Making plans for StratoSpear again

[Adrian A]

Last year at Balls I attempted a 3-stage 38mm flight, which I documented here (in the wrong forum). I decided to go for it again, this time with all Loki motors:


It's mostly motor.

I was astonished by the results from RASAero:



This is for 10 mph crosswind at launch and 1 degree of additional adverse tower angle.

On a 3-stage rocket with potentially long ignition delays, initial conditions are very important, so I looked at a range of simulations with different delays, initial tilt angle, crosswind, etc., and got the following for my current planned Balls rocket:



As I suspected, dowrange distance is primarily a function of the angle at ignition. If 15 nmi is acceptable, then I could keep my usual plan of sustainer ignition at 12 or 13 degrees, or the optimal vertical velocity, whichever comes first. Then again, the horizontal velocity won't go down much until it gets back under 100,000 feet, so it would be unwise to use too much of the budget getting to apogee.

Here's the apogee altitude vs. total ignition delay, for a variety of initial angle conditions:



I didn't run cases with long ignition delays for very bad initial tilts in the above plot because I should be able to avoid that with the Blue Raven's future angle ignition trigger feature. But generally speaking, longer delays give higher altitudes for this rocket, if you can put up with the downrange distance.

 

[StreuB1]

Planning on being at Balls this year, either with a project or not. Can't wait to see this in person!!

 

[charrington]

I'm gonna have to try and make it out as a spectator! I don't want to miss this.

 

[manixFan]

Having a vendor who pushes the limits with his own gear greatly improves what the rest of us get to use. Also looking forward to seeing this at BALLS.

I’ve said it before, and I’ll say it again: for such a small hobby we have an incredible number of dedicated vendors who design equipment far beyond what the potential market would seem to warrant. For any price point there is something for everyone, with a wide array of technologies to choose from. (And not just electronics, nearly any aspect of the hobby.)

Keep up the great work!


Tony

 

[kjhambrick]

Adrian --

What is the discontinuity at about t = 119 sec in the Mach Number -vs- Time Plot ?



Is it a transition into the 'Stratus Spear' ( ) ?

Kool Stuff. Wish I could be there.

Thanks.

-- kjh

 

[Chuck Rogers]

Adrian --

What is the discontinuity at about t = 119 sec in the Mach Number -vs- Time Plot ?

Konrad:

It's the parachute opening. The RASAero II parachute model is for a rocket in nearly vertical flight with the parachute opening at apogee, not the rocket under the parachute at Mach 0.5 continuing a parabolic trajectory through apogee (the result of an increased launch angle from vertical) before slowing down and transitioning to a vertical (nearly vertical) descent. With a better model the Mach number with time would look fairly similar to that above (you are of course opening a parachute), just not nearly as steep, not a vertical discontinuity, especially at high altitude (152K ft) with a low dynamic pressure.


Charles E. (Chuck) Rogers
Rogers Aeroscience

 

[kjhambrick]

Thanks Chuck.

The graph makes perfect sense now.

-- kjh

 

[Adrian A]

I just finished writing up my class 3 application for the flight. Holy smokes, that was a lot of work. Almost all of it is stuff I was planning to figure out and document later though, so it's good to have it done this early. After I make required changes required (if any), I'll post the package here if that's allowed.

 

[Adrian A]

The package is getting reviewed, but I'm in the mood for some building, so I'll get started with the build thread.

Last summer I built a nosecone that I have been using on my 38mm rockets starting with BALLS. It has survived a number of flights including some hard landings, but it's a lot heavier and longer than it needs to be because I overdid the casted nose weight. Also, when I cast the noseweight I embedded a hanger bolt for the harness, but the opening was so small I could only thread through two loops of 350 lb Kevlar and I have been eyeing the wear and tear on this harness suspiciously.

