building the "afternoon delight"

[joebot]

I've been building rockets for a long time, but this is my first high-power attempt. I am basing my build off of the PML "callisto" kit, and I'm incorporating their CPR 3000 system. This was my compromise between my urge to build from scratch and my conservative side telling me to take small steps. My modifications to the kit will be the CPR 3000, an extended booster section that will be large enough to house a full CTI 38mm 6XL, tip-to-tip glass, a Von Karman FG nose cone, and an ogive tailcone. The plan is to qualify lvl 1 using a smaller motor, but build the rocket so that it is strong enough to take a J motor for a level 2 qual attempt for the case that I want to do it all with one rocket. I've been talking with someone at PML, and they say that this kit should survive such a flight if I build it right.

I am aware that this will be a relatively high-altitude rocket, and plan on incorporating a tracking system (later on) for supersonic and high-altitude flights. I am also aware of the "low and slow" argument for first attempts - I don't want this thread to be another debate about that. My ultimate goal in rocketry is to go very high and very fast with big level 3 motors, so I'd like to learn as many lessons about that as quickly as I can.

The second reason for trying to build it to take a level 2 motor is that I think it will be great practice for the other rockets I have on deck (I managed to get my hands on the very last 3 Dogs Rocketry KestreL kit!). If I screw up the fiberglass, I'll still have a rocket that can get me my level 1 cert, and I'll have learned a lesson.

So far, things have been going great. I've got the fin can (mostly) all built up, and I'm about to start the glassing. I've never done it before, and I don't have any "rocket buddies" to spot me. I have, however, been reading up on the subject, and I'm generally "pretty good at stuff".

So, that leads me to some questions:

1. After reading around, I made the decision to not use JB Weld on the root fillets of the fin-motor tube interface. The motor tube is phenolic, and the fins are g10. Instead, I used BSI 30 min epoxy with milled fiberglass added (see pic). Was this a mistake?
2. I have obtained the following material for the tip-to-tip layup:
   • West Systems 105/206 Epoxy w/ pumps
   • 1 sq. yd. of Hobbico-brand 5oz. FG cloth in what appears to be a smallish, tight weave
   • 1 sq. yd. of Hobbico-brand "3/4 oz" FG cloth with a slightly smaller weave, seems very light weight
   • 6"x10yd West Marine 6oz. FG "tape," very large weave
   • Five 1" "chip bristle" brushes
   • two 16oz. plastic mixing bowls
   • Am I missing anything?
3. I don't know which glass to use, although I'm leaning towards the 3/4oz light stuff in two 45° layers. Thoughts?
4. Considering the radius of the fillets (~0.125") I have at the root of my fins and the weight/weave of the fiberglass, will I need some sort of weight to hold everything down while it sets? I don't want to mess with vacuum bagging (I have done zero research on that).
5. I should attach the tailcone after I glass the fins, right?
6. I need to overlap the fiberglass of the adjacent fin layups at the space fore and aft of each fin, right? then I just sand it flush, right?
7. How do i deal with the fiberglass that lays against the body tube forward of the fins - do I just sand the transition from fiberglass to bare bodytube into a slope? If so, do I then need to add a space between the rail guides and the body tube to account for the increased diameter of the fincan?
8. Similarly, do I need to sand the aft fiberglass transition to the tailcone?
9. I am planning on just laying the rocket on a table to do the glassing - I don't really need a fancy jig, do I?
10. What the hell is pullpeal, and do I need it? That's just a carbon fiber thing, right?
11. My KestreL kit should probably get carbon fiber. When I try to tackle this down the line, how different is tip-to-tip glassing compared to laying carbon fiber? I have read the Jarvis guide, FYI - I'm looking for additional experienced insight.
12. What are some things you wish you would have been told prior to your first attempt at fiber glassing?
13. What sort of paint best survives supersonic flights? I was thinking of using the high-temp spray-on bbq paint that home depot sells. the can says it's rated for 2000°F!
14. Is it worth it to fill the fincan with the good 2-part foam?
15. Do I need to shape (chamfer) the fins to break the sound barrier?

I'd really appreciate detailed answers if you guys don't mind - please assume I know nothing!

