A2MD Rocket

Quick Links
EX Products
Home
EX Forum
Propellants
Rocket Motors
Rockets
Rocket Electronics & Recovery
Ground Support
Launch & Static Tests
Software
Links
Motor Class Table
Iowa Amateur Rocketry Group
Don't Click Here!
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
This new rocket project came about after a couple of weeks of serious consideration between two different rocket styles. The debate was between a minimum diameter aluminum body rocket, and a larger diameter fiberglass body rocket. With the success of my last two rockets, the Cosmo 2 and the Prelude, it would be hard to move to a different construction style. But I had to look carefully at my goals...

What it really boils down to for me, is that I design a rocket, the motor, and the propellant to meet a specific flight plan. In this case my goal is an altitude of 15,000'. The speed of the rocket is a neat extra, anyone that tells you they wouldn't like to break the speed of sound is either lying, or kidding themselves. Another goal of a flight is the ability to launch a payload of some sort. Not that I have any specific payload in mind at the moment, but at some point that becomes an issue. Payload issues are actually getting easier as my rockets get larger, I actually need more mass for maximum altitude.

A lot of people like the show and flash of a launch, a big motor in a big heavy rocket is a real crowd pleaser, but I'm not into what's called "Big Dumb Rockets". An "L" motor in a 120 pound rocket is about the same as a "C" motor in a little Estes kit, with a lot more show of course! Flames, color, smoke, none of those really concern me. I'd be just as happy with a flight that had no visible exhaust.

When I started in EX, I was trying to build rockets as cheaply as possible, with only local materials. I was pretty pleased if I got a rocket to 1,500' and back safely. Now my goals have changed, I'm looking for reliable flights and recovery at ever increasing altitudes. So where does it end? My guess would be with a "P" class flight. Maybe not, perhaps I'll get into staging, or strap on drop boosters.

I looked at what the next project after this one would be, and likely it will be the SBS-6250 "O" motor. The SBS-6250 motor is 6" in diameter, and it's unlikely I'd need more diameter for payload in that rocket. So I decided this new rocket, the A2MD will be a stepping stone for construction of future aluminum body rockets.

Admittedly, I do have concerns over an all metal body rocket. The concern would be if the rocket ever came in ballistic, an all metal body rocket could do some serious damage. But as I thought about it, the more I realized the difference in a fiberglass rocket with a large aluminum motor and an all aluminum rocket is negligible. You wouldn't want either one landing on your car. To minimum this risk, regardless of the rocket body style, I'm going to use a redundant apogee deployment system, RADS for short. This system will use a sensor to detect deployment of the apogee parachute, and fire a back up drogue if it does not detect the primary drogue deployment. I'll get into the design later, but I thought I should bring up the concept early on in the development of the rocket.

Here are some the basics of the new rocket:

Projected Altitude: 15,000'+

Propellant: KNO3/Erythritol

Rocket Diameter: 3.5" with a transition to 3.75" at an all aluminum fin can.

Propellant Mass: 14+ pounds

Total Impulse: 1,750 pound seconds; 7,778 Ns

Recovery: Dual deployment via a PIRM 2, RADS for back up apogee deployment.

Here's an early draft of the A2MD using AreoLab.

Lower Body Tube/Motor Casing: 6061-T6 Aluminum 3.5' OD x 60" long x .125" (inside diameter 3.25")

Fin Can: 6061-T6 Aluminum 4" outside diameter turned down to 3.75" x 9.5" long x .125" (inside diameter 3.5")

Fins: .125" thick 6061-T6 Aluminum, welded to fin can.

Coupler: 6061-T6 Aluminum 3.5" OD turned down to 3.25" x 7.0" long x .125" (inside diameter 3.0")

Nose Cone: Hollowed clear pine with outer fiberglass layer.


To retain all the various parts, nozzle, bulkhead, couplers, fin can etc. I'll be using an undercut flat head 1/4" stainless steel machine screw.

Here's a drawing from McMaster-Carr showing the machine screw.

The advantage to this type of screw is that it won't protrude out from the body tube. I also ordered an 82 degree countersink from McMaster-Carr to match the angle of the screw head. This should make for easy assembly of the rocket, and yet leave a fairly clean profile.

