping nose cone and fin temps in flights above mach:
https://www.gag.com/rockets/airframes/2YikStik/
https://www.gag.com/rockets/airframes/YikStik3/

Temperature sensors were embedded in the fins and in a custom NC made by Shockwave. Fantastic build photos here:
https://gallery.gag.com/rockets/YikStik3/Build/
 
They will be ~3mm thick of G10 FG plate
Don't skimp on fin material. You need all the stiffness you can get. Real men use Dragonplate.

This is just me backseat driving, but I would shrink your fins down. By extending the root chord and shortening the span; your fins would be stronger, more aerodynamic and better for high speed stability.

I am also well aware that I could build a "low and slow," L3 rocket that would greatly increase the chances of success.... but that is not who I am. This flight is more about pushing my limits and knowledge and thus furthering my progression and the progression of the community; I feel this is more important than obtaining an L3 certification.

Well put. I love people who embrace this mentality. I'll be going for my L3 in a similar fashion.

Carry on.

Alex
 
N5800, huh? Daring.

Fins fins fins. If you have sufficient thickness, I don't think flutter and material strength is an issue anymore, but rather you need to focus on attachment.

As of right now, no minimum-diameter N5800 has successfully (or semi-successfully, either) flown that had all-epoxy fin attachment: Don't Debate This had a brazed fincan, Kari Pill and Bare Necessities had all-aluminum fincans (the latter of which survived corkscrewing at M4), and Mad Max had aluminum fin brackets bolted to the airframe from the inside.
Project60K in 2012 didn't use any tip-to-tip, but their fins' bonds exceeded the peeling strength of the FWCF airframe. To be honest, I would step away from the filament-wound stuff unless you can find a source of ribbon-wound tubing that is not ground on the outside.
I recall overhearing Tom Rouse at last year's LDRS saying that he ordered decorative ribbon-wound tubing (gold and black, really cool-looking) made for covering motorcycle mufflers, but he said that the minimum order was 50 feet.

Otherwise, I'd look into alternative fin attachment.

You are a smart man Carlo ;) lol. On a serious note I have studied the other N5800 attempts. Some successful and some not and Nic's Mad Max is the only rocket close to having a more traditional attachment point (minus the brackets). I do not think flutter should be a problem granted the alpha is not high and do agree that the bond of the fin will play an important role in this flight. That is why I am testing various surface preparation methods, epoxies, and additives to find the best fit for this flight. I have the Af made already, but since you have raised concern about the FWCF's peel strength I will have to look into that. Thank you for that information. If need be I will re-make a FG airframe for this flight. Just another place to test :)

From your drawing I'm having trouble understanding your DD setup-- if you have a single airframe and a single coupler tube, how do you get your main out for a second event? The only thing I can think of is some shear pins in the aft-most bulkhead on the NC assembly? The bulkhead would need to have some length to it, more like a coupler within a coupler, and I imagine the shear pins would need to be beefy enough to withstand the shock of the first event. You wouldn't want them shearing off when the drogue shock cord reached its full extension. More drawings and/or pics would be helpful.

Edit: ....or, after a bit more head-scratching, maybe you could break the NC coupler at the middle bulkhead, with a longer coupling "bulkhead" and shear pins at that point? More pictures please

Yes I know I need more drawings of that. I am trying to get some drawn up in solid works by the beginning of next week. All in good time mate. Let me see if I can clear that up now though. The aft bulkhead is in fact connected via shear pins (something I forgot to mention), the bulkhead before it has pyro canisters on it that push out the main and bulkhead like you might normally expect. The charges that separate the NC from the AF would not be able to compromise that main bulkhead due to a lip. However as you mentioned the only concern is that if the shock cord is to short and/or the charge is overly energetic then that force could shear the shear pins. I have built a 10' long vacuum chamber to test the energy of the charges at altitude and will do ground testing and flight testing to make sure that cannot happen. I should also be able to ~calculate the amount of force that is pulling on that bulkhead and might be able to introduce some stronger shear pins. Right now I am thinking 4-40s.
 
In the process of becoming on yes :)



Yeah, one problem we found with the FWCF was that they wrap the tubes, then grind them to be smooth/the right diameter. What that causes is that the entire surface becomes frayed fibers, which are far more likely to tear than a single wrap. Honestly if they inverted the tube wrapping process, and wrapped almost parallel on the outtermost surface, and could make it smooth without grinding, it might work.

If I had to try again I would probably get a no overhang aft closure and glue the fins directly to the case, and use coupler tubes or aluminum tubes to make it nice and flush to the case wall, like Bare Necessities.