 

[Aksrockets]

Can't break mach with quantum tubing. fYI
The mach monster and QT dont get along very well.
Quantum tubing is great stuff. Durable, smooth, reasonably priced. It just shreds at mach .9 ish

 

[joebot]

Can't break mach with quantum tubing. fYI

PML says that it's ok as long as it's less than 2.6" diameter (this is a 2.1" dia rocket). There's no intrinsic property of QT that precludes the rocket from breaking the sound barrier. It's just not as strong as a fully-glassed phenolic frame. If you check rocket materials, you'll see that the axial crush strength of QT is >4,000lbs.

according to the PML FAQ on tube strengthening:

Therefore, PML kits 2.56" and larger should be reinforced for >950 fps or >650 mph
flight. We feel that kits 2.1” and smaller can be flown without body tube fiberglassing,
though all other items listed below should be done, as well as building the rocket with
plenty of epoxy and good sanding of areas to be bonded.

also, i emailed them about this. they said that i'll be approaching the limits of the kit's design limits, but if i build it well, it should be ok. if it doesn't survive, i'll just try with another rocket. the winters are long and boring here in colorado, and i have everything i need to build four rockets before it's flying season again...

Anyway, you've inspired me to run the calcs to see if my rocket will survive...

UPDATE:

ok, after rough calcs, it looks like the most stress my airframe will see is ~500 psi with a J600. add a safety margin of 5 to that (because i probably have no idea what im doing), and peak stress is ~2505 psi. according to rocket materials, the peak stress experienced for QT before crushing was 8500-9000 psi. I'm willing to say that I'm not worried about the strength of the airframe at this point... maybe it might buckle under some condition like warping caused by unsymmetrical heating, but other than that, i'm satisfied. finally, if you compare the data between glassed-phenolic and QT, it suggests that they are very similar in axial strength.

UPDATED UPDATE:
I should also point out that the modulus of elasticity for QT is HALF that of glassed phenolic. that means it's basically half as stiff. This probably explains why people say that they have seen QT rockets "bananna" in flight. I'm still going to try it anyway because meh...

 

[Aksrockets]

Besides that, The only real problem I see is that FG doesn't stick to QT that well. For this reason, PML offers Phenolic airframes on request.
Rough it up good and you should be fine.

 

[joebot]

ok. i still have a ton of questions regarding the whole procedure, though. I'd really like to glass it tomorrow.

 

[blackjack2564]

Ditch the CPR go with a standard Av-bay. Will take more punishment & is simpler to set up & use.

Rough up the QT with 80 grit so the glass will stick. It really is not meant to be glassed.

Use the 5oz for T-T first layer 1-1.5in. smaller than 2nd on edges of fin.
Use the 3/4 oz for veil if you wish, less filling for smooth surface. [really not needed if your finishing skill are up to par]

Peel ply and Vacuum not needed, but used by some.
Tail cone after glass, fill to level.

Rail buttons should work fine, they do for me, with out any needed spacers.

I would just use glass on Kestral.

If you do use high temp paint, don't prime underneath, will defeat the purpose. wet sand to polish finish. I did blister my finish with Mach 2 flights.

Fin can to small to gain anything by foaming. BUT if your NC is plastic, I would foam that.

Yes put a mean bevel on the fin edge & properly taper the T-T also.

Being an engineer you will probably appreciate what I did.

https://www.rocketryforum.com/showthread.php?t=22533

Good luck.

PS you can always line the inside of the QT with couplers to stiffen it up for serious Mach flights. It will greatly increase chance of success with high thrust motors

 

[Tominator 2]

I think "pullpeal" you are referring to is actually called peel ply. It's used on fiberglass and carbon lay ups to let the excess epoxy soak through so it can be soaked up by a breathing material. It also leaves a nice finish for adhesion. Or you can just sand it down some.

Carbon fiber is much more difficult to do especially on tip to tip. It is usually thicker than fiberglass and i have a difficult time telling where I have applied epoxy unlike fiberglass where it is easy to tell. I wouldn't start any CF layups until you have lots of practice with the fiberglass.
I recommend you do some practice runs before doing it on the project. That way if you mess up you won't be sorry.