This time I'm designing for a longer sustainer body and approriately-sized fins, so I won't need so much mass up front. I'm also revising my chute cannon and nosecone ejection design, which puts the GPS tracker closer to the tip than the current nosecone. Here is the plug I used about 12 years ago to make my nosecone mold:

It started as a Von Karman nosecone (Sorry I don't remember where I got it 12 years ago) with a long L/D ratio, which is an under-rated feature for supersonic altitude shots IMO. I added a scrap piece of airframe to extend the nosecone into the cylindrical region, then filled and sanded everything to a glossy finish. The mold (not shown) is silicone cast with this plug in the middle inside a 54mm F/G tube that I later slit lengthwise. The outer tube keeps the silicone mold straight, and the mold's flexibility lends itself to easier mold release. I would recommend this approach except that doing the layup inside such a long, skinny mold is a bit like making a ship inside a bottle. I made last fall's version of this with braided carbon sleeves, which usually make things easier, but I used Cotronics 4461 which is super sticky and viscous, and it drove me crazy trying to get the sleeve to lay down inside the mold because my tool kept sticking to it and pulling it out. I'm hoping that these triangles will be easier because I can keep plastic on the inner side to keep the tool from sticking until the carbon is stuck to the mold surface. (I hope) The 0/90 fiber direction is also more optimal for this application than +/-45.

The line across the template is where the carbon will transition to fiberglass for radio transparency. That's where my GPS antenna will be. A lot of the LoRa whip antenna will be forward of that boundary under the carbon section, but based on the last nosecone it should be o.k. to transmit out through the RF window section. Behind the FG section is a pre-made carbon tube. This is cut off from the tube that will become the sustainer airframe, a 1.500 ID thin-walled mostly-unidirectional tube from RockWest composites. With a 1.500 ID, I needed to use a flap sander to remove some material on the inside in order to fit one of my 3 Loki cases. An Aerotech case can be squeezed in also. My CTI case can't. This close fit should help keep the nosecone better centered than it was on my shortest sustainer body. With my current plan, the motor case will be most if not all of the coupler interface for the nosecone. Part of the idea here is to avoid anything (an aluminum nosecone tip edge, for example) that will prematurely trip the boundary layer turbulent. I haven't calculated the Reynold's number but I think there is a chance I can keep the boundary layer laminar all the way to the back edge of the cone, at least at the lower speeds. That reduces both drag and aeroheating, which will be no joke for this flight.

 

[Adrian A]

For vacuum bagging the noseone layup inside the mold, I use a silicone inner bag that I had cast around the mandrel way back when I made the mold. It's frustrating to use though, because it's necessary to seal the opening to the rest of the bag, but nothing sticks to the silicone, including the sticky vacuum bag tape. Last time I ended up sort of squishing the end to the bag with some weight. This time I used some scrap fiberglass plate to make a 2-part flange that clamps the edge of the inner bag between the two pieces of fiberglass. Then I can seal the bag to the fiberglass.



I prepared my layup material. At the top are some blue pieces of peel ply, and two sections of braided FG sleeve, and at the bottom are carbon fiber pieces. The triangles are cut out of unidirectional material, and on the right is plain weave.




I mixed up the resin and poured it out onto the CF pieces on a cut-open garbage bag, and then laid the rest of the garbage bag on top, then squeegeed the resin though the CF between the two clear plastic sheets. That made it relatively easy to cut out the pattern from the cloth without the fibers shearing around.



I only got as far as laying down 4 wedges of the CF cloth before I ran out of time before a meeting. So I put a couple of wedges of peel ply down into the mold and bagged it. I like to just use a garbage bag, toss everything in there, put a vacuum attachment on, and then twist and clamp the open end of the bag. It beats endless application of vacuum tape. The new inner bag worked like a charm, and I had good vacuum pressure right away.



Either later this afternoon or tomorrow when I think it's cured enough to safely pull out the peel ply, I'll do another layer of carbon, then the FG sleeves in the midsection, and then the pre-cured CF tube at the bottom. For the carbon at the front of the nosecone, the 4 wedges that I put down in there this morning have several sections with little/no overlap between them. And now that the front of the cone is all black, it's hard to see where I would need to put more cloth wedges. So I'm thinking of going back to biaxial CF for the next layer. I have a variety of diameters of sleeve, and I think I may be able to join everything up except for the very tip using one piece of the the 1.25" diameter sleeve. The tip will be cast in later, so I think that would work out.

