Carbon Prepreg Airframe build for the Loki 54/4000 - Fins Attached

[watheyak]

Nice Wayne! Wanna do a 2-stager??? Thanks for the pics, they're inspiring.

There is no progress since the last post. I've been super busy with vacation and new work stuff. But I can feel the itch. I'll be full blast again here soon. I don't think it will be ready for Octoberfest, but end of this year or early next, I hope. Weekends are getting scarce lately!

 

[jus_rockets]

What are the plans for the electronics, deployment setup, tracking, fincan? I really hope you can tame that motor and break a record!

 

[watheyak]

What are the plans for the electronics, deployment setup, tracking, fincan? I really hope you can tame that motor and break a record!

Hey Juan,
The electronics are as yet undecided. They need to be tiny and include tracking. I want to mess with some combinations of separate items to see if I can get it smaller that a combo unit like a Telemetrum or Aim XTRA.

Deployment is unconventional to say the least. A few years back I did a bunch of ground testing of rear ejection of long motors. Even with a motor this long, at the time I was hoping to use a Kosdon L-3000, and it worked just fine. This allows a one piece airframe. The Av-bay screws into the tip of the nose with a loooong tool. The advantage is a strong, seamless joint between the airframe and nosecone. As well as no airframe that isn't supported by either motor or coupler, as the coupler is the thrust ring.

Weird, I know. Lots of testing and experimenting was done, even taking into account expansion of the motor case so it doesn't get stuck at deployment. Dual deploy with a cable cutter, unless it flies at Black Rock, but as of right now it looks like it'll fly at Eagle Eye. Maybe even then single deploy....

As for the fins, they're complicated, too. Many layers of prepreg of decreasing size to make a biconvex airfoil. And spread tow tip-to-tip. With this stuff-


A tall order. I gotta take my time with it. More to come soon!

 

[jus_rockets]

Hey Juan,
The electronics are as yet undecided. They need to be tiny and include tracking. I want to mess with some combinations of separate items to see if I can get it smaller that a combo unit like a Telemetrum or Aim XTRA.

Deployment is unconventional to say the least. A few years back I did a bunch of ground testing of rear ejection of long motors. Even with a motor this long, at the time I was hoping to use a Kosdon L-3000, and it worked just fine. This allows a one piece airframe. The Av-bay screws into the tip of the nose with a loooong tool. The advantage is a strong, seamless joint between the airframe and nosecone. As well as no airframe that isn't supported by either motor or coupler, as the coupler is the thrust ring.

Weird, I know. Lots of testing and experimenting was done, even taking into account expansion of the motor case so it doesn't get stuck at deployment. Dual deploy with a cable cutter, unless it flies at Black Rock, but as of right now it looks like it'll fly at Eagle Eye. Maybe even then single deploy....

As for the fins, they're complicated, too. Many layers of prepreg of decreasing size to make a biconvex airfoil. And spread tow tip-to-tip. With this stuff-
View attachment 183344

A tall order. I gotta take my time with it. More to come soon!

It looks like you got mostly everything figured out and have past experience using those methods. I like the way you think outside the box! I hope to learn a few tricks from this thread, good luck and keep us posted.

 

[watheyak]

Ok, in the past few months, a lot has changed with the design of this rocket. I spent some time at Balls talking with Curt Von Delius about my ideas, as well as the background of some of his altitude record rockets.

Above K impulse, weight is the enemy. This seems to have been backed up by many simulations in Openrocket, Rocksim and Rasaero. So for that reason I am switching to a flying case design. Less outside the box, and easier as well.

So I got to making an actual part. This past weekend I made the tube for the fin can. With the practice tubes in the past I had difficulty tightly wrapping the prepreg carbon around the mandrel, as it wouldn't stick to the mold release. So I devised a clamp with angle aluminum so hold the carbon on the mandrel while I tightly wrapped it around for the first layer. It sticks to itself, so this was not an issue on subsequent layers. I also left the backing on the carbon for stability while wrapping. Here's some pictures;


The clamp.


Wrapped around about 370 degrees, stuck to itself, clamp removed. Backing being peeled away as it's wrapped.


Tube wrapped in heatshrink tape. Much tighter than previous tries, about 1/8" spacing. About to go in the curing oven.


Out of the oven. As the aluminum mandrel cooled I could hear it "un-sticking" itself from the carbon tube. Once it cooled, the mandrel literally dropped out of the carbon tube. Awesome. I will never use a mold release other than Frekote 700. Ever.


Finished tube on the motor case. Those are raw, unfinished edges. Nicest tube I have made from the prepreg so far and is very strong though less than half the wall thickness of commercial tubing. In a super-scientific squeeze test it is stiffer than filament wound carbon tubing from PR, as well as stiffer than Shadow Composites tubing. Which is cool since it's so thin, right? But the Shadow Composites stuff flexes in a good way I think. My tube would shatter and crack, while the Shadow stuff would be pretty resilient. Both are okay, I just need to make sure I'm using this stuff to it's advantages.