 

[joebot]

PS you can always line the inside of the QT with couplers to stiffen it up for serious Mach flights. It will greatly increase chance of success with high thrust motors

this sounds like a fantastic idea! ill see if i can buy long lengths of coupler from PML. Thanks! i briefly looked at your thread - zoinks! - im going to read it in bed tonight.

 

[stantonjtroy]

this sounds like a fantastic idea! ill see if i can buy long lengths of coupler from PML. Thanks! i briefly looked at your thread - zoinks! - im going to read it bed tonight.

I did the same thing on my rocket with 4" QT at the erea where my DDejection charge is located. I've seen them blow out so.... Not mentioned but worth pointing out. QT does NOT like cold and the thermal expansion/contraction rate is rediculous.:2:

 

[joebot]

honestly, i bought the kit without doing much research on the QT. I actually had it confused with the blue tube stuff, and only realized my mistake after it was too late. oh well

Any tips on shaping the fins after they are attached to the rocket?

edit: coupler ordered!

 

[cbrarick]

glue some sandpaper to a block of wood, then take it slow....

 

[joebot]

i took a break from the booster section today to install a really nifty magnetic switch that i got from Featherweight Altimeters. I'm going to use this switch to arm my recovery electronics.

The avionics system i got for this rocket is the PML CPR3000. A lot of people don't like them, but I bought it before I knew that. Despite this, I decided to use it anyway since i have it on hand. It comes with a simple panel-mount switch made of plastic that you are supposed to mount outside the bodytube with a plastic standoff on 2.1" (54mm) kits. Since I plan on eventually flying this thing supersonic, I don't want the switch that controls my recovery system to melt off mid-flight. The mag switch (above) allows me to activate my avionics by waving a magnet over the airframe, and allows me to keep everything internal and out of the airstream.

Since I'm anal, I don't want to simply glue the switch inside a bulkhead that I will never be able to access again. I want to be able to inspect the switch and it's wiring, replace it if it becomes faulty, or remove it for use on another project if this rocket dies (it's like a $25 switch or something). Therefore, I am building an access hatch that will allow me to do all of these things.

The first step is to mount the switch to the aft CPR coupler. I did this by attaching some modified blind nuts to the CPR coupler. In order to make sure the switch sits tangent to the CPR center tube, I put the coupler on my drill press so that the axis of the mounting screws are perpendicular to the mag switch. To make sure my crappy cheap drill press doesn't let the bit wander off the side of the tube, I used a spring-loaded awl to pop three dents in the tube. These helped guide the pilot bit straight down.

Marking the CPR coupler with my awl.

The three guide pits made by the awl.

One of the modified blind nuts on the right. Note that I scratched up the barrel with my dremel so that the epoxy has something to hold on to.

After I drilled the holes, I test fitted the modified panel nut to make sure it fits and doesn't protrude into the interior of the CPR coupler.

test fitting. note that the holes are tangent, and not radial.

I then used the switch itself as a jig to hold all the nuts in place as the epoxy dried. I used a piece of tape under the switch to make sure that the switch didn't accidentally become a permanent part of the assembly.

drying...

Then I cut out my access hatch in the body tube. I used the "roll-paper-around-the-circumference-to-get-a-straight-line" technique to mark the lateral cuts, and an aluminum L-channel to mark the longitudinal cuts. The actual cutting of the lateral cuts was done with a razor saw, while the longitudinal cuts were made by repeatedly going over the line with an exacto knife.

Man did this come out well! (ok i practiced first on a piece of scrap tube)

finally, I test fit everything together.

note how little room there is... a fit is a fit, though, and everything clears. once I'm done, I'll have plenty of room to access the wires and inspect the switch.

now time to add fasting hardware to the hatch...

screw locations measured out

guide dimples in place...

back from the drill press. i used some countersinking bits i found at home depot for this. I placed the taped assembly on top of a scrap tube to hold it in shape as the drill press did it's job.

some simple 4-40 nuts epoxied to the back of the nutplate ought to hold the cover on just fine.

another test fit. my screws are too long, and i don't have an allen wrench that fits that tiny socket oh well, it's starting to look really cool, if i do say so myself!

EDIT: done!