I think the biggest problem with a 98-6XL rocket is that you're working with the smallest margins of error possible. Extremely high thrust coupled with low mass and low altitude means stupid high acceleration and drag, which is what caused all the minimum mass rockets to fail. The ones that did succeed used heavy aluminum fincans to reduce speed to pull the loads down.

I have read that and do believe it is a thing of the past. They are now able to build the tubes to the correct OD when winding them. Whether mine have been or not is something I need to look into. DDT and Bare Necessities both had fin cans on their N5800 flights, that is not something I am looking into doing. I am not interested in throwing a fin can on a motor casing for this flight, however I see myself in the very short future jumping on the fins to motor casing train.

Yes that is per se what ultimately lead to the failure of other N5800 flights. However not all of them were Al fin cans; Nic's Mad Max was 98% composite and handled it arguably better than Mike's rocket. Both survived but Nic's retained less heating damage however that was mostly due to the paint he used vs the paint mike used. So I retract that statement lol. I have no doubt that composites can handle a N5800 and possibly better than metal, the doubt is with the construction methods. Carbon-Carbon would be ideal but it is very expensive and extremely hard to make. It has been used on Mach 7 vehicles that experienced temperatures at cruise around 3000F; handled it no problem. Graphene seems to be very promising but also is hard to make and expensive. I like to believe that our future will be a composite one :p
 
Thank you for posting your build thread. I really love threads like this and will be watching it closely.

Just please scour the internet for composite bonding techniques and anything you can find from high mach flights. There are a lot of myths about bonding on this forum and the best thing to do is to research the science not just "what he said" on this forum.

My two cents.
 
I have read that and do believe it is a thing of the past. They are now able to build the tubes to the correct OD when winding them. Whether mine have been or not is something I need to look into. DDT and Bare Necessities both had fin cans on their N5800 flights, that is not something I am looking into doing. I am not interested in throwing a fin can on a motor casing for this flight, however I see myself in the very short future jumping on the fins to motor casing train.

Yes that is per se what ultimately lead to the failure of other N5800 flights. However not all of them were Al fin cans; Nic's Mad Max was 98% composite and handled it arguably better than Mike's rocket. Both survived but Nic's retained less heating damage however that was mostly due to the paint he used vs the paint mike used. So I retract that statement lol. I have no doubt that composites can handle a N5800 and possibly better than metal, the doubt is with the construction methods. Carbon-Carbon would be ideal but it is very expensive and extremely hard to make. It has been used on Mach 7 vehicles that experienced temperatures at cruise around 3000F; handled it no problem. Graphene seems to be very promising but also is hard to make and expensive. I like to believe that our future will be a composite one :p

If not metal, then what are you going to do to combat the complex stress state in the fin roots? There is radial tensile stress in the fin roots which work to peel tip-to-tip reinforcement off. Aluminum is anisotropic, so it doesn't care, but the inability for composites to handle out-of-plane tension is a fatal flaw in these locations due to limitations on the fiber orientations.

One thing I've been thinking about is manually braiding a very deep fabric for tip-to-tip which would reduce its tendency to peel apart. Unfortunately, that would be really slow, risk damaging the carbon fiber (which you can do by bending it too tightly), and it would do nothing to help the whole reinforcement from peeling off the airframe.

The other thing you could do is monster sized fillets, which defeats the point of composites which would then weigh more than an equivalent aluminum fin root.


The other other thing you could do is lay up a CF fincan layer by layer from back to front in a mold a la 3d printing. This could have arbitrary fiber orientations, except you would not be able to have any fibers oriented forward-aft. (those could be applied on the surface though)
 
If not metal, then what are you going to do to combat the complex stress state in the fin roots? There is radial tensile stress in the fin roots which work to peel tip-to-tip reinforcement off. Aluminum is anisotropic, so it doesn't care, but the inability for composites to handle out-of-plane tension is a fatal flaw in these locations due to limitations on the fiber orientations.

One thing I've been thinking about is manually braiding a very deep fabric for tip-to-tip which would reduce its tendency to peel apart. Unfortunately, that would be really slow, risk damaging the carbon fiber (which you can do by bending it too tightly), and it would do nothing to help the whole reinforcement from peeling off the airframe.

The other thing you could do is monster sized fillets, which defeats the point of composites which would then weigh more than an equivalent aluminum fin root.


The other other thing you could do is lay up a CF fincan layer by layer from back to front in a mold a la 3d printing. This could have arbitrary fiber orientations, except you would not be able to have any fibers oriented forward-aft. (those could be applied on the surface though)

The "monster fillets," is the path I chose. I did think about a CF fin can and actually want to try that for next years BALLS but with some twists…. more to come on that in the future. I am also skeptical that they will have ~ the same mass as a Al fin can. Not that cannot be made light (especially the Bare Necessities design looked light) but with all the t2t, additives, and metal leading edges I could see it getting close. I do however think it would be interesting to see you create a deep braided fabric but would be tedious and I am thinking the peeling off the airframe is the more likely of the two failures.

