How to build a fiberglass rocket, part 13: parachutes

Rocket recovery

The recovery system in a high power rocket is extremely important. Needless to say, launching a large rocket extremely high into the air is pretty fun. But that thing is going to come down again, sooner or later, and the recovery system you’ve designed and built for your rocket will determine whether it comes down like a ballistic missile or floats down gently for a soft landing. It could potentially injure or even kill someone. More importantly, your valuable rocket could be completely destroyed.

If the rocket separates in the air, even with no parachute, that’s a good start. A rocket separated into two (connected) parts loses its aerodynamic design; it will fall, but awkwardly and more slowly than if it were in one piece. Of course, if it’s heavy, it’s still going to hit the ground pretty hard.

Drogue vs. main

Better to separate and have a parachute. A single chute can be sufficient – it depends on the rocket’s expected altitude and how heavy it is. Ideally, though, the rocket will be capable of “dual deploy,” which just means deploying two separate parachutes, a drogue and a main.

person running in street, pulling orange parachute behind in the air
a running start

The drogue chute is smaller and deploys at apogee. The idea here is that, at apogee, the rocket’s speed has ground to a halt. It’s no longer shooting upwards, but it hasn’t yet started falling back down very fast either. The rocket separates at this point and the smaller drogue chute deploys, slowing the rocket’s descent to some degree.

After the rocket continues its descent and is closer to the ground, the airframe separates again and the main parachute deploys. This larger chute slows the rocket’s descent considerably and allows for a softer, gentler landing.

This order is important. You wouldn’t want to release the main (larger) parachute at apogee because even a slight wind would carry it very, very far away by the time it landed. But if you only used a drogue chute during the flight, the rocket’s descent would not slow sufficiently and it’d have a rough landing.

person in street, holding open orange parachute above head
I am actually being pulled backwards here

A weighty problem

This brings us to my current predicament. The Darkstar Extreme is a fiberglass rocket, and it’s pretty heavy, at roughly 14 lbs before adding the recovery system or the motor. That’s about what two gallons of milk weigh. Imagine the force it would take to accelerate two gallons of milk vertically, thousands of feet into the air – and likewise, the size parachute you’d need to significantly slow the descent of something that heavy.

There are online calculators that can help determine the right parachute size (diameter) based on the weight of your rocket and how fast (or slow) you want it to land. There are a lot of trade-offs in rocketry, some of which may not be obvious. All else being equal, for example, the softer the landing, the better – but there is a trade-off here. You can slow the descent and get as soft a landing as you’d like by increasing the size of the parachute. But large parachutes take up a lot of space. It’s increasingly difficult to stuff a gigantic parachute into your rocket without increasing the size of the rocket (and its weight, which then requires an even bigger parachute, etc.).

I ended up buying a 2 ft diameter drogue chute and an 8 ft diameter main chute, both from Rocketman Parachutes. In the above pictures, I opened them up as soon as they arrived at my house and ran around in the street trying to catch the wind and fly them like kites – with mild success.

They’re also a very vibrant orange color, which should make the rocket a little easier to see during its descent in the sky, and easier to locate once it touches down.

I’m looking forward to launching the Darkstar Extreme in the near future and seeing these parachutes deploy! I should also be able to capture a lot of data with my flight computer, so I’ll be able to tell how fast the rocket was descending with the drogue, as well as with the main chute.

High Power Rocketry: L1 Certification Flight

Finally! According to the National Association of Rocketry, I now officially have my level 1 certification in high power rocketry.

rocket launching into the air with fire and smoke below

I finished building my first high power rocket, the HyperLOC 835, back in December, but getting certified requires a successful flight and recovery of the rocket. But clubs don’t often host launch events in the winter months, and those that do are still subject to weather conditions (e.g., snowstorms). It’s helpful to have a club host a launch because you need (a) access to a large suitable area of land, (b) a waiver from the FAA to launch up to a certain altitude, and (c) launch equipment, such as launch pads and rails and an electric ignition system.

Clubs often start hosting launch events in the spring, but in spring 2020, COVID-19 hit, and things were cancelled or postponed.

I was finally able to attend a launch in June in south central Washington, about a 4 hour drive from where I live in the Seattle area.