 

[dixontj93060]

Nice work on the hatch.

 

[patelldp]

I will just leave this here

[YOUTUBE]HQwzX8oLUnw[/YOUTUBE]

 

[joebot]

hahahahhaah thanks - those last two posts made my day!

 

[Handeman]

ok, after rough calcs, it looks like the most stress my airframe will see is ~500 psi with a J600. add a safety margin of 5 to that (because i probably have no idea what im doing), and peak stress is ~2505 psi. according to rocket materials, the peak stress experienced for QT before crushing was 8500-9000 psi. I'm willing to say that I'm not worried about the strength of the airframe at this point... maybe it might buckle under some condition like warping caused by unsymmetrical heating, but other than that, i'm satisfied. finally, if you compare the data between glassed-phenolic and QT, it suggests that they are very similar in axial strength.

I don't think the crushing strength is what is lacking in the QT for Mach + flights. I believe the issue is it's strength and stiffness when passing through Mach and the fins begin fluttering. Even if the fins are stiff enough to handle it, the QT BT will probably be the weak point.

I would suspect the reason the small tube QT doesn't have the same restrictions is the smaller size gives better thickness to span ratios between the fins and holds up better.

Just my :2:

Good Luck

 

[patelldp]

I don't think the crushing strength is what is lacking in the QT for Mach + flights. I believe the issue is it's strength and stiffness when passing through Mach and the fins begin fluttering. Even if the fins are stiff enough to handle it, the QT BT will probably be the weak point.

I would suspect the reason the small tube QT doesn't have the same restrictions is the smaller size gives better thickness to span ratios between the fins and holds up better.

Just my :2:

Good Luck

I do believe it's the lack of sufficient bond between the G10 fins and the QT. compression strength of the tube is fine.

 

[joebot]

I don't think the crushing strength is what is lacking in the QT for Mach + flights. I believe the issue is it's strength and stiffness when passing through Mach and the fins begin fluttering.

well, crush-strength is the most important metric regarding tube strength in the case of rocketry. not to get tangential, you can have a very strong material that buckles under loads significantly less than that material's yield strength. axial crush strength takes that into account (mostly - tube length and surface finish would play into this also).

stiffness, on the other hand, is described by the materials modulus. i mentioned in the update of that post that QT is half as stiff as glassed phenolic. so yes, QT's stiffness (or lack there of) would play a big part in the rockets ability to survive the highly-dynamic environment of trans/super-sonic flight.

I would suspect the reason the small tube QT doesn't have the same restrictions is the smaller size gives better thickness to span ratios between the fins and holds up better.

almost right on, but it doesn't have to do with the between-fin span. it has more to do with the fact that the wall thickness more or less remains constant through out PML's standard tube size schedule. let's look at an example:

because the wall thickness for both the 2.1" and 3.9" QT is almost the same (0.063 vs. 0.062, respectively), the 2.1" has more "cross-sectional area per diameter" if that makes any sense.

x-sectional area = [pi*(OD/2)^2] - [pi*(ID/2)^2]

2.1":
ID: 2.152"
OD: 2.278"
x-sectional area: 0.438 in2

3.9":
ID: 3.637"
OD: 4.076"
x-sectional area: 0.383 in2

when we evaluate how much stress material experiences given an applied force, we divide the incident force by the area under stress to normalize that value. regardless of the incident force, you can see that the 3.9" tube will experience more stress (mathematically described as force / area):

x <- any force size, be it 1,000,000 lbf, or 500 lbf

which is bigger, (x lbf) / (.438 sq. in.) or (x lbf) / (.383 sq. in.) ? the answer will always be the later. note that the resulting units is pounds per square inch (aka stress).

hence, the smaller diameter in this case is stiffer and stronger than the larger diameters. also, later on in the discussion, i mention that i plan on reinforcing all my body tubes with phenolic coupler tube. that ought to give me tons of extra margin on the stiffness/strength required to survive a supersonic flight.

 

[Handeman]

.. but it doesn't have to do with the between-fin span. it has more to do with the fact that the wall thickness more or less remains constant through out PML's standard tube size schedule.