Update: Looking at the biaxial sleeve diameters, I think I will want a 0.75" section and a 1.5" diameter section. These combine for 6.4 grams of dry CF.

 

[Adrian A]

In my warm garage the epoxy cured enough that I could get the peel ply off, so I decided to go ahead with the next layup. With a bright light shining into the mold, I could see o.k. what was going on:


It looks kinda ugly in the photo, but I actually think it's turning out well so far. There's just one substantial wrinkle on the right side, and it's just past the point where the leading edge of the chute cannon will go, so I don't think it will interfere.

The two braided sleeves (0.75" and 1.5") went down with only a couple of false starts. This is my third nosecone attempt in the last year, so I'm starting to get enough skill with my tools that it didn't seem totally impossible and desperate this time. I'm not sure I got the two CF sleeves and 2 layers of FG sleeves positioned with optimal overlap, but hopefully I got some overlap and the FG is positioned close enough to where it needs to be for the radios. The FG sleeves seemed to lay down against the mold o.k. The only thing I didn't get done on this layup was gluing in the cured CF tube at the aft end. When I tried to stick it in, it was just scraping up the edges of the rearmost FG sleeve, which shouldn't be surprising. When I have put in a premade tube into the aft end of a nosecone mold before, I had wet sleeve going back all the way to the mold opening, so I could hold onto it while the tube section slid in. This time I didn't want that extra thickness and weight of a fully-overlapped layup, so I'll just need to wait for this layup to cure and then glue in the tube as a third step. If the ID of the cured layup is too small I may need to take the partial noseone out of the mold and dremel out the ID before attempting to glue the premade tube section inside it.

 

[Adrian A]

Christmas morning here, as I unwrapped the nosecone mold and pulled out the nosecone:



I had good mold contact over most of it, but not at the tip where the inner bag couldn't reach. The cone feels appropriately strong, and I think that will get better after I re-mold it to cast the tip and fill in the low spots in the surface. It's 45 grams so far. The budget for the nosecone in the model is 85 grams, and that doesn't include any of the cylindrical section. I have another 65 grams budgeted for the tungsten epoxy nosecone tip and the stainless steel eyebolt I'll embed in it, which will be one of these:

https://lokiresearch.com/secure/storeDetail.asp?id=749618.
The inner bag, since it was originally molded around the outside of the mandrel, is too large to fit inside the layup without a wrinkle, and the wrinkle left a high spot on the inside. A flap sander on my drill took care of that pretty quickly in the first few inches of the nosecone. I'll want to go back with a threaded rod extension for the flap sander to remove more of it where it could interfere with the inner structures.

Speaking of which, now I can check whether the model is right for how much room I have inside for the chute cannon and nosecone ejector piston. The forward end of the 29mm chute cannon is supposed to fit 6.25" from the tip of the nosecone. The nosecone tip doesn't exist yet but with that red 29mm airframe tube pushed in as far as it goes, I lined up the ruler with where the tip is supposed to be. It's close to where it's supposed to be, but the tip looks like it might be a bit longer than that.



Also, checking the edge of the carbon, it's about 10.5 inches back from the tip, where I wanted it to end about 9.7" back. Bottom line is that the overall sustainer length will probably have to be about 3/4" longer than what's in the model right now. I'll know more after I make the av-bay, chute cannon and nosecone ejection piston. Looking ahead, I think the steps will be:

1. Make the nosecone ejection piston, since the hardware for that arrived yesterday
2. Lay up the 29mm carbon tube for the chute cannon
3. Add reinforcement and the baseplate to the chute cannon
4. Make the structural adapter that connects everything to the end of the motor
5. Put the av-bay together to cut the right length for the av-bay tube
6. Put all the sustainer innards together to see where the end of the nosecone ejection will wind up
7. Weigh, measure and simulate to figure out how much nosecone weight I want
8. Cast the nosecone tip with the eyebolt in the right location
9. Glue in the motor coupler tube to the back of the nosecone and taper the fiberglass overlap to meet it
10. Re-mold the nosecone's outer surface
11. Weigh and measure everything and put it back into the model
12. Tweak the fin design based on the actual CG and dimension

 

[Adrian A]