If you don't already know about it, you should read up on this project over at Altus Metrum:
https://altusmetrum.org/TeleScience/

YikStik 2 and 3 were designed with exactly this goal in mind, mapping nose cone and fin temps in flights above mach:
https://www.gag.com/rockets/airframes/2YikStik/
https://www.gag.com/rockets/airframes/YikStik3/

Temperature sensors were embedded in the fins and in a custom NC made by Shockwave. Fantastic build photos here:
https://gallery.gag.com/rockets/YikStik3/Build/

Yes I know about the TeleScience and have actually talked to Bdale about his brother doing this. I might work with them on something like this. I completey forgot I could use my Tele to acquire data from the sensors. Duh!
 
Last edited:
Oh yeah, what is your dry weight? Just curious...

Right now dry weight is ~15lbs. Rounded up.

Cool project. Thanks for sharing.

Thank you very much.

Thank you for posting your build thread. I really love threads like this and will be watching it closely.

Just please scour the internet for composite bonding techniques and anything you can find from high mach flights. There are a lot of myths about bonding on this forum and the best thing to do is to research the science not just "what he said" on this forum.

My two cents.

Thank you also. I couldn't agree with you any more. That is why I have started this thread https://www.rocketryforum.com/showthread.php?65106-Battle-of-the-Epoxies-A-Road-to-BALLS-Experiment take a look. I hope to find out as much information as I can, not just testing but also by reading.
 
It is possible to move up to better fabric for some of the work than what is typically used. Here is an example: https://shop1.r-g.de/item/190265 R&G in Germany is where I get most of my carbon fabric. Prices are much better than US sources for spread tow carbon fabrics.

I see you ran stability computations based on speed. Did you run them based on non-zero yaw angle also? The results might be different.

Gerald
 
It is possible to move up to better fabric for some of the work than what is typically used. Here is an example: https://shop1.r-g.de/item/190265 R&G in Germany is where I get most of my carbon fabric. Prices are much better than US sources for spread tow carbon fabrics.

I see you ran stability computations based on speed. Did you run them based on non-zero yaw angle also? The results might be different.

Gerald

I ran them also on different wind directions and alphas, yes. The results are different and it can still be flown safely if the alpha is not too high. It is all about watching the conditions and how they move the CP.

As far as the fabrics go. I use high modulus aerospace grade CF from Soller Composites. Thanks for the link and I will check them out!

EDIT:
Those fibers are 12K… that scares me lol. But look very nice.
 
Don't skimp on fin material. You need all the stiffness you can get. Real men use Dragonplate.

This is just me backseat driving, but I would shrink your fins down. By extending the root chord and shortening the span; your fins would be stronger, more aerodynamic and better for high speed stability.



Well put. I love people who embrace this mentality. I'll be going for my L3 in a similar fashion.

Carry on.

Alex

Thank you Alex. I spent about a month on the fins and have gone through at least 15 different fin designs. This one I feel offers the most performance and amount of stability (at least stability I am comfortable with). At higher velocities the CP moves faster and much more dramatically than at lower velocity flights. I am nervous with anything under 1.5 calibers of stability. That is all assuming an alpha of 90 degrees, if that changes then the stability and CP are different values and can ultimately become unstable. I am trying to prevent that. But I hear you lol.

Note: I also considered 4 fins, btw.
 
You probably don't want high modulus for this application. High modulus is brittle. Some fibers will break doing the T-T. Actually quite a few may break. That weakens the result. High modulus is also lower in strength. HT is quite a bit stronger and more tolerant of bends. Ditto IM.


Spread tow fabrics are exceedingly thin. These fabrics are two fibers thick. That is all. Their tows are spread to wide ribbons one fiber thick and then woven. There are far fewer fiber crossings so the fabric is not weakened nearly as much as is the case for conventionally woven carbon fabrics. Even using the same fiber, a spread tow version of the fabric will achieve a higher fiber ratio, greater strength, and greater stiffness. Of course, also at a greater cost.

Gerald
 
You probably don't want high modulus for this application. High modulus is brittle. Some fibers will break doing the T-T. Actually quite a few may break. That weakens the result. High modulus is also lower in strength. HT is quite a bit stronger and more tolerant of bends. Ditto IM.


Spread tow fabrics are exceedingly thin. These fabrics are two fibers thick. That is all. Their tows are spread to wide ribbons one fiber thick and then woven. There are far fewer fiber crossings so the fabric is not weakened nearly as much as is the case for conventionally woven carbon fabrics. Even using the same fiber, a spread tow version of the fabric will achieve a higher fiber ratio, greater strength, and greater stiffness. Of course, also at a greater cost.