I agree with your calculations and the premise that the smaller diameter tube with the same wall thickness is going to be stronger then the larger diameter one. Unless I'm mistaken, you are saying that extra strength from the small diameter tube thoughout the entire length of the BT is the difference.

That may be the case, but I was thinking that even the larger diameter QT tubing is strong enough for Mach speeds between the nose cone and fin can. The forces there are primarly crush forces between the motor thrust and the drag and weight of the nose cone. It is when you get to the fin can and you add the extreme twisting and torque forces from the fins that the QT doesn't hold up. I just didn't think determining the strenth needed over the whole BT is sufficient, I would think you have to look at the non-linear forces the fins apply when they flutter to get a real feel for the failure point.

 

[blackjack2564]

Just some thoughts:

You should be OK with this size airframe and most 38 motors.

The weak point is the CPR as in all rockets that use it. There simply is not enough Av-bay going into the airframes for proper support. Extreme forces & leverage take over & that's when it breaks.
The fincan/payload joint is where they all let go.

I helped an employee of PML prep for their L-3 flight with a carbon fibered, heavily modified 3in kit. He should have known better. The CPR module only extended 1.5 in. into the fincan. I recommended he take the time and swap it out for a conventional av/bay with standard length, to no avail.
Care to guess where it failed/folded under thrust? A lot of expensive hardware went to never, never land.....never to be seen again. LoL

Normally you want at least a caliber of airframe extending into the body. I.E.: if airframe is 3in. then 3inches of couple/av-bay extends into payload & 3in into fincan. The threads on CPR are woefully lacking depth for this. This is a "sport rocket" and is designed for under .85 mach flights. Yes with some beefing up you can pull off mach+ in this diameter.
It's great stuff when used as it was intended, everyday sport flying.

Larger QT tubes rarely if ever [I've never seen one] survive mach.
Wanna turn a 4in QT rocket into confetti? Stick a K-1100 in it.
Just this past season I have witnessed 3 QT rockets [1-3in & 2 4in] destroyed.

Think of it like this: you can stand on a beer can,BUT if someone pokes at the side with a finger it will immediately collapse. Under high thrust flights this material [QT] is subjected to wind shears and off angle stresses, rarely is a flight perfectly vertical.
Putting the stiffy couplers inside will help eliminate the "poke in the side"

Cold with weaken QT by making it brittle.
Hot days 85 and up... if painted dark color surface can easily become 110 and above, will soften QT and really weaken it.

I may not be able to "run the numbers" to prove things, but practical experience & thousands of witnessed flights have it's advantages.

One of the best mods you could do to pull this off besides using couplers internally is to sim the project while cutting down the fin tips. I'm going to guess you can cut at least 1in. off the span and still maintain stability. I would whack off as much as possible.

Just fly the darn thing, don't push the limits too far and enjoy this kit. Save the "pushing the envelope" for materials made for it.

Or let us know when you find the speed of QT.

Ps: I have developed a real simple mach ready test. If you are willing to lay your rocket on a hard surface & stand on it, take a bounce or 2, without damage, your odds for mach success just went up a few hundred %. Lol

 

[joebot]

The CPR module only extended 1.5 in. into the fincan.

you are talking about the phenolic coupler that links the booster section to the recover section, right? (see below)

i was planning on making the following modification to the booster section coupler (see below). basically, the stiffy goes all the way down to the centering ring, and then protrudes beyond the QT body tube by at least 2 calibers. the coupler PML provides with the kit only gives you about 2 inches of coupler interface (which i guess is about one caliber).

do you think that ought to be strong enough, blackjack?

this leads me to another important design consideration. if i line the entire QT rocket with stiffy tubes, I lose the ability to use PML's piston system (because it's basically a free-sliding coupler with a bulkhead in it). what is your opinion of the piston system? is it worth keeping, or should i just get some nomex wading?

EDIT:

Think of it like this: you can stand on a beer can,BUT if someone pokes at the side with a finger it will immediately collapse

this is what i was talking about when i mentioned buckling earlier, and why i completely agree with you that i need to add the stiffy tube everywhere i can! QT is strong, but it isn't very stiff at all - they really created a funny material...

but practical experience & thousands of witnessed flights have it's advantages.

that's why i'm here - I want to learn from vets like yourself! thanks for all your help!!