The last nosecone ejection piston that I made worked pretty well. It was 18mm carbon fiber tube that I laid up and reinforced. It sits on the front end of the chute cannon and pushes against the nosecone eyebolt to push the nosecone off of the front of the sustainer. When I first started making chute cannons and before I thought of the nosecone ejection piston, I just had a bare deployment charge here, with the expectation that the pressure generated in nosecone would blow it off. I found this to be not that reliable, because the friction of the nosecone varied a lot and there were opportunities for pressure leaks. It also took a relatively large amount of BP to get things moving. Using a smaller diameter piston allowed me to put the BP in the bottom without any other containment, and be confident that it would all ignite, even for a high altitude apogee. This time I'm taking it a step further and trading in the hand-laid CF piston for two nesting aluminum spacers from McMaster. The outer one is 2" long 1/2" OD and is threaded for a 10-32 screw. I drilled out the middle to 1/4". The inner piston is a 2" long 1/4" OD spacer from McMaster threaded for a 6-32 screw. I drilled that one out also, to join up the two threaded sections and make it easier to stick the wire in there.


I had to chuck the inner piston into my drill to sand it down for a loose running fit in the outer piston. Then I cross-drilled it for the ematch wire to come out of the side so that it doesn't get jammed against the nosecone eyebolt when it fires. The red plastic ematch cover keeps it from shorting. Even after trimming it, it's a little longer than I had envisioned, so there is about 1/2" of wasted length:



I'll probably cut down the length of the inner piston to free up some more length. I could also drill out a little more of the outer piston so that the inner piston can sit down deeper. There is currently abut 1/4" of threaded section for the screw that seals bottom of the outer piston and connects it to the bulkhead that will sit on the front end of the chute cannon. There is some room under the red plastic protector around the ematch for some BP, but from past experience I'll want about 0.25 grams in there, which could push the inner piston out a bit further.

 

[Adrian A]

Construction of the chute cannon:

I wrapped a 29mm motor casing with one layer of mylar, with some thin kapton tape on the inside. It took a couple of tries, but eventually I got it on there clean and snug, or so I thought. I used plain weave carbon, 13 grams IIRC. I thought it was 17, but the cured tube is only 22, and I don't think the resin ratio is that low. The larger piece makes two wraps, and the shorter piece adds 2 more wraps around one end. This is so that the base of the cannon where the charge goes off can be thicker than the forward end. I may add additional reinforcement at the base end. The carbon is inside a clear garbage bag. I pour out the epoxy over the carbon, pull the bag over the top, and then use the bag to squeegee the resin around and get a consistent resin amount.



I wrapped both pieces around the tube, and it seemed like everything was sticking down nicely. Then I spiral-wrapped the whole thing with 2" wide strips of peel ply. Then covered it with paper towels to promote gas escape, cut off the epoxied end of the garbage bag and then put the tube into the remaining bottom of the bag. I attached the vacuum port, twisted and clamped the open end of the bag, and the pump settled out at 20 in Hg.



This morning I peeled off the bag and the peel ply, and discovered a lot of wrinkles.



I think the tube is usable, but unfortunately the wrinkles do show up on the inside, where they could make a gas leak path around the piston. If I knew a way to get a totally smooth ID I'd re-do the tube today, but I'm not sure what I would do differently to improve the result. Anyone have a suggestion?

On the plus side, the tube seems strong and it's only 22 grams at this extra-long length before I cut it down.

 

[Adrian A]

Here's a photo of the inside of the tube:



Maybe if I slather on some filler-thickened epoxy before the first wrap, that will fill in the print through and any wrinkle divots. I think I will try again, and this time be extra sure that the mylar is wrapped snugly around the mandrel.

The wall thickness is about 14 mils on the 2-wrap end and 28 mils on the 4-wrap end. The OD on the forward end is 1.16", vs. 1.24" for a 29mm fiberglass tube I have. This is the motivation for laying up my own tube; I need to fit a kevlar line around the front end of the chute cannon and small differences in OD make a big difference in length where this slender nosecone narrows down. The 2-wrap end feels like it's probably strong enough, but without much extra margin. I think I'll re-do this with 3 wraps all over and then another 1-2 wraps at the bottom.