The construction of this rocket will require turning down the fin can and coupler to the proper diameters. To make this a little easier, I ordered a bull nose live center for my lathe tail stock. The bull nose center is like a big cone on bearings, that allows parts with a large inside diameter (pipe or tube) to be turned in the lathe.


The above is a conceptual drawing, not entirely to scale of the Redundant Apogee Deployment System (RADS).

In a nutshell, the nose cone will house a complete back up drogue deployment system, including it's own drogue chute. I will have a sensor at the base of the nose cone that will deactivate the system should the nose cone deploy normally. I'll use a timer that is magnetically started at launch to fire a deployment charge located above the upper bulkhead.

Hopefully, the RADS will never be needed, as the primary drogue itself has redundant electronics. But should the RADS fire, it has the advantage that it will still be able to deploy the main chute via my PIRM2 dual deployment system.

For the nose cone I glued 5 layers of 3.5" x .75" clear pine together to form a square blank.

The blank was then turned on my wood lathe to form a 3.5" diameter cylinder.


Here I am using a Forstner bit to bore out the inside of the nose cone.


A plywood disk was screwed to the base of the nose cone so I can turn it again in the wood lathe. The area below the shoulder has been turned down to 3.25" to fit inside the upper body tube. To start turning the profile of the nose cone, I plotted out the dimensions of the nose cone, then printed it full size. I used the print out to map the dimensions at one inch intervals. On the blank you can see the 1" marks laid out.


Now it's just a matter of cutting the proper depth at each mark.


Once all the depth cuts were made, I removed all the material between the cuts, then sanded. I didn't even try to make the last 1" of the nose. Experience has taught me to quit while I'm ahead, rather than risk breaking the tip off in the lathe. The last cut I made was to a diameter of .62". The tip will be formed with Bondo body filler. I could also turn a metal tip, but the thermal heating shouldn't be great enough to be a concern with this rocket.


I decided against the RADS system in the nose cone. The Forstner bits I ordered weren't long enough to be able to bore the length I needed. I could have still done it, but it would mean using pins of some sort to retain the top on the nose cone, and I didn't have enough diameter to work with to get the pins in. The very narrow profile of the rocket and nose cone made it a tight fit regardless, I'll give the RADS concept a try in a future larger diameter rocket.

Not using the RADS in the nose will also free up the nose cone space for the RDF transmitter. The nice thing about having this long nose cone is now I'll be able to use a long antenna with the transmitter, and it will be inside a wood nose cone that is pretty much RF transparent. That was another concern going into this project, with an all aluminum body, the transmitter would need to be placed outside the body, I suppose I would have placed it in the drogue chute, but that wouldn't have been ideal anyway.


For the nozzle, I started with a round of 1020 steel 3.5" in diameter. Here the convergent cone has been turned to 60 degrees, two o-ring grooves have also been cut, and now I'm reducing the outside diameter in the throat area.

Aside from some more polishing, the nozzle is done.

In the box to the left is my new bull nose live center for the lathe. The left aluminum cylinder will be turned down to 3.75" OD, and will form the can part of the all aluminum fin can. The aluminum cylinder to the right will be cut to 7" and turned down to an outside diameter of 3.25", this will form the coupler between the upper and lower body tube sections.

Here is the coupler tube being turned down in the lathe.

Here I drilled out 12 holes through the (motor casing/lower body tube) into the coupler. The holes are then tapped and counter sunk and my undercut flathead machine screws are inserted. When the motor is assembled, the coupler will be removed and the upper bulkhead will be inserted from the top, the coupler will be reinstalled and will serve not only as the coupler, but also as the motor upper bulkhead retainer. While the 1/8" thick 6061 aluminum may seem a little light to use as a retainer, I calculated the loading on the coupler and it has about a 9 times safety factor at my maximum expected operating pressure.

Here is an aluminum fin perched next to the as yet unfinished fin can. I had real concerns about cutting the fin plan form from the sheet aluminum. In the end, I decided to try my power miter saw. Now, it was just a carbide tipped wood blade, but I figured if I ruined it, it wasn't that big of a deal. It actually worked quite well and the blade is still fairly sharp! For the leading and trailing edges, I simply went to my big disk sander with a course grit.


