The Defiance-H project will re-use portions of the original Defiance rocket, converting it from a solid propellant to hybrid propellant rocket. My first hybrid flight, the Aestus, was designed as a quick and inexpensive flight test platform for my first O class hybrid. The Defiance-H is intended to test some design changes necessitated by a minimum diameter design.
I want to keep the ball valve to initiate oxidizer flow into the engine. I'm not a huge fan of burning off a fill line and injector lines to initiate flow in a large engine. While it might be fine in small engines, I have seen cases where the nylon lines didn't burn through completely, resulting in reduced thrust and possible burn through issues from the N2O being injected off axis into the side of the fuel grain. I've also heard of cases of line fittings letting lose prematurely, or lines cracking from bend radius strain.
At the moment, I'm leaning towards running higher thrust on the engine by increasing either the size of the injectors, or increasing the number of injectors. I had concerns prior to the Aestus flight of weather cocking if we had much ground wind. As it turned out, we had almost launch scrubbing wind levels and the rocket weather cocked as much as I've had a rocket weather cock. Increasing the off rail speed of the rocket is the best bet to reduce weather cocking. I need to be careful though, as too much N2O mass flux could result in a flame "blow out" at ignition. So I'll have to increase the port diameter to keep N2O mass flux within acceptable levels (~.8 to ~.9 pounds per square inch per second).
Increasing the port diameter leads to another problem though, the HR5-B (short) engine almost ran out asphalt fuel during the Aestus flight. So I may have to revert back to the HR5 (long) engine and use a fuel grain of polyester resin/HDPE again. The down side to that is a slight reduction in Isp. Of course, the poly/HDPE fuel grain can sustain a much longer burn, so I can simply add a couple extra pounds of N2O to bring the total impulse back up.
As it stands now, I'm looking at a 7' N2O tank section, a 3' engine section, a 4' upper body tube section and a 2' nose cone. That's an additional 2' longer than the Defiance flew on solid KNER propellant, with about 81% of the total impulse of the KNER flight. I certainly could go longer still, and increase the N2O load and total impulse. But again with this flight, I'm not looking to set any altitude records and would like to keep apogee at about 20,000' with this flight.
Here's a rough drawing of the valve assembly which will be positioned in a coupler between the N2O tank and the engine.
Here are the valve parts after I machined the slip-coupler for the N2O connection between the engine and tank. Notice the taper on the edge of the upper slip-coupler. Since one end of the coupler is attached to the N2O tank, and the other end on the engine, I will need to slide these two parts together on a body tube coupler to assemble the rocket. The slight taper on the end should assist in assembly by guiding the two parts together. You can also see the o-ring just inside that taper. At the moment, I have only one o-ring as a seal. I'll pressure test this o-ring seal, and if needed add another internal o-ring.
Here they are assembled. Notice I did make a change from the drawing. I tapped the bottom of the slip-coupler with female threads rather than male threads. It was so much quicker and easier to tap the internal threads, rather than set up the lathe and cut a taper on the shaft end, then cut external pipe threads on the lathe. So I had to add one 1/2" nipple, but it saved me a lot of time and fuss.
I discovered a 3/4" drill bit was just perfect for the bore for 1/2" female NPT threads. I was also pleased to find the 1/2" tap still fit into my lathe tail stock drill chuck. It's very difficult to get perfectly straight threads cut by hand tapping, but chucking up the tap in the drill chuck and the part in the lathe chuck, keeps everything in alignment and cuts a perfect set of threads.
To test the slip coupler I plugged one end with a pipe cap, running an air line in the other end. Since the air pressure would turn the slip coupler into a small "hydraulic cylinder", I chucked it up in the lathe on the air inlet end, and used the tail stock to to hold the other end. I could turn the tail stock feed in or out to move the coupling under pressure.
Here's the air line coming out the head stock. After pressurizing with 125 psi, I did detect a very slight leak at the o-ring on the coupling. That may have been a sliver of metal, or perhaps not enough compression on the o-ring. I'll go ahead and cut a second o-ring inside the coupling, this time cutting the groove shallower to give more compression and a stronger seal.
