29mm Min-Diameter High Performance Scratch Build Thread

The Rocketry Forum

Help Support The Rocketry Forum:

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

Adrian A

Well-Known Member
TRF Sponsor
TRF Supporter
Joined
Jan 21, 2009
Messages
3,157
Reaction score
2,806
Location
Lakewood, CO
I've always liked high performance 29mm minimum diameter rockets. They're cheap to build and fly, there are some kick-butt motors to use with them, and design-wise they are challenging to fit everything in efficiently. The last time I made a 29mm rocket, I tested different nosecone shapes and collected Cd vs Mach number data using a Raven altimeter. Later I it went to 14,818 feet on an H (https://www.tripoli.org/Records/Commercial-H-Records) once I made my own nosecone with a Von Karmon shape. That rocket got destroyed in a 2-stage attempt.

For this build, I just want a reliable, high-performance flight test vehicle for my altimeters and GPS trackers. I'm going to trade some performance for cost and ease of building, but I'm hoping it will still be useful for those interested in going fast on a budget. My goal is to get this done in time for a launch on Nov 17, and show how inexpensive a really high-performance rocket can be. Below is a screen shot with open rocket, some rough guesses, and a motor loaded for a 132G flight.

upload_2018-10-28_20-46-52.png

Design:

For this test vehicle I'm going to make it long enough to fit the longest 29mm motors I'm aware of, and to use the "chute cannon" concept I've used before. I like the chute cannon approach because it allows dual deployment with a single airframe break, it makes the best use of the nosecone volume, and it keeps the electronics well protected. Both of my official altitude records used this technique. I'll show more of that as I go along.

For electronics, of course I will be using a Featherweight GPS tracker and a new Raven4 that I'm testing. I haven't decided yet whether I will use a stand-along magnetic switch or a 29mm Featherweight av-bay bulkhead. More on that later.

My strategy is to use carbon in the back end, so the fins and body tube can be light and thin for stability and low drag, and use fiberglass in the front end, to provide useful density, while being RF transparent, easy to work with, and electrically non-conductive.

Materials:
For tonight, I'm ordering airframe materials. The main body tube should ideally be about 15 inches long in order to fit the longest 29mm motors on the market. I'm going to do an inexpensive and easy woven carbon sleeve over thin cardboard for the main tube. Unfortunately, Apogee only sells their thin 29mm cardboard tubes up to 13 inches long. I think I will tape 2 together before doing the carbon over-wrap, or I might change my mind later and just stick with 13", and let my largest motors overhang a couple inches.

Main tube:
A 6-pack of cardboard tubes is $9.60.
The carbon overwrap from Soller composites (1.25 in diameter, 3k weave) is $5.70 for 5 feet.
The 1.9" clear heat shrink tubing is $12 for 5 feet. Note that for $27 in materials not including epoxy, I can make 5 feet of carbon-reinforced tubing, which is only about $6 per rocket.

The epoxy is actually the expensive part. The epoxy at Soller composites seems promising, with arated temperature of 210F. I have used Aeropoxy with success, and its Tg is 193. Every little bit helps when you're planning for supersonic flights. A quart of 820 system sold at Soller is $50, and it will bump up the shipping, too. But I'm due for some fresh stuff. Also buying some microballoons to make a lightweight putty for fin fillets, $13 for a many-rocket supply.

Fins:
I already have some precut carbon fiber fin core stock from a previous project, so they're free for this build. But if I were to make them again, I would start with thin carbon sheet stock like this https://www.cstsales.com/a-carbon-f...M6-vSXawsEZI9BR_D2Qr2reM6MJwPhoBoCTDYQAvD_BwE, which is $12.65 for 5.75" x 5.75" , plenty for 4 fin cores on a rocket this size. Most of the fin strength comes from unidirectional tip-to-tip carbon which I will add. A 12" x 12" square of lighter uni is $1.49. Release fabric for compressing the tip-to-tip layup is $7.

Nosecone:
Since I have a nosecone mold, I'll use fiberglass sleeves to make the cone structure. Carbon fiber would also be o.k., except I want to maximize my GPS performance, so fiberglass it is. I'm buying biaxial braided sleeves in nominal diameters of 0.75, 1, and 1.5 inches. Smaller ones go toward the front, and so on. 1 foot minimum quantity for each one adds up to a whopping $2.52. Most people will just buy a cone, but so far nobody makes 29m Von karmon cones, so you would need to settle for conical ones.