I was curious if it would be possible to attach the fins to the fin can with epoxy, probably the best epoxy for the job would be J.B. Weld. So I decided to do a little test. I cut a one inch strip from my 1/8" aluminum fin material. The span of each fin is 4", so I left the 1" wide strip at 4", I roughed up both the strip and the tube with some 240 grit sandpaper, then mixed up the epoxy and set the fin stock strip into an epoxy bead on a scrap piece of 3.5" aluminum tubing. The strip was taped to the tube and allowed to cure for 24 hours.

Here is the strip epoxied to the tube.

I wanted to measure the force required to break the joint. I set up my load cell and recorded the data as I pressed the far tip of the aluminum strip against the load cell. I applied pressure only from the tube, so all the force was directed at the joint from 4" out.

Here's the load cell recording.

Here is the broken joint.

The joint failed at about 27 pounds. Interestingly, it failed in the epoxy itself, rather than in the adhesion to either surface. I have to admit, it wasn't a good, filleted joint. And the joint was a lot thinner than I would have used in a real application.

My conclusion. Well, I think it would work. Considering each fin is 9" long at the root, that would give me 243 pounds of force required to break the joint. A larger, better formed joint would likely double the strength as well. For the flight of A2MD, I used software to calculate the dynamic loading on the rocket at 25.7 pounds per square inch. Providing the rocket flew straight and true, the loading on the fins would be minimal. But should there be an anomaly during flight, and the rocket had an angle of attack of some 30 degrees, things might get a little dicey.

I've got the fin can tube turned down to 3.75", or .125" thick, and tapered and sanded smooth the transition area to 3.5". But what I am having problems with is welding the fins to the tube. I knew this would be a problem with a small stick welder, but I have welded aluminum in the past with success. Unfortunately, I'm having no luck this time around. If I had a TIG welder... I could have it welded up at a machine shop, but I really hate to do that. It may sound silly, but it seems like cheating if I didn't do it myself.

So it's plan "B" for the fin can. I ordered some carbon fiber cloth, the plan now is to do a composite fin can. Using only surface mounting (as in, not through the wall) I think I'll need the strength of the carbon fiber.

The rest of the rocket nearly ready to go. The electronics module is mostly done, with a new PIRM2 just finished.

Here's the new PIRM2. No great changes to the design, but I did use a slightly thinner key at .125". I also increased the length of the charge holder inside the device. That leaves more room for a longer e-match head. The charge holder cover uses (2) #10-32 x .5" machine screws. Retention of the device is through (4) #8-32 x 2" machine screws. The pin is a 1/4" bolt with the threads cut off. The pin is retained by stranded steel picture hanging wire that goes through a hole drilled in the bolt head and to a stud hole wire retainer on the other end. Both ends were soldered for security, I don't need a bolt falling down from 15,000'!

I ordered some chemicals for anodizing aluminum. Not that I really think it's needed, but I thought it may be fun to do, and does add some extra resistance to corrosion. I'll start a new page on anodizing once I get started.

Look at that rocket go! Ok, so Cosmo actually has a squirrel up a power pole. He'll stay there all day if need be, or until the squirrel is dumb enough to try to jump down and run for it.

Sorry, back to the rocket...

I've been working on the composite fin can, and decided to start a page with details on just the fin can. Check it out on the Carbon/Glass Composite Rocket Fin Can page.


Not that the rocket is anywhere near done, but I thought it'd be nice to get an idea of what the rocket would look like when done. So I assembled the rocket for the first time.

Pictures really don't do it justice. When I assembled the rocket on my work bench, my first thought was on the lines of, "Oh my God, what have I done!". Compared to the Prelude rocket on the left, the A2MD really looks lean and mean.

Here's a shot from the other end. Everything about this rocket has the look of speed. And rightly so I guess, compared to the Prelude it's smaller in diameter, about the same weight and holds 50% more propellant. If this rocket flies straight and true... Let's just say, it should be an awesome flight!

Here the lower body tube/motor is getting a snap ring groove cut. I was going to retain the nozzle with screws, but the screw holes need to be perfectly aligned to keep stress on the screws equal and to make sure the nozzle is perfectly in line with the length of the motor. This tube section is 60" long, and I was concerned about getting the holes drilled properly, so I opted for a snap ring to retain the nozzle instead. As you can see in the picture I'm reaching the length limit on my lathe to cut snap ring grooves.