 

[boatgeek]

For such a light part, I'd probably just skip the vacuum bagging entirely. Even if you have a 25% increase in the total amount of resin, you're only adding a few grams. That said, I use a completely different work flow so my approach may not work for you. I typically paint unfilled epoxy on the tube, then lay dry cloth on to it. I add epoxy over the top with a paintbrush, and roll vigorously with a grooved roller to force out air bubbles in the matrix. I back off on adding more epoxy on teh last layer or two. You might be able to do something similar but with rolling the pre-wetted carbon on to the mandrel.

 

[Adrian A]

In my next attempt I spread some colloidal-thickened epoxy on the first circumference of the carbon and then it rolled the pre-wetted CF onto the mandrel. It seemed like I was able to get it down nice and snug. Then I spiral wrapped it with the peel ply and everything still seemed smooth. I put it in the vacuum bag and turned on the suction and that's when the wrinkles appeared. I'm pretty sure they're just in the carbon and not in the Mylar. I could smooth some of them by hand with the vacuum on, but then decided to try just curing it from here with the vacuum off. Hopefully the short period of vacuum got rid of voids in the layup and if I'm really lucky, it will come out round and smooth on the inside. If this doesn't work, I think I will try spiral wrapping the mandrel with the cloth at a small angle so that I can get better tension on it.

 

[kalsow]

For me, vacuums and tubes don't usually work.

I second @boatgeek 's suggestion to skip the vacuum.

A different path is to cut the carbon on the bias. With 0-90 thread directions the ones running around the tube have no where to go under the vacuum. Their only choice is to wrinkle. With 45-45 thread directions every strand has a path to escape under the vacuum. The bi-axial weave tubing relies on this technique.

 

[rocketace]

Maybe if I slather on some filler-thickened epoxy before the first wrap, that will fill in the print through and any wrinkle divots

That is exactly what I would suggest. When I started doing this based on JimJarvis's write up my tubes have been coming out perfect on the inside.

Also, as boatgeek just stated, I would skip the bagging and just do a peel ply. I have personally never had good luck bagging anything below 54mm without getting wrinkles. Cant get epoxy percent as low, but its always been low enough for me.

 

[watheyak]

I'll pile on and warn against vacuum bagging tubes.

 

[Adrian A]

My latest attempt came closer, but without the vacuum bag I had some voids against the mandrel.




You can also see some voids in the edge of the layup. I had a clean inner surface on a 29mm CF tube I made last fall with biaxial sleeve and shrink tubing. I may use that one or make another one for the inner surface and then reinforce with 0/90 plies on the outside.

 

[Adrian A]

I re-did the tube this morning and I think I finally broke the code. I slathered on some thickened epoxy over the mandrel, then installed a wetted lightweight biaxial sleeve, then did 2 layers of wetted 0/90 cloth, then at the base of the cannon where the pressure and stress will be highest I spiral wrapped wetted 1.25" wide 0/90 cloth with about 50% overlap which made it about -15/75 angles. I spiral wrapped peel ply with 2 layers to compress the layup, covered with paper towels and then put it in the oven at 160F for a couple of hours. It finally came out just the way I wanted, pretty much perfect. Here it is before any trimming:



The edges of the peel ply make it a little ugly, but it's super strong and has a great ring to it like low-void composites should. When I trim it down to the necessary length it will be about 20 grams.

 

[Adrian A]

I re-did the tube this morning and I think I finally broke the code. I slathered on some thickened epoxy over the mandrel, then installed a wetted lightweight biaxial sleeve, then did 2 layers of 0/90 cloth, then at the base of the cannon where the pressure and stress will be highest I spiral wrapped 1.25" wide 0/90 cloth with about 50% overlap which made it about -15/75 angles. I spiral wrapped peel ply with 2 layers to compress the layup, covered with paper towels and then put it in the oven at 160F for a couple of hours. It finally came out just the way I wanted, pretty much perfect.

View attachment 590319
View attachment 590320

 

[Adrian A]

I'm finally getting over my cold from the last 2 weeks and I made some progress today on the front end of the sustainer.