Coupler/Chute Cannon
A strong fiberglass coupler is important for those > 100 G flights. One from Madcow looks promising. And it's only $6. The chute cannon itself will be a leftover 18mm Estes cardboard tube reinforced with more fiberglass. I should have plenty of biaxial sleeve left over from the nosecone. The av-bay will need some fiberglass sheet stock that I will reinforce with more fiberglass. Madcow sells 1/16" plate in 8.5" x 11" size for $11.

I'm going to ignore the shock cord and chutes for now, and add up the damage:

Apogee rockets: $9.60
Madcow Rocketry: $21.43 (not including a replacement 38mm cone I'm getting for another build)
Soller composites: $104.6, most of which is enough epoxy and release fabric for a lot of builds.
 

Attachments

  • upload_2018-10-28_20-46-17.png
    upload_2018-10-28_20-46-17.png
    60 KB · Views: 101
Exciting stuff, I've only pushed the unreinforced Aspire tubing to about 500mph. Looking forward to your results!

Isn't the Madcow Go Devil 29mm a VK shape?
 
I like it.. I modified an Aspire with double diamond fins and epoxy clay fillets and took it supersonic to 4700ft on a G80, and got it back again. Recovery was on 3m mylar streamer an it had an eggfinder mini and JLA3 in the cone...

For Thunder Downunder next year, I have plans for an H135 and an I200.. I plan to find what speed cardboard and balsa shreds... Best part is, in testing the G80, I had two flights where I lost the airframe, but the EFMini got me a nosecone with my electronics back...
 
Nice. I’m putting a chute cannon in my newest build. Can’t wait to see how yours turns out.
 
Nice. I’m putting a chute cannon in my newest build. Can’t wait to see how yours turns out.
Is there an easily located thread that shows how a shoot cannon works?
 
Is there an easily located thread that shows how a shoot cannon works?

I don’t know. You can look at my build thread. I have some sketches in there. It’s just a tube with a bulkhead on one end and an ejection charged at the bottom. Just pack the harness and parachute in on top of it.
 
I found a carbon-reinforced nosecone of mine that seems pretty stout but has some pretty severe surface bubbles in a few spots. I'm going to test the GPS in it tonight to see how much attenuation I get. If it works o.k, I'll just fill the voids and that would save me a lot of time and headaches (applying wet composites to the inside of a 1 foot by 29mm mold is not especially fun)
 
Will be following along! Looking forward to trying the new raven when it becomes available to the masses
 
Eager for Raven 4 testing and release. Know a college team that put five competition multistage flights on a Raven 3. Three flights were this year. Survived some crashes. And its been through a swamp. They’d like a Raven 4 for sale by next year fall for SEDS. You really built an easy to program altimeter that is durable. That Raven saved our lives. Thanks for the help this year with GPS tech support. Any rocket that helps you get the Raven 4 online quicker I’ll sub to. Best of luck with this min diameter build.
 
Eager for Raven 4 testing and release. Know a college team that put five competition multistage flights on a Raven 3. Three flights were this year. Survived some crashes. And its been through a swamp. They’d like a Raven 4 for sale by next year fall for SEDS. You really built an easy to program altimeter that is durable. That Raven saved our lives. Thanks for the help this year with GPS tech support. Any rocket that helps you get the Raven 4 online quicker I’ll sub to. Best of luck with this min diameter build.

Thanks, Andrew. Did you guys ever find the rocket from that last flight whose last packets showed it heading toward the swamp?
 
Yes sir. They used waders and a drone and got the Raven 3 and Featherweight GPS back from the swamp. More importantly the competition altimeter. We were all annoyed at how that last packet was treetop level for some odd reason. Other than that it tracked the flight okay. We just had a lot of bad luck with your GPS units this year. Once they sync’d it was generally good. Of all places it landed in a chest high swamp.
 