 

[CarVac]

There is no need to worry that much about brittleness because it is always going to be on the tube, so it would actually fail by tension if the fins manage to stay on.

 

[flynfrog]

That looks really good scott. Be careful on your prep work not to break the fibers uni can unwind from the tube easier than cloth. Your weight to strength and stiffness ratio should be much better than anything I have seen posted on here.

 

[REK]

Man I really need to make me an oven and make some carbon tubes with more layers.

Nice work on that tube.


Alexander Solis

TRA - Level 1
Mariah 54 - CTI RedLightning- I-100 - 6,345 Feet

 

[watheyak]

Thanks flynfrog, that that means a lot.

Today I'm spending some time cutting the tube to length and working on a boattail/thrust ring concept.

Up next are fins. Curt and I are going to make plate using the prepreg for our respective projects. Curt has been doing a lot of leg work on the plate fabrication, including a cut list drawing in Solidworks for the quasi-isotropic layers.

Curt also purchased some caul plates to facilitate the smashing of the stack of prepreg. They're 7.5" X 21" X .75" thick made from Alcoa Mic 6 precision machined cast tooling plate. Very cool!



After the fins are cut from the plate, they will be clamped to a sine plate and milled to an as-yet-to-be-determined diamond airfoil. I have abandoned the biconvex airfoil. Sine plate-

 

[flynfrog]

If you have machining capability why not cut a mold and mold your fins net?

 

[watheyak]

If you have machining capability why not cut a mold and mold your fins net?

That was the original plan. I was going to mold the fins using successively smaller shapes cut from the carbon, putting them in a vacuum bag with a very thin breather and then smashing them between two female molds of the biconvex airfoil made from MDF. Complex, right? I'm having doubts about how precise I can make the fins. It might work great, I'll just have to try it to see.

In addition, the wedge airfoils are better for mach flights than the biconvex. And if all I have to do is a wedge, that should be pretty easy with the sine plate and manual mill that I have free access too.

In the end though, I would love to have the biconvex even though it's not ideal. I really don't want a 1/4" bevel like every fiberglass kit ever produced.

I'll sleep on it.

 

[watheyak]

Finally got some time to work on this project. Made the carbon fin stock today. 24 .005" layers stacked quasi-isotropic.


The goods


Cutting pieces


Stacking


In the bag


Vac Pulled


Setting up the plates


C-Clamps


Torque Wrench on the Clamps. Much measuring ensued.


Ready for oven


The oven

It'll be done and cool tomorrow morning. Keeping my fingers crossed.

 

[watheyak]

Forgot an in the oven shot...

 

[REK]

Is that a kitchen oven?

 

[watheyak]

Forgot an in the oven shot...

View attachment 249455

Formerly a kitchen oven, now a mad scientist oven.

 

[Aksrockets]

Excellent Composites work! It's great to see a higher level of composite construction other then the usual 5.7oz carbon and Aeropoxy.

Alex

 

[ksaves2]

All's I can say is Holee Guacamole!! Beautiful work. Kurt

 

[watheyak]

Here's the completed plate-


In the bag


Unwrapping my Christmas present


Completed plate

It is wicked. Dense, with lots of sustain when tapped. Unbelievably, the plate measures .125" +/- .001" or less!! No voids or imperfections.

Mission accomplished!

 

[thobin]

Here's the completed plate-

View attachment 249471
In the bag

View attachment 249472
Unwrapping my Christmas present

View attachment 249473
Completed plate

It is wicked. Dense, with lots of sustain when tapped. Unbelievably, the plate measures .125" +/- .001" or less!! No voids or imperfections.

Mission accomplished!

Dang! That's some fine work!

TA

 

[Just_Keep_Flying (Masonh)]

Can I have one too?? I swear I've been good this year!

 

[maxvelocity]

This is an awesome project!

 

[watheyak]

Bad news!

The plate has a SLIGHT twist in it. Time to do some more research into ply orientation and hopefully fix it next attempt.

 

[REK]

Bad news!

The plate has a SLIGHT twist in it. Time to do some more research into ply orientation and hopefully fix it next attempt.

What kind of SLIGHT twist? Sorry to hear about that.

 

[watheyak]

What kind of SLIGHT twist? Sorry to hear about that.

Opposite corners are lifted about 1/16".

Still not sure what the issue is, but I'm speculating there are some tension issues in the layup after it cooled. I have much reading to do.

 

[REK]

Opposite corners are lifted about 1/16".

Still not sure what the issue is, but I'm speculating there are some tension issues in the layup after it cooled. I have much reading to do.

Awe man, that is strange.

 

[flynfrog]

Opposite corners are lifted about 1/16".

Still not sure what the issue is, but I'm speculating there are some tension issues in the layup after it cooled. I have much reading to do.

the issue is with uneven heating between the tool and the carbon. The tool will grow more than the carbon thus building stresses in the part. Using Steel plates will lessen this a bit. Ideally Invar if you have a NASA budget. A carbon tool would also help.