It won't take a lot of new materials to make the conversion to hybrid propulsion, but I did need one new bulkhead, a coupler and a new engine casing section. To make the coupler I ordered a 12" length of 5" tube with a .25" wall thickness, then turned the tube down to an outside diameter of 6.75" to fit inside the body tube sections. In the picture above you can see the new coupler inserted into the N2O tank section of the upper body tube.
With the coupler inserted and the N2O bulkhead installed, you can see where I need to drill a hole through the body tube to gain access to the engine valve. The coupler will retain the bulkhead so I calculated the maximum force on the bulkhead at 22,863 pounds, the edge (or load bearing area) on the coupler is 1.82 square inches for a load of 12,562 psi. Since 6061 T6 aluminum has a UTS of 45,000 psi, I'm well within the load bearing limits.
This is the new engine bulkhead with the bottom half of the slip coupler installed.
I'm using an engine casing with a wall thickness of .125" on this rocket. I had used a .1875" thick wall casing on previous HR5 engines. The hybrid engine runs at a fairly low pressure, so I don't need the thick walled casing, I used the thicker casing in the past for more thermal mass. But going with a thinner casing will allow me to add nice heavy layer of EPDM rubber as an extra thermal layer, and reduce weight as well. That does mean I'll need a new nozzle though. I considered using the Defiance nozzle from it's last flight, but it did have a slight stress crack, so I'll likely make a new nozzle.
As my engines get bigger, I'm running into some other issues that will need to be addressed. A big issue is getting far enough from a test/launch with the launch control equipment to be safe. The way it is now, I'm running 5 sets of wires, if we start looking at a set back distance of 1,000' or more, that's a lot of wire! So I decided to start on a wireless launch control system. I'll probably do a web page on the system, so I won't go into much detail here. But the basics are a Basic Stamp 2 controlling a relay box at the pad, with a set of radio modems creating the link back to a PC application at the launch control table.
Here's a screen capture of the PC application.
This is a screen cap of my hybrid calculator showing the main parameters of the engine.
On the far left is the original HR5 injector bulkhead, the new injector bulkhead is to the right. The new injector is the same basic design, but it's slightly larger diameter to fit into the new combustion chamber case which is .125" thick, as opposed to the old casing which is .1875" thick. The new injector bulkhead also has larger, .145" injector holes and a larger inlet to accommodate the greater N2O flow rate of this engine.
In the foreground is the N2O tank lower bulkhead, it has the upper end of the slip coupler/valve assembly installed, the brass device is the one way fill valve with a 90 degree elbow to connect to the external fill line. In the back you can see the tank to combustion chamber coupler, behind it is the combustion chamber and the N2O tank.
Here is the lower end of the N2O tank, at this point I've drilled out another hole for the fill line attachment. Since this tube was used as a solid motor casing in the past, I had to make sure it was thoroughly clean inside. I used a cylinder hone with two extension arms to polish the inside, then I cleaned the inside with soap and water, then I wiped the inside down with IPA. At this point the other bulkhead is installed. The forward bulkhead had a tap for a pressure transducer, so I plugged that port, cleaned the bulkhead with IPA, lubed new Viton o-rings with Krytox grease and bolted the forward coupler/retainer in place. That coupler/retainer uses 40 3/16" stainless steel bolts. I also disassembled the one way valve and the ball valve to thoroughly clean them.
Here you can see the lower tank bulkhead installed. On the top is the fill line port, on the bottom is the ball valve linkage. The fill line needs one more 90 degree fitting not shown, I installed an old tube length to keep out any dust. To the left of the combustion chamber case is the graphite nozzle. I decided to reuse the original Defiance nozzle, since it's so close to exactly what I needed anyway. The throat is slightly larger than I had planned for, which will bring the throat area to injector area to about 14 to 1, where I had been running about 12.5 to 1. The greater ratio will drop the chamber pressure slightly, but should also help stabilize combustion.