I tested the effect of a carbon nosecone on GPS and LoRa tonight. It was pretty interesting. The GPS was able to lock, but it was sketchy. Likewise, the LoRa signal was down about 15 dB (a factor of 32 in power). Here are some pictures and screenshots:

First is the ground station and a tracker inside the carbon rocket but with the nosecone off. This should be pretty representative of the configuration if the nosecone were fiberglass rather than carbon fiber. The GPS tracker is situated in the coupler above where the carbon of the lower part of the airframe.
IMG_3832.JPG

Below is a screenshot showing the GPS performance of the ground station on the left, and the tracker on the right. Pretty comparable performance. The colored bars are a summary of the GPS satellite information that the GPS receiver collects. The size of the blue section represents a strength over 40 dB, the green is between 32 and 40, the yellow is over 24, and the orange is under 24 dB.
IMG_3827.PNG

Then when I put on the nosecone, the GPS strength in the tracker dropped dramatically:
IMG_3834.PNG Now none of the tracker satellites are received with more than 32 dB. In this case the ground station is showing more satellites than it did before because it had been powered up longer at that point and had time to lock onto more satellites. I tried orienting the tracker in different configurations but got pretty much the same result each time. Once I took the nosecone off, then the tracker and the ground station came back together again. Based on the way blue went to yellow and green went to orange, it looks like a change of about 16 dB in the GPS signal strength (which corresponds to a factor of 40 in power). The effect of this is that if a GPS is allowed to get a lock with the nosecone off and then stayed powered, it would likely keep its lock on the pad. But a GPS needs at least 32dB signals to acquire a lock the first time, because it needs to download the satellite ephemeris (orbit location data), which is transmitted every 30 seconds and I have read that it only can do this with a strength over 32 (green or blue satellites). I just confirmed this by power cycling the unit, putting the nosecone on right away and then leaving it on. Even after a few minutes there is no lock. The effect in flight with this weak signal is TBD, but it's more likely to lose lock or give poor-accuracy solutions. For this rocket, once the nosecone comes off, it would get back to good strength again, but I'd like to see some good GPS data on the way up to apogee, so I'm leaning now more toward rebuilding the nosecone with fiberglass.

Likewise, I measured the LoRa signal strength with the nosecone on and off, and there was about a 15 dB difference there as well. That corresponds to a 7dB max range ratio, or a factor of about 5. So if the usual LoRa range is around 300,000 feet, then with the carbon nosecone on I would expect to lose signal at 60,000 feet, or earlier in an RF-noisy environment.
 
Adrian, that is a very good demonstration of why one does not put a radio in a carbon fiber nosecone or body tube. Attenuation, and lots of it.
 
IMG_3852.jpg A small update tonight, planning out the av-bay sled. After experimenting with a bunch of different configurations, I'm going to go for the H-shaped fiberglass sled shown above, with the Raven4 and the Featherweight GPS Tracker on top of each other and the batteries in the tray behind each one. In the background is the fiberglass 29mm coupler from Madcow. The round bulkhead in front fits inside it. As usual, everything looks bigger in the photo than it does in real life. The tracker, the Raven4, batteries, a magnetic switch and wiring should fit into about 3.8" of 29mm coupler. I'll include a banana for scale in later photos. The thin sheet material making the H-shaped sled so far is 15 mil FR4 fiberglass from Apogee. It's thin enough that it handles like cardstock and you can cut it with scissors. I'm going to lay up some fiberglass over everything in place of the blue tape in order to make a nice solid structure.

The front bulkhead will have a 0.1" pitch pin header embedded in it that the Raven terminal block will screw onto. Before I lay up the fiberglass I think I'll run some wires from it down to the back end in case I want to use the 3rd and 4th Raven channels later for a stage separation charge and sustainer airstart.

The chute cannon will be permanently attached to the rest of the sled with more fiberglass. For the tracker antenna I'll run a 1/4 wave wire through a hole in the bulkhead and alongside the chute cannon. Making terminals for the main and apogee terminals will be tricky. So far I'm thinking about using 3 2-56 screws arranged radially at the base of the chute cannon, with nuts and soldered wires embedded inside the base. This will be better shown with photos later than described now.
 
Just wondering about the chute cannon, after reading over the design thread you linked (direct link here: https://www.rocketryforum.com/threa...ign-and-build-thread.26383/page-4#post-238112) - where does the drogue chute go? Is it free-floating inside the nose, or is it nested inside the main chute tube? Is there just no drogue?

Build looks great so far. That is one heckofa snug fit on that AV bay. Can't wait to see it up in the air!
 