First up, some planning for how to trim the cute cannon. On the left is the nosecone so far. In the middle is the hardware that will get embedded in the nosecone tip along with tungsten/epoxy mixture, and underneath is an update to the nosecone ejection piston. This time I bought a 3" x 3/8" unthreaded spacer from McMaster, and drilled it out for most of the length, and tapped for 1/4-20 on the other end. This way I'll get more pressurized travel and more room for the nosecone shockcord up in front of the chute cannon. The chute cannon tube is on the right, before trimming to length.


Despite thinking about this for weeks, and re-measuring my chute with the actual shock cord I want to use to make sure I had enough room in the chute compartment, I somehow forgot that I also needed to budget for the piston and the deployment charge at the bottom. D'oh! But I think I left enough margin, and I have some other options for more room if I need them, so I don't need to re-do the chute cannon tube yet again.

After trimming the chute cannon tube, I prepped some pieces for the bulkhead that will withstand the deployment charge. I have a small coupler ring to provide extra glue surface area. You can see the chute cannon tube is pretty thick-walled at this end where the pressure will be highest. It's considerably thinner-walled at the other end.



After gluing with JB weld:



Now moving on to the nosecone. I went after it hard with the belt sander to remove areas that were not well consolidated and bonded with the inner layers. Unfortunately, the bad areas were a lot more extensive than I had expected. I uncovered a nasty dent that you can see below, and I even pretty much sanded through in a couple of places. Not structurally sound.


Not confident in my ability to do better if I tried laying up from the inside of the mold again, I decided to experiment with laying up an external reinforcing layer with braided CF sleeve. The disadvantage is that it may be larger than appropriate for the mold in places, but I am doing so much rework anyway that I hope I can make up for it with extra surface layer at the end.

I didn't take any pictures during the layup process, but I'll show some pictures I took as I unwrapped it, in reverse order as if I were building:

I knew I wanted to compress the layup from the outside, but I didn't have a way to withstand that pressure and press back from the inside. I ended up improvising, starting with with a nosecone and 29mm cardboard tube I had used as a layup tool earlier. I wrapped it in paper towels to increase the diameter and give it some springiness to hold the shape from the inside, and then wrapped it all up in plastic to release the inevitable epoxy that was going to make it through the holes in the layup:



After a little trial and error I got the kind of internal pressure against the weak areas of the cone that I was looking for.

Next I mixed up my epoxy and then de-gassed it using a bell jar. Lots of bubbles were removed. Then I slathered up the nosecone and then the sleeve with plenty of epoxy and squished the sleeve down into place. I used thin strips of peel ply to provide the compression. Because of the conical shape, a long strip that starts out with a large wrap angle on the wider-diameter end will end up stopping and turning back the other direction after it gets near the tip. Below is all one piece that's about 4' long.



I had to cut more strips to get enough material to fully cover the reinforcement area, but eventually I got it wrapped snugly with one or two layers of peel ply. Here's how it came out of the oven:



On days like this I set my oven for "warm cookware" at 140-160F and put each epoxied piece in as I go along. In this case I also had mixed up a bunch of extra epoxy, so I put that into the refrigerator while I put the nosecone in the oven. After an hour or two I pulled it out, and it looked like it came out great. You can see it a little bit about 3 photos up where I had unpeeled all but one layer of peel ply

With a little more time in the day and some refrigerated epoxy available, I decided to go for the nose tip hardware installation. I sanded down the new reinforcement on the belt sander, opened up the nosecone tip a little, prepped the shockcord attachment hardware with grease and tape, and taped it to the end of a hollow rod I could use to position it inside the mold. Then slathered up the nosecone, put some extra epoxy inside the cone where the hardware will bond, and then mixed up some of the epoxy with tungsten powder for the tip, which I slathered on the hardware and then dripped into the tip of the mold (and a little on the sides) after adding more mold release. I jammed in the nosecone and poked around with my handle for the nosecone tip hardware until I got satisfied that the noseone is in as far as it can go and the tip is centered. Fingers crossed that I guesstimated the right amount of epoxy and tungsten to drip in there to solidly attach the hardware but not cover the hole for the shock cord. If worse comes to worst, I guess I can always cut it off and try again. I'm not expecting a great surface to come out of the mold tomorrow, but success will be if it is a continuous sealed nosecone with a strong tip and a place for the shock cord to go. I still need to add the back end of the nosecone that will go over the motor case. After I get that bonded on I'll go nuts with the belt sander again so it fits properly in the mold, and then attempt a near-perfect molded surface by squishing the nosecone back into the silicone mold with a bunch of extra epoxy one last time.