(this assumes your ply stack is balanced)

What are you doing with the clamps? You obviously have access to a vacuum pump why not put caul sheets inside the bag.

 

[watheyak]

the issue is with uneven heating between the tool and the carbon. The tool will grow more than the carbon thus building stresses in the part. Using Steel plates will lessen this a bit. Ideally Invar if you have a NASA budget. A carbon tool would also help.

(this assumes your ply stack is balanced)

What are you doing with the clamps? You obviously have access to a vacuum pump why note put caul sheets inside the bag.

Just to clarify, you're saying the problem stems from the aluminum tooling plates expanding and causing stresses in the layup, not that the plates actually warped?

The ply stacks were balanced and alternating 0, 90, +45 and -45. (I forget the exact order) The clamps were to provide pressure beyond what the bag was providing. Extreme care was taken to ensure that all 4 clamps were providing equal pressure, but I can see how this might have at least been overkill, if not the cause of the warping problem.

This weekend we're going to square up one end with a diamond burr and check for voids.

 

[watheyak]

So Curt and I have both been doing lots of reading about internal strains in layups. Pretty dry reading, but a picture of the cause is starting to form.

This from Curt-

So far with everything that I've read, I suspect that the diagonal plies need to be at the center of the plate to reduce their possible contribution to diagonal stress. I.E., I don't think the 0, and 90 plies cause twist.

Following this type of schedule may be our best bet: Quasi Isotropic 0(3), 90(3), +45(3), -45(3), -45(3), +45(3), 90(3), 0(3)
or possibly: 0, 90, 0, 90, 0, 90, +45,-45,+45,-45,+45,-45, -45,+45,-45,+45,-45,+45, 90, 0, 90, 0, 90, 0


We're going to try the second layup schedule, with alternating 0,90s.

 

[flynfrog]

My background is more in processing not so much the internal laminate stress I generally go alternate 0 and 45s where possible. It is possible to have issues with stresses from tool CTE mismatch where part of the tool will grow faster than other parts and add internal stresses in your part. Your clamps are probably not helping this as its not letting things grow and shrink freely. You could buy some stainless steel shim stock and use it as a caul sheet inside your bag. Your pressure from vacuum is orders of magnitude greater than what those clamps will do.

You might find your plate straightens out after you cut it. The uneven edges can cause warpage. [/wordvomit]

 

[bobkrech]

Forgot an in the oven shot...

View attachment 249455

Bad news!

The plate has a SLIGHT twist in it. Time to do some more research into ply orientation and hopefully fix it next attempt.

Opposite corners are lifted about 1/16".

Still not sure what the issue is, but I'm speculating there are some tension issues in the layup after it cooled. I have much reading to do.

the issue is with uneven heating between the tool and the carbon. The tool will grow more than the carbon thus building stresses in the part. Using Steel plates will lessen this a bit. Ideally Invar if you have a NASA budget. A carbon tool would also help.

(this assumes your ply stack is balanced)

What are you doing with the clamps? You obviously have access to a vacuum pump why note put caul sheets inside the bag.

Just to clarify, you're saying the problem stems from the aluminum tooling plates expanding and causing stresses in the layup, not that the plates actually warped?

The ply stacks were balanced and alternating 0, 90, +45 and -45. (I forget the exact order) The clamps were to provide pressure beyond what the bag was providing. Extreme care was taken to ensure that all 4 clamps were providing equal pressure, but I can see how this might have at least been overkill, if not the cause of the warping problem.

This weekend we're going to square up one end with a diamond burr and check for voids.

My background is more in processing not so much the internal laminate stress I generally go alternate 0 and 45s where possible. It is possible to have issues with stresses from tool CTE mismatch where part of the tool will grow faster than other parts and add internal stresses in your part. Your clamps are probably not helping this as its not letting things grow and shrink freely. You could buy some stainless steel shim stock and use it as a caul sheet inside your bag. Your pressure from vacuum is orders of magnitude greater than what those clamps will do.

You might find your plate straightens out after you cut it. The uneven edges can cause warpage. [/wordvomit]

It has nothing to do with your plies. Your plates are aluminum, you raised the aluminum plates off the oven grates with c-clamps so the ends were unsupported, you used clamps instead of vacuum bagging, and you cured the composite in a home oven that heated from one side only. All of these factors contributed to the warping.

Aluminum has a very large coefficient of thermal expansion. Raising it off the oven grates leaves the ends unsupported, allowing the aluminum plates to bend. The clamps also point load the plates. Because the heating in the oven is from the bottom, and it's not a convection oven, the bottom plate will be hotter than the top plate and will bend more, warping the aluminum. The composite is no longer flat, and once cured, will remain curved.

The best plates are glass, preferably borosilicate glass at least 1 cm thick. The coefficient of expansion of glass is at least an order of magnitude lower than any metal. By vacuum bagging the glass plates you have a very stiff package and will not warp if laid on the oven grates, although the ideal situation is to use a convection oven to make the oven temperature uniform.

Bob