It's for a drogueless design. If you really want a drogue, you would need to leave some extra space for it in the tip of the nosecone and then you would probably wrap it in a nomex burrito. With these droguless designs with some decent ballast weight in the tip of the nose, I think the nose leads the way during the descent and the back end of the rocket flops around behind. My 75mm almost-all-motor rocket came down at 150 feet/second before the main came out and then 19 feet per second under the main.
 
Tonight I got set up for the tube reinforcement, but I didn't pull the trigger because I need to borrow or buy a hair dryer or a heat gun for the shrink wrap tubing.

Below is my layup table and the results of some resin experiments. Under the table I set up my hot box with a space heater that blows through the open box, and a couple of barbecue temperature probes. On the low setting, I get 130F in the airflow and about 105 on the bottom of the box. When I put a small amount of neat resin in there, it went off in about half an hour, melted the bottom of the cup and was smoking a bit. Too hot for this resin. The middle cup was mixed with some colloidal silica, which also melted the cup after being at about the same temperature, and on the right is the epoxy with low-density filler, which I didn't heat as much and it came out o.k. When I do the layup the motor mandrel will absorb a lot of the heat, and the resin will be more spread out, so I'm not too worried about the actual layup overheating. I'm planning on using a little of the colloidal silica to thicken the resin when I mold the nosecone, and I'll use the low-density filler for my fin fillets.IMG_3857.jpg
I also prepared my cardboard airframe tubes for reinforcement. I like the idea of making the tube a little long and then cutting the ends off, so I needed more than one tube and more than one motor case for a mandrel. I put some tape on the inside of one full-length tube and was happy to find that I could still fit the motor mandrel in there:
IMG_3853.jpg

I slid the other cardboard tube on and pressed the tape tight against the mandrel. Hopefully this will be enough to keep the epoxy sealed out of the joint when I do the reinforcement. Then I added a shorter motor case so that the joint is in the middle of the longer case cut off the excess tube:
IMG_3854.jpg
Then I masked off the ends and sanded off the glossy glaze from the tube. Below is the cardboard joint:
IMG_3856.jpg

I'll sand it a bit more right before the layup tomorrow.
 

Attachments

  • IMG_3856.jpg
    IMG_3856.jpg
    86.2 KB · Views: 70
I reinforced the airframe tube last night. Here is before shrinking the shrink wrap:

IMG_3859b.jpg

I had a borrowed heat gun that was a monster, and overheated the shrink tubing, which resulted in holes. So I cut it off and used a spare length of shrink tube. With a little more distance and care, I got a pretty good result:
IMG_3861.jpg

This morning, here's what it looks like to peel off the shrink tube:

IMG_3864.jpg
I was concerned that I would have trouble getting the motor casing mandrels out, but since I didn't have shrink tubing for the whole length, I was able to peel away the blue masking tape at each end and slide the motor out of the cardboard easily. There wasn't any epoxy leakage through the taped cardboard joint.

This carbon 29mm tube used about $12 in materials, and the carbon was actually the cheapest part.
 
Are you just making a carbon fibre tube, or are you carbon reinforcing the 29 mm motor tube?
 
I’m reinforcing the cardboard tube. The cardboard doesn’t add much strength compared to an all-carbon tube, but having it in there makes it super easy to remove the mandrel after the layup. The outer diameter of the reinforced tube came in at about 1.205”.

After trimming the tube last night, it’s 15” long and weighs 31 grams. That’s the same weight as my 4” long fiberglass coupler. When I finish my av-bay sled I might trim down the coupler a little for some weight savings to make it fly better on an F10.
 
Today I laid up the nosecone and the chute cannon tube.
I added a chimney to the hot box so that I could heat up the nosecone standing up (I wanted excess resin to go to the tip rather than one side) IMG_3865.jpg
IMG_3866.jpg
Here are the fiberglass woven tube sections when pushed together for near-maximum diameter:

IMG_3867.jpg
And here they are when stretched out:
IMG_3868.jpg

In a nosecone these are great because you can stick them into the mold in the skinny configuration and then push the back edge down to make it expand to fit the side of the mold. Here's how it came out:
IMG_3870.jpg