I ended up launching the HyperLOC 835 on an Aerotech I-140 motor. The rocket is capable of dual deploy using a flight computer, but for this L1 certification flight I wanted it to be as simple as possible, so I didn’t use electronics. The recovery system was a parachute that deployed when the rocket separated using the motor ejection charge.

white and red rocket on launch pad
maiden flight

The weather looked ominous: it was cloudy, and we felt a few raindrops hitting us periodically, but it seemed to be holding steady.

The rocket launched, the parachute deployed, and it landed without a scratch in the tall grasses. The only tricky part was locating it. But since I was able to see the general area where it landed, it wasn’t too difficult to find.

yellow parachute in a sea of green tall grasses
a sea of tall grasses

Luckily the bright yellow parachute was pretty easy to spot from a distance, even though the rocket had sunk into a sea of tall grasses.

white rocket with text "improbable ventures" lying in tall grasses
a venture most improbable

Overall, it was a textbook launch and went as smoothly as could be expected! I’d estimate the rocket went about 1,700 ft in altitude, but as mentioned above, I didn’t use electronics for this flight so I can’t say for sure.

Immediately after this, I took the level 2 written exam, which is required prior to the level 2 certification flight, and I passed that (not difficult considering I’d had six months to study). It started raining more heavily, though, and we weren’t sure if we would need to call it a day and head out. But we waited another 30 minutes for the rain to stop, and then the skies cleared up and the sun came out. Perfect timing for my L2 certification flight, which I’ll summarize in my next post!

How to build a fiberglass rocket, part 12: primer and paint

All the difficult and time-consuming rocket construction steps are basically complete. The drilling and sanding fiberglass is done, the epoxy has cured, and technically, you can fly it “naked” at this point. But a painted rocket just adds that extra touch of class, and we are nothing if not classy.

Before getting started here, a couple of tips and some basic prep. There are a few parts that you might want to cover before spraying anything: things like the aluminum tip on the nose cone, the aluminum motor retainer, and the rail buttons. You can use masking tape to manually cover up these things pretty easily.

rocket disassembled on cardboard and grass, no paint or primer
“naked” rocket

Also, keep in mind that where a coupler slides into another section of the airframe with a snug fit, you don’t want to build up multiple layers of primer and paint, or things won’t fit at all anymore, without sanding the paint off. You may want to cover coupler ends with masking tape as well to save yourself trouble later.

So, to begin: lay out the rocket on cardboard or somewhere that you don’t mind getting turned into a rainbow of colors. Spray a thin layer of primer over everything (I chose a simple white primer) and use a quick back and forth motion. Don’t spray too close, and keep it continually moving while spraying, so that paint doesn’t build up too much in any single area. You can always come back and spray again and again, lightly with a thin coat each time.

rocket disassembled on cardboard and grass, white primer applied
applying primer

Of course, since the parts are lying on the ground, you can’t get underneath and will need to wait for them to dry and then rotate them. Depending on how much you’re able to coat the pieces each time, you may need to rotate them just once, or perhaps twice.

Finally, once everything has a nice layer of primer and it’s dry, you can begin spraying paint. The colors and design are totally up to you, but I would certainly recommend a high gloss finish.

rocket disassembled on cardboard and grass, painted glossy navy blue
roses are red, rockets are blue

In my case, I went with a glossy navy blue for the rocket body. I then used “sunshine yellow,” also glossy, for the fins and the vent ring around the e-bay. Note that it didn’t matter if the layers of navy blue got all over the fins, but once this was finished, I had to use more masking tape to very carefully and thoroughly mark off the fins from the rest of the body. I also used some brown paper grocery bags, tearing them up into approximate sizes to cover the rocket between and around the fins, with masking tape at the edges sealing it off to create sharp and exact lines.

I’m no expert painter; this is only the second rocket I’ve ever painted. But I think once finished, it turned out pretty well.

completed rocket standing vertically, painted navy blue with yellow fins
the rocket stands on end taller than you

Of course, this paint job is bound to get scratched, scuffed, and generally deteriorate over time. The rocket gets disassembled and re-assembled, and parts bump and bang into each other, and of course upon landing after even a single flight it will get dirty and a bit dinged, no doubt. I can always touch up the paint in the future when that happens, and it doesn’t really matter – it’s purely cosmetic. But it is fun and it just completes the look.

How to build a fiberglass rocket, part 11: fin fillets

The epic fin-ale to the fin attachment series! (Pun intended.)