 

[Handeman]

this leads me to another important design consideration. if i line the entire QT rocket with stiffy tubes, I lose the ability to use PML's piston system (because it's basically a free-sliding coupler with a bulkhead in it). what is your opinion of the piston system? is it worth keeping, or should i just get some nomex wading?

I have the PML Callisto and I like the piston system. I've added a piston to a scratch built rocket because of my experience with the Callisto. With that said, I wouldn't keep the piston if I had any idea that I needed to use the stuffy tubes. The pistons have pluses and minuses and are not necessarily better then other types of deployment methods. If you need the extra strength, and it certainly seems like a great idea in your application, I would drop the piston and not think twice about it.

 

[joebot]

ok thanks! that's what i'll do! o7

 

[joebot]

i finished the CPR coupler today:

I think it looks really cool. I like that it makes the rocket seem like it's more than just some cardboard tubes glued together - very show-off-able.

the recovery section joined together. during high-speed flights, I think I'm going to cover the o-ring with aluminum tape to keep it from melting out.

it's a very sturdy joint, with zero-slop. it seems way more solid than a simple coupler joint.

 

[Handeman]

i during high-speed flights, I think I'm going to cover the o-ring with aluminum tape to keep it from melting out.

I don't think you'll have to worry about that. I suspect the melting point of the QT tubing is lower then that of the o-ring. Besides, I don't think you'll be able to get that rocket fast enough for long enough to have any heat issues with the o-ring.

 

[joebot]

anyone want to take a look at my open rocket file? i've decided to chop 1.5" of of the fins due to a recommendation made earlier in this thread - but it's made me nervous. as far as I can tell, the sim says that this thing is stable, but im a noob and want a second set of eyes on this. particularly, is this rocket stable as it leaves the rail? thanks for your help.

btw, the "short body" is in reference to an alternate configuration i've been testing called "long body." basically, it has an extra 12" of airframe for recovery electronics. I really don't think I need it though.

View attachment Callisto_joe - short body.ork

 

[blackjack2564]

First off when running sims you should correct for proper rail length & temperature.

That will give you velocity at rod/rail clearance.


I ran multiple sims from a small H-123 to a large 6xl grain load.

You have plenty of stability in all ranges with the rocket shown. [as I predicted]

The shorter /lighter reloads move CG to rear. Larger/longer reloads will move it forward. Always check with smallest & largest load to be flown.

Here are some screen shots showing different rod lengths. Default is 39 in., it should be changed to what you will be flying on. For me 8 or 10 ft. Also default temp is 59 degrees. Change to a more realistic profile to get more accurate sims.

I changed the rail length from 39 [default] to a more realistic 8ft and your off rail speed goes up from 80fps to 140fps.

In winds less than 10mph you can get away with a thrust ratio of 5-1 over 10 you should use 10-1 or greater. Otherwise most rockets will suffer from extreme weather cocking. Longer rails also improve this problem by giving you a higher velocity before you lose guidance.

 

[blackjack2564]

If you really want to "play", optimize your design by changing the rocket length & fin span, then running the drag coefficient vs mach numbers.
Use the various motors planned to fly.
This will yield a usable safe flight range.

Kits almost always come over stable, designed for the widest range of motors & conditions. Play around & find what suits ya!

The shorter airframe and smaller fins will give you better mach speeds.

Your tailcone is getting you an extra 500-800 ft in altitude with larger motors & flights over 6000ft & some extra velocity. Probably the best thing you have done so far. [cut down on base drag]

Take off the lugs, launch from a tower, use a polished finish and get another 5-800.

This one did much better than sims predicted.... No tip to tip. Just surface mounted fins.

Went almost 23,000, over mach 2 .

Stability was only 1.3


Less is better with rockets. If it ain't needed...get rid of it!

 

[joebot]

hahaha thanks for the great advice! i think you are totally right about going overboard... i've been planning my fiberglass layup all night

 

[joebot]

what do you think of this? the idea is to try and create a 3D airfoil with three layers...

i experimented with the various FG cloth i have. the hobbico stuff is JUNK! it was stiff as a board. luckily i have some 6oz from PML that is real nice.