Structurally it's great, it feels very rigid despite only being 31 grams. My main mistake was to shove the cardboard airframe tube in a little too far as the last step after all the FG sections were in the mold, so it distorted the soft silicone mold, causing a ridge at the end of the cylindrical section. Also, there are the expected surface divots due to not vacuum bagging this. I sanded down most of the cyclindrical ridge, and I can take care of the divots by filling them with some thickened epoxy and then putting it back in the mold with some extra runny epoxy so that the whole front section will get a new epoxy coating. This will add some weight, but should fix all the surface imperfections. If this were a record attempt nosecone, I would be carefully adding tungsten powder/epoxy mix to the tip in order to get the rocket up to optimal mass, while moving the CG far enough forward for smaller fins. For this one, I'm more interested in high Gs high speed, and maybe some smaller motors, so I'm keeping it light and using some oversized fins for extra stability.

Below is the 18mm chute cannon, after getting 2 layers of FG. It's very solid, too. It's 11 grams now, but when I cut it to length it will be probably be around 6-7.

IMG_3871.jpg
 
One more picture...If I insert the chute cannon all the way, you can faintly see where it stops inside the nosecone. When this is done, it won't be in quite that far.

IMG_3874.jpg
 
Well, this didn't turn out quite the way I wanted it to...;):rolleyes:

IMG_3880.jpg

It's belt-sanding time.

Once I get it back into shape, I'll try again, but this time I'll start by squeezing the air and excess resin out of the silicone mold before I put the mold back into its rigidizing plastic tube. I have gotten this to work one other time years ago, and I think that's how I did it.
 
Well, I decided to spring for a benchtop belt sander. I have a nice Porter Cable portable belt sander, but it's currently in another state, and a job like this requires clamping a portable sander sideways on the bench, which is always a little sketchy. I read some reviews and got a Ryobi benchtop disk/belt sander from Home Depot, and I'm pretty impressed with it so far. It made quick work of all the extra epoxy from the last miscue, and it should be better at squaring up the ends of my tubes better than anything I've used before. So now I'm just waiting for my epoxy to warm up in the hot box and then I'll try again at using the mold for a finish coat. Since I didn't explain clearly before, my mold has two parts, a semi-rigid polycarbonate outer mold which keeps the inner mold straight, and a flexible silicone inner mold which I created by pouring it in between the outer mold and my aluminum mandrel. I put a slit in the outer mold to make it easier to get the inner mold in and out. See below:
IMG_3888.jpg

The picture also has the shows nosecone with the offending bits sanded off, ready to give another try to using the mold to give me an accurate epoxy outer surface.
 
3rd time is the charm:

IMG_3891.jpg

No bubbles, no airgaps, nothing to fill, glossy finish. There are a couple of high spots where some dried epoxy spots from the last attempt were stuck in the mold before I put it in, but those will be easy to knock off with some sandpaper. The mass is up to 38 grams now. Unfortunately, the inner mold gave up its life in an emergency C-section in order to get the nosecone out. Silicone molds have nice self-release properties for a handful of epoxy moldings, but to get them to last longer you need to add mold release, which I didn't have available this weekend. I still have the mandrel if I want to make another one later, but this rocket won't need the mold any longer. I guess the tip end of the mold is still o.k., so if I break off the tip with landing damage I can use that part again to re-cast it. This was the most challenging part of the build and it's a big relief for it to be done.

IMG_3892.jpg

Not sure yet if I'm going to do a tip-to-tip layup on the fins. The West 407 low density fairing filler plus the Soller composites 820 epoxy is a stronger and harder combination than other low-density fillets I have done in the past. I have used Aeropoxy and Cotronics epoxies before, and I wonder how this compares quantitatively, but it sure seems hard. The airframe with nosecone, 4" FG coupler, fins and fillets tips the scale at 123 grams.
 
I updated my sim with the new lighter masses, assuming I don't put tip-to-tip on the fins, and I can get the chute cannon and av-bay to come in at 40 grams or les:

upload_2018-11-12_8-23-16.png

The highest impulse motor will take it to just under Mach 2.0, and the highest thrust motor will get it going supersonic in 0.22 seconds, 100 feet above the pad. :eek:
 

Attachments

  • upload_2018-11-12_8-12-1.png
    upload_2018-11-12_8-12-1.png
    349.4 KB · Views: 77
Back
Top