So far, we’ve attached the fins using the “through the wall” method with epoxy at several points: (1) the fin root, where it directly touches the motor mount tube, and (2) all along the inside edges of the fins and motor mount tube using a syringe to inject it. This created an incredibly strong foundation (especially by mixing some chopped carbon fiber into the epoxy), and at last we can turn to (3) the outside of the rocket to create fin fillets.

fiberglass rocket and fins on workbench, with masking tape creating lines
making tape to create lines

The first step here is to measure and mark approx. 3/8 inch from the joint, along both the airframe and the fin itself, and draw parallel lines on each. Then follow this up with masking tape along the full length of each line. Using this technique, you can apply the epoxy fillet, and when you remove the tape afterwards, it will leave a very clean edge. Generally you want to wait before removing the tape so the epoxy has a chance to partially cure – but it’s easiest to remove the tape if you do so before it fully cures.

rear view of rocket with epoxy fin fillets
aerodynamic and aesthetically pleasing

For the epoxy mixture, the difference this time will be the addition of a thickening agent so it’s not quite as runny and it maintains its shape. Specifically, this means first mixing the two part epoxy (resin and hardener), then mixing in some chopped carbon fiber as before, and finally adding the thickener.

As I mentioned earlier, I’m using West System resin and hardener, and also for the thickener. The stuff is extremely lightweight – so much that it’s almost difficult to even take the can’s lid on or off, as the slightest breeze or air movement will cause it to fly up into the air like dust. Of course, this is something that should be done only while wearing a respirator or face mask.

Once the epoxy is sufficiently thickened, it should have the consistency of peanut butter – spreadable, but will more or less hold its shape.

Finally, a round length of wood (e.g. a broomstick) or plastic (e.g. metal pipe or PVC pipe) will be very helpful at this stage, particularly one that has a 3/4″ or 1″ diameter. I happened to have some spare 3/4″ PVC pipe lying around from my earlier project running electrical wire to the shed to build the workshop. The idea here is to spread some epoxy on at the joint where each fin touches the rocket body, and then to use the PVC pipe to run along the full length, creating a nice, smooth, rounded fin fillet.

a first pass

Above is a picture showing my partial progress with this technique. It can take some practice getting the right epoxy consistency with the thickener, and also using the pipe to create the rounded fillet. But once finished, this will provide the third and final bond of the fins to the rocket body.

As noted above, you can wait before removing the masking tape, but don’t wait so long that the epoxy fully cures.

After one set of fins is complete, as shown above, you can rotate the rocket 120 degrees and repeat (although give the epoxy some time to cure, first, before rotating). Repeat a second time, and then rotate and repeat a third.

Once all the fin fillets are completed, you’re done with the actual rocket build! The rest is cosmetic work or just attaching things to this newly constructed rocket: priming and painting (which, frankly, is optional), attaching recovery harnesses and parachutes (slightly less optional but simple), and conducting ground testing.

How to build a fiberglass rocket, part 10: epoxy injection for fins

Fair warning here: I’m actually splitting the “attaching the fins” information into three separate posts. This is not out of a sense of malice or sadism, but simply because there’s a lot going on with the fins.

The prior post (part 1 of 3 in this epic fin series) was basically the prep work and first few steps to secure the fins to the rocket body by using some epoxy and inserting the fin root through the wall where it can bond against the motor mount tube inside. Along with a printed fin alignment guide and a bunch of heavy objects, this keeps the fins in place and serves as a starting point for the multi-step approach for attaching them.

This post (part 2 of 3) involves epoxy, a syringe, and a bunch of strategically drilled holes. This might sound like the setup to a bad joke, but it’s actually literal and straightforward.

If you’ve been following along so far (whether with your actual fiberglass rocket build, or just conceptually in your head), you know that inside the airframe there is a motor mount tube, held in place and centered with the aptly-named centering rings. The top and bottom of each fin should just barely be touching a centering ring, inside.

fiberglass rocket with fins attached, sitting on workbench, partly suspended in air
ready to inject!

The idea here is to take the syringe and inject the epoxy into each hole. There’s two holes per fin (so 12 total in this case for 6 fins), with one hole on either side of each fin, roughly 1/2 inch away from the fin. Using a typical plastic syringe that you can find at a local drugstore, you can inject 10 ml at a time, and you should inject roughly 25 ml for each side of the rocket, split evenly among 4 holes. The rocket should be positioned horizontally and completely level, as in the above picture.