 

[Adrian A]

It came out about as well as I had hoped. The tip hardware seems solidly embedded. I was hoping it would be a little better centered (you can see the front of the threaded rod, peeking out below. Also, after sanding off the high spots I'm not liking how thin it is in front of the reinforcement I made. I think today I will extend the reinforcement further forward with some 0.5" dia and 0.75" dia braided carbon sleeves with the same method I used yesterday.



I got the hollow rod off of the shock cord attachment hardware, but I can't see down there well enough to see what's going on. I ordered a borescope from Amazon that should arrive tomorrow. I'm kicking myself a little for not attempting to install the shock cord attachment with some cord already threaded through the hole. That would have added some risk of the hole getting hopelessly clogged if any epoxy had gotten in there, but it's going to be a bit of work to remove the masking and thread the hole.

 

[Adrian A]

Some progress today. I used the bore scope to help clear out the masking tape. But when I tried to put a weighted thread through the hole in the hardware, it wasn’t going in and I was afraid it was clogged with epoxy. I ended up drilling a couple of small holes to help get the hardware threaded, but I got it done.

This afternoon I filled the holes and filled a divot from some material that didn’t lay down against the mold correctly.

Then I cured that in the oven and went on to attach the aft airframe tube and take a shot a re-molding the surface.

Here’s how I left it, with extra epoxy dripping out of the bottom of the mold, and plenty of weight on the mold to get the nose one to seat as deeply as possible into the mold. Fingers crossed.



In the last couple of days I have been working on the av-bay too. Not a lot to show for it yet, but I have some small Al spacers coming from McMaster to space the two 29mm Blue Ravens apart. It looks like I’ll be able to get 2 Blue Ravens, a GPS tracker, a 150 mAhr battery for one Blue Raven and a 400 mAh AAA size battery for the other Blue Raven and the tracker into the 29mm diameter av-bay with 3” long threaded rods.

 

[Adrian A]

The nosecone came out with the surface in good shape. The epoxy is continuous, with no bubbles or divots, all the way to where the aft extension is glued in. De-gassing the epoxy really makes a difference, I think. The next 2 photos show the new longer, lighter nosecone compared with the existing nosecone. The new one weighs 90 grams, vs. 265 grams for the older one, because of all the tungsten I cast into the older one's tip. The new one's tip has a little bit of tungsten, along with the steel harness attachment hardware, so it balances slightly nose-down, compared to the very nose-down attitude of the previous nosecone.





The RF transparent window is moved forward quite a bit with the new nosecone, consistent with the shorter overall length I'm shooting for. Some of the length reduction comes from using a 29mm chute cannon rather than 24mm, and some comes from just having a lot less of the tip volume taken with noseweight. On the new nosecone there's a black sharpie line drawn where the 3" rods of the av-bay will go. The GPS tracker will be toward the front of the av-bay so I think it's o.k. that the carbon section I installed last night went in a little farther forward than I had intended. There's still a couple of inches of RF transparent area.

The aft end of the nosecone still needs work. The carbon fiber aft section that is so obviously smaller came from the airframe tubing I plan to use for the sustainer. It's only 1.59" OD, vs 1.61" OD that the the nosecone mold was designed for. Typical tubing is 1.60" OD. I will probably fill and taper it down on the nosecone.

 

[Adrian A]

I seem to be missing a post or two. Oh well.

The innards of the StratoSpear sustainer are almost all done, so I took some photos, in reverse order of assembly:



Here's the av-bay and chute cannon bolted onto the end of the Loki 38-1200 motor case. The shock cord is missing but the base of the chute cannon is cross-drilled for it. The nosecone fits over everything on the left, to the depth shown. There's quite a bit of overlap with the motor case. I may want to shorten the nosecone somewhat so that the nosecone ejection doesn't have to do quite so much work. I'll ground test first with it the way it is now though, to maximize the distance to the first joint, delaying the turbulent flow transition as long as possible.