So, where I'm at now. The N2O tank is sealed, that's nice to have done. Now I'll do an inert gas pressure test to make sure there are no leaks. If all goes well with the pressure test, I'll have to cast a fuel grain and assemble the engine so I can do a cold flow test using N2O. That will give me a chance to pressure test the tank, and give the new ground support equipment a test as well. I ordered some aluminum stock for a nozzle support ring, once that arrives the engine will be complete.
This is the original Defiance fin can. When I inspected it closely, I found a deep, pea sized ding in one fin. That fin also felt a little loose to me. My guess is the Defiance hit a steel fence post when it landed on its last flight. Better safe than sorry, I sanded down to bare composite, and added four layers in each fillet of glass and carbon cloth. Now that it's cured for a couple of days, I gave it the push, pull and tug test. It's back to it's original inflexible self. Now I just need to sand and finish again.
No pictures yet, but I've been working on the recovery system as well. I think I'll go back to a single drogue and main this time. So I won't need to use the deployment mortar tube like I did on the original flights. Sometimes I think keeping it simpler may be more reliable, to much redundancy can cause problems too. I'll use dual altimeters for recovery in the electronics bay, with a GPS telemetry system in the nose cone. The telemetry system will be either the new system that's being developed now, or if it's not ready I'll use the same system I used in my balloon flights and the Aestus flight.
Speaking of telemetry. I have the new radio modems for the remote launch controller installed. So I decided to run both the launch controller and the telemetry radios at the same time, just to make sure they didn't interfere with each other. With GPS telemetry coming in on one PC, I still had full control of the launch control system with the second PC. So, no problems there.
If all goes well, look for a cold flow test in about a week.
It was time to start thinking more about the recovery system. I decided to go with a single drogue to main configuration using a PIRM2 with both chutes out the upper body tube. This has worked very well for me in the past. But I did have a problem with this design in the Aestus, as the main chute came out at apogee. I have had this happen in my early testing of this system several years ago, and I simply forgot lessons learned at that time. The primary lesson learned was that the main chute needs to be well retained. In the Aestus, the main chute was pulled out from the apex of the chute, rather than by the shroud lines I usually use. When using the shroud lines to pull the main out of the deployment bag, the lines were pulled out the bag and retained snugly by the PIRM2, this had the effect of pulling the shroud lines over the top of the chute in the bag, holding it in place.
The Aestus didn't have those shroud lines to hold it in place, the deployment bag was also a little oversized, making it that much easier for the main to come out of the bag. For the Defiance-H, I decided to make a nice snug deployment bag with the added security of a Velcro closure.
In the above series of pictures you can see the bag open, Velcro'd closed and the opposite side in the last image. The bag was made from Nomex cloth, one side is a very heavy flame proof Nomex, the other side is material from an Air Force flight suit. I sewed in a loop at the bottom using 9/16" tubular nylon, this loop keeps the bag in the air frame while the main gets pulled out. I also added 2 elastic straps to hold the shock cord from the main chute to the anchor.
Here's the bag attached to the deployment bulkhead. The shock cord is 5/8" tubular Kevlar, with a breaking strength of 10,000 pounds.
Here I'm folding the shock cord from the PIRM2 to the drogue chute. When folding shock cord like this, you want to fold it back and forth over itself (not in loops), the folding allows the cord to pay out and won't tangle. To keep it nice and neat, I use a small rubber band on each end. The rubber bands snap off as the line pays out and are expendable.
Here I'm testing what happens when the PIRM2 releases. The main chute pulls out with maybe 4 or 5 pounds of force. Keep in mind, the force available to pull the main out the bag is going to be close to the dry weight of the rocket. The rocket while under the drogue chute will accelerate until it reaches terminal velocity, at which point the force on the drogue will equal the weight of the rocket (minus the drag on the rocket body).
Here it is all laid out as if it went through a deployment sequence.