Once the epoxy has been injected into these 4 holes, you tilt the entire rocket forwards and then backwards, slowly, in order to move the epoxy around inside and completely coat the area where each fin touches the motor mount. The centering rings on each side should create a “dam” to prevent the epoxy from going any further past the edge of the fin, if everything is aligned reasonably well.

(If not, well, the epoxy may ooze out some of the other holes below a little. Not a huge deal, but may require a bit of extra cleanup.)

Once the epoxy is spread evenly inside, it needs some time to cure. Come back a few hours later. At this point, you can rotate the rocket 1/3 of the way around and repeat the process for the next 4 holes, and then finally a third time after that. Ensure each time that the rocket is level as the epoxy cures, so it doesn’t slowly ooze and collect in a lopsided fashion. This would not be ideal for a uniform fin attachment, it could also throw the rocket off balance in its weight.

epoxy in plastic mixing cup with popsicle stick on workbench
two part epoxy mix

Above is a picture of the two part epoxy mix (resin and hardener) when combined and thoroughly stirred in a small plastic cup. The syringe I used is pictured as well. You may need several since they can get clogged over time.

One final note here: you can also mix some chopped carbon fiber (pictured below) with the epoxy, and again mix thoroughly. The color will darken noticeably. This epoxy mixed with chopped carbon fiber will significantly strengthen the bond as it cures. In other words, those fins are never coming off.

epoxy darkened with chopped carbon fiber, in plastic mixing cup with popsicle stick on workbench
epoxy mixed with chopped carbon fiber

This epoxy injection technique is pretty cool, and it’s been tested and used successfully for many years. Try it!

In my next post (part 3 of 3 in this series on attaching fiberglass fins) I’ll briefly cover the final step: creating external fin fillets where each fin touches the rocket airframe. It gives one final layer of protection to ensure the fins are secured, and also looks more aerodynamic.

The mystery of the sewing machine: How someone nearly stole my identity


A few weeks ago, I received a surprise: a brand new sewing machine showed up at my front door.

It was a surprise because I didn’t order it. I’ve never owned a sewing machine, and I have no overwhelming desire to own one. I’m not even sure what one does, exactly, with such a machine.

singer stylist 7258 sewing machine, box packaging
surprise package

And yet, there was a UPS label on the front of this box, and it was very clearly addressed to me; it had my name and my address. Everything was spelled correctly and appeared to be in order. I was unambiguously the intended recipient.

So I assumed, logically enough, that this must either one of two things: (1) a gift from some friend or family member, currently anonymous but soon to surely reveal themselves as the generous benefactor, or (2) a mistaken shipment.

Let me also quickly point out that this was also no ordinary sewing machine. It came in packaging directly from the manufacturer, where the return label said SVP Machines. A quick Google search revealed this to be the parent organization of Singer Sewing Company. This was a Singer Stylist 7258, and a quick glance at the Singer website revealed this machine’s retail price to be about $300. The Rolls Royce of sewing equipment!

singer stylist 7258 sewing machine
sewing machine with hieroglyphics

Given that these seemed to be the only two possibilities, I left it in the box and waited for someone to contact me. Surely either a friend who was trying to surprise me would coyly ask me if I’ve received a package recently, or (less likely but still possible) the manufacturer would contact me and say they somehow accidentally sent me their product, and would likely ask me to ship it back (hopefully at their own cost).

But days went by, and then a week, and then two. I heard nothing from anyone. This sewing machine was a real mystery. Where did it come from? Was it truly a mistaken shipment from SVP Machines (i.e., Singer) who didn’t even realize anything was wrong?


After about two weeks, I happened to be checking my recent credit card transactions online. I do this periodically, usually once a month when I receive my statement and I grow outraged and indignant that my bill could be so high. I’ll glance through the transactions looking for something amiss, only to sheepishly realize that one after another is justifiable and I simply spent more than I realized.

But this time, something was different. I saw a charge from May 20 from “SVP Machines” for $826.

That can’t be right.

I’ve never heard of SVP Machines before receiving the sewing machine package a few weeks before this, so I immediately connected the two. This was clearly a charge from Singer Sewing Co. for the machine I received.