Here's a closer shot of the cutting tool. A little hard to see, but it has a flat hook ground into the tool to cut the groove. I made the groove a little deeper into the casing to accommodate four screws that will retain the fin can.

With all the parts of the rocket pretty much complete, I've been casting grains so I can measure the center of gravity with a full propellant load. The grains are 70% KNO3/ 30% erythritol as opposed to the usual 65/35. The increase in oxidizer should increase the Isp by about 5 seconds. I'm going to test 6 of the new grains in the SBS-1350 in the next couple of days. If all goes well, I'll use the new high oxidizer propellant in the A2MD.


Here are some preliminary numbers:

Rocket Weight: 21.55 pounds fully assembled no propellant or thermal liner

Rocket Length: 110.5" tip of nose to end of motor casing

*Center of Gravity: 72.88" from nose tip

*Center of Pressure: 93.61" from nose tip

*Using a propellant load of 18.4 pounds including casting tubes, no thermal liner.


Here's the top of the new electronics module for the A2MD. One thing it still needs is an apogee charge holder extension tube. I've got one on the way from McMaster-Carr, it's a 1/4" threaded aluminum pipe nipple. When it arrives I'll drill and tap threads in the bulkhead to retain it. I was very pleased with the upper bulkhead fit in the body tube, it slides in easily, but the tolerance is so close you can put a hand over the body tube end and the air pressure stops the electronics module from moving.

This is inside the top bulkhead. A little hard to see because the aluminum is so reflective. The electronics mounting board is 1/4" plywood with 2 layers of glass on each side. I thought about using aluminum plate instead, but I was concerned about shorting out the electronics on the aluminum, I also like the wood as it tends to dampen some vibrations.

I didn't want to leave anything to chance, so I went with a bulkhead turned from solid 6061 aluminum bar. The bulkhead thickness is .25" with a .2" thick shoulder that extends down .5". I'll drill through the outside aluminum upper body tube and tap threads into the shoulder of the bulkhead to retain it. The main anchor is a 1/4" stainless steel U-bolt. I ordered some 5/8" tubular Kevlar from Pratt Hobbies for the shock cord, it has a breaking strength of about 10,000 pounds. While that's some real overkill, what I really wanted the Kevlar for was its resistance to cutting and its heat resistance. Now I won't have to worry about the shock cord burning through from the deployment charge, and it eases my concerns about the shock cord getting cut on the upper body tube when it deploys.

I ordered a Transolve P6 altimeter kit to back up my Accel Techronics ARB-3D recording flight computer.

Upper body tube to left, lower body tube to right. I used a buffing wheel and compound to lightly polish the upper body tube. This was after only about 15 minutes on the wheel. I was curious how well these tubes would polish up, it looks like with a little time and elbow grease, these aluminum tubes can have a mirror finish.

Here again I couldn't resist getting a look at the rocket assembled. The nose and fin can only have a first coat of color on, and the body tubes need a little more time on the polishing wheel, but it's looking pretty good.

My PK6 altimeter kit came in today, it was a breeze to put together. In less than 30 minutes the unit was soldered together and I did a quick vacuum test on it. It worked great! The documentation didn't say the amp rating of the outputs, so I looked up the output transistor specs on the web and found it to be 5 amps. The new e-matches I've been making fire at .5 to .6 amps, so I'm well within the units capacity.

Here I have the electronics module laid out as it would be mounted in the upper body tube. The lower body tube has 18 screws to retain it, as the motor bulkhead will be retained by the coupler. The upper body tube is attached to the coupler with 8 of the same screws. The upper bulkhead in the electronics module will have deployment load stress on it, so it is retained by 6 of the stainless steel screws.

As you can see, I'm not leaving much to chance with this rocket, as it should be incredibly strong. The electronics module has (3) 9 volt battery holders on the back side. The main power switch is a double pole single throw 20 amp switch. Two of the 9 volt batteries are tied in parallel to power the computers, the third battery is deployment charge only for the ARB3D. Access to the power switch is through one of the 1/4" ambient pressure ports, allowing easy arming once the rocket is on the pad and ready to launch.

My next project was a test of the deployment charge in a partial vacuum. I understand black powder can fail to fully burn properly at low ambient pressure. I read a report on deployment charges by Scott Aleckson, his report was on using confinement of gun powder (not black powder), while this isn't the same problem as black powder at altitude, it is similar. His conclusion was confinement of the charge would allow the charge burn fully, a latex or rubber cover of some sort was recommended. I needed to use my charge holder extension with the PIRM2 design, so I opted to cover the opening of the holder with both tape, and a couple of fingers cut from a latex glove.