Above is a closeup of the base of the nosecone ejector. I still need to drill it for a kevlar tether. I think I'll attach it to the nosecone deployment to reduce the risk that it could tangle the chute.


The base of the av-bay is shown. Normally the bolt that attaches it to the motor is the first thing to be assembled and the last to be disassembled, but for this test fit it was loose enough that I could undo it by hand. After my single-stage attempt this weekend (weather-permitting) I'll add screw switches to the inside of band on the left for arming the sustainer head-end ignition.



Now I have unbolted the chute cannon from the av-bay using the long t-handled hex driver. Normally this is the first step in disassembly. The main chute and apogee deployment charges are connected to 3 threaded rods that are exposed in this view. I drilled the chute cannon above and below the base you can see here so that the main charge wires get routed out of that cavity to the outside, and then back into the cannon from the side just above the bulkhead. This is so that any gas escaping through the hole will just get dissipated outside and won't go blasting into the av-bay. The harness for the main chute and the nosecone is one piece of Kevlar that gets routed through two more holes in the bottom of this cavity, which are both (barely) visible in this photo. I usually put stopper knots in both sides so that when the nosecone is deployed it doesn't pull all the main chute's harness through, and vice versa. The circuit board you see on the end is just a 29mm Blue Raven attachment bulkhead. It doesn't have any electronics on it but it's there for structure. I have a 10-32 T-nut glued into the middle that is used to screw the chute cannon to the av-bay.


After removing the red av-bay outer sleeve, you can see some of the av-bay innards. At the bottom are 2 29mm Blue Raven altimeters. I'm waiting for some spacers to arrive to separate the boards. The GPS tracker fits in there too, with its power coming from the Blue Raven that is not doing the sustainer ignition. Each Blue Raven has a built-in magnetic switch. In the back you can see a AAA-sized 400 mAhr Lipo battery, to which I have soldered a wire at each end to power the tracker/Blue Raven. The other Blue Raven gets a normal Raven battery that plugs directly into the JST 1.25 mm connector on the board.



Here's pretty much everything taken apart. The av-bay baseplate with the threaded rods is the first thing to get assembled, and then it is bolted onto the motor. Then the av-bay gets built up, the red av-bay sleeve goes over the top, then the av-bay top plate is fastened to the threaded rods. Then the main and apogee charges are connected to the threaded rods and the chute cannon is bolted to the top. Then the piston and chute are loaded, then the nosecone ejector is put in place, and the nosecone ejection harness is reefed and taped to the side of the nosecone ejector. Then the nosecone goes on and it's ready for the pad.

 

[Adrian A]

With a lot of weights and dimensions nailed down, I updated and finalized the design in OpenRocket and RasAero, and tweaked the fin design. The rocket is coming out a little lighter than when I submitted my class 3 application, so the altitudes are even a bit higher. OpenRocket design attached.

The fins are quick to make so far. I printed and cut out templates and applied spray adhesive, then stuck them down on a mostly-unidirectional CF I had purchased a year or two ago:


You can see the unidirectional grain, which is going left to right. The rocket is designed for 3 fins but I cut 4 to have an extra for just-in-case.

Then I cut them out on my we tile saw.


A rough-cut fin compared to my previous sustainer fins. The sweep angle is about the same, but these keep going. I needed more fin area on this one because the nosecone weight is so much lower. But the tradeoff in extra drag is worth the weight reduction in this application.

 

[Adrian A]

Today made a fin alignment jig from two sets of aluminum angle from McMaster. I just put it together by drilling and tapping 1/4-20 holes and using knobs for fasteners.


To glue the fins on as parallel to the body tube as I could, I used some calipers to adjust the tube guide pieces to be exactly the same distance from the fin at all 4 ends.



Since the nosecone is so extended, the aft body tube is shorter, conveniently short enough to fit inside my oven, which is speeding up this process. I'm using JB Weld for the initial bonds and then tomorrow I'll use Cotronics 4461 thickened with colloidal silica for the fillets.