I stared at this charge and tried to mentally re-calibrate the likelihood of my previous two possibilities. This was definitely not a friend or family member offering me a generous gift (though the chances of that had already faded more with every day that had passed). My credit card was being charged. That’s not typically how a gift works. That only left the manufacturer making a grievous error, somehow incorrectly charging me and shipping me their product.

But the dollar amount didn’t add up, either. I checked the Singer website one more time to be sure I was looking at the right machine. The Singer Stylist 7258 was still going for $300. So even if this was somehow a mistake, where was the $826 charge coming from? Was I missing something?


At this point, I decided I should investigate a bit further. I looked more closely at the UPS label on the package and yes, my name and address were correct. But among all the bar codes, tracking numbers, and other information on the label, I noticed in the upper corner that it said “2 of 3.” Did this mean there were two other packages in the shipment? That might explain the $826 charge for what appeared to be a $300 product.

I checked the UPS website and entered the tracking number from the label. Sure enough, this was one of 3 packages in the shipment. In the tracking details, it showed that there was a request from the shipper the day after the order was placed (while the packages were still in transit) to re-route them to another city in Washington state that is approximately a 4 hour drive from the Seattle area, in the central part of the state. The other two packages were successfully delivered the day after mine to that address. UPS confirmed they were left at the front door.

This certainly seemed like credit card fraud, although it is still baffling that someone is committing this crime and going to these extreme lengths solely to purchase sewing machines. And even then, why did one of the machines get delivered to me at all?

But I noticed two other things that chilled my blood and removed any doubt that this was fraud, and part of a bigger scheme. My credit card transactions indicated a brand new charge as of June 8: the United States Postal Service (USPS) was charging me $1 for “change of address.” And my credit card account stated that I ordered a replacement card, and the card was on the way.

I never attempted to change my address with USPS, and I never ordered a replacement card.


At this point, I was certain it was fraud and some sort of identity theft scheme, and I started taking action to correct everything with all the parties involved. I spent the better part of a day making phone calls, alternating between navigating automated menus and speaking to real people.

I disputed these unauthorized charges with the credit card company and explained I never ordered a replacement card (someone else must have done so), and so they cancelled the card and issued me a new one. I made sure USPS was aware that I am not changing my address and someone else fraudulently attempted to change it on my behalf, so that my mail would get routed to them.

I spoke to folks at both UPS and Singer, trying to get a name or the exact street address for the location that the other two packages were sent to, but both refused to tell me. All I know is the city. They did say that they can tell law enforcement, and so I also filed a report with the local police, who should get this information and track down the perpetrator.

While I have a new replacement card, and USPS is aware that my address didn’t change, there are still some deep unanswered questions here.

  • Who was responsible for this unwarranted attack on my person?
  • Once they obtained my credit card information and decided to make unauthorized purchases, why did they start (and end) with sewing machines?
  • Why did one of these machines arrive at my door, when the other two were rerouted to the perpetrator?
  • What are the relative advantages and disadvantages of chainstitch, lockstitch, overlock, and coverstitch?

These are questions that may never be answered, and we may need to leave them to the philosophers. But this concludes the saga of the sewing machine and thwarted attempt at identity theft.

Let me know in the comments if you have any thoughts about this strange occurrence, or if you’ve ever experienced something similar.

How to build a fiberglass rocket, part 9: attaching fins

Exciting times! I’m ready to attach the fins.

If you’ve never put together a rocket before, well, I’m baffled that you are reading this blog. But typically, a rocket will have either 3 or 4 fins, which are placed symmetrically – equally spaced out, in the 360 degrees around the center. If 3 fins, then they’d be 120 degrees apart; if 4 fins, then 90 degree spaces.

This rocket has 3, and then another 3 aligned above them for a total of 6, each spaced out by 120 degrees. I think having 6 in this arrangement is purely aesthetic, as opposed to just having 3. Who knows?

I’ve used this two part epoxy before (resin and hardener) in one or two previous steps with this rocket construction. But this is the first time I’m using it in larger quantities.

rocket with fins attached, lying on workbench and partly suspended in air
back that up

Basically, I used a fin alignment guide (which I found online for free and printed out) to ensure that the fins were aligned properly, spaced exactly 120 degrees apart all the way around. I then prepared some of the epoxy and applied it with a popsicle stick (sophisticated technology), applying it to the edges or “roots” of each fin as if I were buttering a piece of toast. I inserted each fin into its slot, where the fin edge or root with the epoxy is pressed up against the motor mount inside. This is the first of several steps to ensure the fins are securely attached, starting with the interior.