Gerald

Yes they are more brittle, I forgot about that; that could be a bad thing in this case. Strength is defined here as the amount of force that can be applied to material before it breaks (fails). Often mistakenly interchanged with strength, think about stiffness as the amount a material deforms when a given force is applied. A stiffer material will deform less under the same force compared to a less stiff material. I made the mistake of using them interchangeably. So the STF spreads a tow in a thin and flat uni-directional manner like a piece of tape. Then weaves the tape into a woven "spread," and thus increasing the mechanical properties and lowering the weight. STF also allows for better wetting and I have also read that it is more restraint to delamination. This is something I cannot over look and I will most certainly make an order and play around with it. I have used 3k, 6k, 12k 2x2, 4x4 twills, sleeves and triaxial CF before but never STF. I do believe that the biaxial and triaxial CF has high strength properties but is harder to work with and its bending qualities are not as ideal (12k).
 
Cool project. Good luck!



Maybe consider it some more?...

Jeroen

Thanks. Luck… oh how I prefer the odds of probability (that are not on my side). :)

I thought about it for awhile, I must admit. Having 4 fins allows the fins to be smaller and reduces the aerodynamic loads that they will experience and thus reduces the chances of failure. However it does introduce more drag and thus takes away from performance. Right now it designed a tad bit more on the performance side. However it is summing 10kft shy of your new N5800 challenge lol. 3 fins are easier for me to achieve the desired performance and are easier for me to make (less materials and less work). I am not against having a 4 finned rocket, I just think this design is better suited for 3 fins. I need only be careful of the higher aerodynamic loads they will experience and that is what I am hoping to overcome.
 
Thanks. Luck… oh how I prefer the odds of probability (that are not on my side). :)

I thought about it for awhile, I must admit. Having 4 fins allows the fins to be smaller and reduces the aerodynamic loads that they will experience and thus reduces the chances of failure. However it does introduce more drag and thus takes away from performance. Right now it designed a tad bit more on the performance side. However it is summing 10kft shy of your new N5800 challenge lol. 3 fins are easier for me to achieve the desired performance and are easier for me to make (less materials and less work). I am not against having a 4 finned rocket, I just think this design is better suited for 3 fins. I need only be careful of the higher aerodynamic loads they will experience and that is what I am hoping to overcome.

There is a lot of anecdotal evidence in the hobby that three fins may need to be significantly larger in the Mach 4 region in order to be as effective as 4 fins. Something with the interaction of shock waves coming off the body tube limiting the effectiveness of the fins that are downstream.

Jeroen might have a little more experience in this regime than all of us put together...
 
The CP shift is more dramatic with Openrocket than Rsaero, at least in sims I've run.

Thank you Alex. I spent about a month on the fins and have gone through at least 15 different fin designs. This one I feel offers the most performance and amount of stability (at least stability I am comfortable with). At higher velocities the CP moves faster and much more dramatically than at lower velocity flights. I am nervous with anything under 1.5 calibers of stability. That is all assuming an alpha of 90 degrees, if that changes then the stability and CP are different values and can ultimately become unstable. I am trying to prevent that. But I hear you lol.

Note: I also considered 4 fins, btw.
 
There is a lot of anecdotal evidence in the hobby that three fins may need to be significantly larger in the Mach 4 region in order to be as effective as 4 fins. Something with the interaction of shock waves coming off the body tube limiting the effectiveness of the fins that are downstream.

Jeroen might have a little more experience in this regime than all of us put together...

Yes I believe I read something similar in a paper about turbulent fin structures and their interactions at Mach 4. Cannot remember exactly and cannot remember who it was by. I am not going M4 and will not be close for awhile.

I am sure he does; working for CTI would be very fun indeed.


The CP shift is more dramatic with Openrocket than Rsaero, at least in sims I've run.

In all the flights similar to this one I think OpenRocket has been a little closer to the predictions. Now for this build I am using both but I am using primarily OR, so if it is more dramatic (which I have yet to detect a dramatic difference per se) then that only works in my favor in terms of stability… if it is true.
 
Last edited:
In all the flights similar to this one I think OpenRocket has been a little closer to the predictions. Now for this build I am using both but I am using primarily OR, so if it is more dramatic (which I have yet to detect a dramatic difference per se) then that only works in my favor in terms of stability… if it is true.

CCotner and I trusted RASAero which declared that Bare Necessities would always have at least 2 calibers of stability... OpenRocket predicted 0 calibers of stability >M4.2. It went unstable.
 
Back
Top