Here's the charge holder. 2" from the end is an epoxy plug to hold the charge, 2 holes were drilled through the epoxy so e-match wires could be run out the bottom of the holder tube. The open end with the wires coming out was sealed with about 1/3" of RTV and allowed to cure overnight. 1.8 grams of black powder was used in the holder, a little paper wadding to keep the black powder in place, then a wrap of masking tape to hold the wadding. The the two latex glove fingers were slipped over the end and secured using electrical tape.

Here is the vacuum chamber used for the test. The chamber was pulled down to about 400 mb, about 24,000' with my vacuum pump. The lid was not retained by the clamps, and allowed to displace (like a nose cone) when the charge went off.

Here is the end result, the lid popped up when the charge went off but didn't go anywhere. No unburned black powder was found in the chamber.

A close up of the business end.

The wire end sealed with RTV held fine as well.

Conclusion: Black powder simply needs some limited confinement to burn at lower pressure. The latex glove fingers worked fine, as would a couple layers of electrical tape I'm sure.


My tubular Kevlar arrived today, so I set about rigging the deployment harness. I knew things were going to be a tight fit, so I had to be careful and pack everything as neatly as possible. When done, I just had enough room! The recovery harness includes; 18' of 5/8" tubular Kevlar going from the nose cone, attached to the drogue chute then to the PIRM2 release key. 22' of 9/16" tubular nylon going from the PIRM2 release key to the main anchor. With a 36" drogue chute, 80" main chute, a Kevlar drogue chute protector and a Kevlar main chute deployment pouch.

The dry weight of the rocket, ready to fly without propellant or thermal liner is 21.2 pounds. Using a 36" drogue the calculated descent rate under the drogue is 60 fps, under the main and drogue landing should be just under 25 fps.

I'm getting quite a pile of parts on this rocket, I wondered how many there were. So I did a parts count, each part that could be unbolted or unattached was counted. That's not unsoldering parts, an altimeter counts as one item as does each parachute, I didn't count each shroud line or panel. At the moment I have 170 parts. Remember when I started this page I said this would be my easiest rocket to build? So much for that idea. It has been a lot of fun, I got a chance to use new materials and techniques and I have learned a lot in the process. The fin can and nose have the last coats of paint on them, if nothing else the rocket looks really good. I my quest to make this a really solid rocket, I have gone over on the weight from what I originally predicted. The extra weight won't affect the maximum altitude much, but the peak velocity will be down to about mach 1.5.

Here is a capture of the first deployment test. Click Here (118 KB) for a short video of the test.

For the deployment test I used 2.4 grams of 2F black powder. The upper body tube was set in a flight ready configuration. If you look closely at the picture, you see can the red parachute and the green Nomex cloth parachute protector. Unfortunately, the charge really wasn't sufficient as the deployment was a bit anemic, the drogue chute also suffered a lot of heat damage.

A couple of things were vividly apparent; first: I needed to free up a little more room in the upper body tube, second: when space is at a premium a piston really helps. With that in mind, I replaced the Kevlar shock cord with 9/16" tubular nylon, and used a 3' length of Kevlar over the tubular nylon to protect the nylon that is exposed to the deployment charge. Next I made a piston like I used in the Cosmo 2 rocket, the piston is all composite, 5 layers of 6 ounce glass and 3 layers of carbon fiber for the base, and 4 layers of glass and 2 carbon for the walls of the piston.

Once all was assembled I again tested the apogee deployment with a 2.4 gram BP charge.

Click Here for the video (285 kb).

Much, much better this time. The parachute even opened in the mild breeze. You can just see the bottom of the drogue at the top, then the nose cone (which I covered in foam to protect) and then the piston cup.

Here is the upper, or drogue part of the deployment system. You can see the Kevlar protecting the tubular nylon shock cord below the piston. The Kevlar sleeve was epoxied to the base of the piston to keep it in place.


30" drogue chute at 23 pounds= 74 fps

30" drogue/60" main at 23 pounds= 33 fps

30" drogue/80" main at 23 pounds= 26 fps