To hold everything in place while this initial round of epoxy cures, I had a couple options. I saw some fancier solutions that other people have done involving using jigsaws and drills to cut out holes in large plywood sheets, and lots of vises and clamps.

That seemed like a lot of work, so I just used some rubber bands and propped the rocket/ fins up against some heavy objects like cans of paint or bricks while it cured (making sure nothing could move, and that the fins were aligned perfectly according to the alignment guide).

For a closer look at the epoxy, I included this photo as well, since it’s critical to this and the next few steps. I used West System epoxy as it was highly recommended, and it works great. The 105 is the epoxy resin, and 205 is the hardener. Each comes with a pump, and you just combine one pump of each into a mixing cup, and mix thoroughly for several minutes. It begins to cure pretty quickly, and the chemical reaction causes it to get extremely hot as it cures (to the point where it will burn you if you touch it, even through the plastic mixing cup, and steam is visibly coming off the top).

mixing epoxy on workbench
mixing two-part epoxy

For most of the fin attachment points, I’m also mixing in some chopped carbon fiber (pictured here as well) which is, in certain places, injected inside with a syringe. The carbon fiber greatly strengthens the epoxy as it cures.

Next, I’ll continue using the epoxy to attach the fins via this injection method, along the inside. After that, a final application of carbon fiber-infused epoxy on the outside of the rocket to create fillets (i.e., just a narrow strip of epoxy along each area where the fin touches the outer rocket body, shaped into a curve to minimize drag).

Things are really coming along – with the fins finally attached, it’s starting to look like a rocket!

How to build a fiberglass rocket, part 8: rail buttons

Before I jump into the riveting details of rail buttons, I’ll take a step back and explain what this is, and why it matters.

Every rocket has a “center of pressure” and a “center of gravity” (or center of mass). I won’t go into detail about these concepts here, but basically, the relationship between these two things is important for a rocket to remain stable in the air. When it’s moving at a fast speed, the fins help keep it going in a straight line (i.e., up) because of the way the air pushes on them. I’m oversimplifying these concepts, but this is the point:

When the rocket is sitting on the launch pad and first lifts off, it is not moving quickly enough to be stable. If you tried launching a rocket from a pad without any kind of support, there would be a pretty good chance that it would not ascend perfectly vertically. It’s entirely possible it would not ascend at all, as it might tip over and fly horizontally (perhaps into a crowd of spectators). This is not ideal for your rocket, or for the spectators.

The solution to this is to provide just enough support for the rocket to keep going vertically as soon as it launches and just begins to (quickly) gain speed. With small model rockets, a thin metal pole is all you need, just a couple of feet high. The rocket will have a small launch lug (basically like a plastic straw) attached to its side, which slides down over the metal pole, ensuring the rocket takes off using the pole as a guide.

For larger rockets, it’s the same concept but with slightly fancier hardware. Instead of a thin pole, the launch pad will have a much bigger rail standing vertically for support. And instead of a plastic straw glued to the rocket, it will have rail buttons, made from plastic and secured by drilling a hole in the rocket body and attaching with metal screws.

metal screws and black plastic rail buttons on workbench
the parts

The concept is extremely simple, and installation is fairly simple as well. It just requires measuring where you want the two rail buttons to be located, marking the spots, and drilling to insert the hardware. In general, you want the rail buttons exactly halfway between two fins, with one very close to the bottom (aft) end of the rocket, and another some distance up the side.

closeup view of rail button attached to red rocket body
the finish

Often one or both is drilled and screwed directly into a centering ring. Whether that’s possible or not on your particular rocket, it also helps to add a small amount of epoxy just to make sure it’s secured in place. Here, you can see where I attached the rail buttons on this fiberglass rocket.

further back view of red rocket body with two rail buttons secured
the alignment

And that’s it! Rail buttons installed, and the rocket can be flown from a standard launch rail.

The next step will require slightly more work: attaching the fins, which are of critical importance in achieving that fashionable “rocket” appearance.