Nozzle retention has always been tricky. You need to make sure the nozzle retainer is mounted perpendicular to the rocket body. If the retainer is off, you have excessive loading on one area of the nozzle and retainer, and also risk to a slight degree some off axis thrust. For the Defiance-H, I toyed with the idea of a retainer that slid into the casing, butted up against the end of the combustion chamber case, then had an external boat tail. The problem with that approach is that my casing isn't wide enough in diameter to do all that, and still have a large seating (load bearing) area for the graphite nozzle. I also have an existing nozzle, and I didn't really want to do extensive re-working of the nozzle.
I decided to compromise. I used a length of 5" OD x .25" wall tubing to make a retainer for the 1/2" steel retainer used on the original KNER motor. I turned the outside of the heavier tube down to .125" for about 1/2 of it's length. That allows the aluminum retainer to slide into the case and butt up against the larger diameter. It also allows the nozzle/retainer assembly to be perfectly square to the rest of the case.
Here's the aluminum part of the retainer. To make sure it was fully seated in the combustion chamber case, I held it in place with long pipe clamps while I drilled and tapped the threads. The bolts will go through the fin can to retain it, then through both the case and the aluminum retainer to retain the nozzle. Notice I did manage to get a little bit of a boat tail on the aluminum retainer.
From the end you can see the aluminum retainer in place, which is holding the steel retainer in place which in turn retains the nozzle.
For the most part, the rocket body and engine are now done. As you can see, I need to re-finish the fin can as well as the nose cone. Now that the combustion chamber is done, I can make a fuel grain and also do the cold flow test.
Here's a screen capture from AeroLab showing the dimensions and the center or pressure.
This is an updated screen capture from my hybrid calculator. Peak thrust should be about 850 lbf, burn time of the liquid phase about 8.2 seconds.
It's time to start thinking about the gps telemetry system. It doesn't look like we'll have the new telemetry system in place in time for the May 17 launch. So I needed to get my old balloon/rocket system mounted in the Defiance nose cone. I do need to use the nose cone for this, since the rocket has an aluminum body tube I need the RF transparency of the nose cone. Using a full sized hand held GPS along with the radio transmitter and a battery pack was going to put a bit of a squeeze on things size wise. The nose cone had been filled with foam, so I needed to come up with a mounting system that was both solid, and provided for some degree of shock absorption.
After a lot of thought, I decide to make an electronics payload tube from 3" fiberglass reinforced cardboard. Then use a sled to mount the electronics on. I used 1/8" plywood covered in 8 layers of 6 ounce glass cloth for the sled board, and 1/2" plywood bulkheads at either end.
Here's the nose cone after I removed enough foam to allow a 3" tube 12" long to be inserted. I tried melting the foam out with a hot steel rod, that's worked in the past on Styrofoam, but it wasn't very effective on this type of foam. So I ended up using a straight router bit on an extension to "route" out the foam. It really worked well, not perfect, but certainly adequate.
Here's the radio side of the sled, with the 3" tube next to it.
Here's the GPS/battery side of the sled. I'm not overly fond of using battery holders like this, but in this case I think they'll be fine. I have cable ties over battery packs and the GPS to retain them, the radio is screwed to the board.
Here's the unit assembled.
Here it's inserted into the nose cone. I tested the telemetry as you see it here, both on its side and upright. I'm getting good signals from the GPS receiver and the radio transmitter, and that's indoors.
I did some checking on batteries, it looks like regular old alkaline batteries have as much capacity in a AA as anything. The Energizers have 3 amp hour capacity at 1.5 volts, that gives me 4.5 watt hours, at 8 batteries that's 36 watt hours. The radio uses about 3 watts at full power, so that should give me around 12 hours of transmit time. Lithium AA batteries have the same amp hour rating, while they are lighter and work better in the cold, the extra weight of alkalines won't matter in a rocket this size, nor should the cold have a chance to effect the batteries.
I tested the remote launch controller last week for range. I set up the PC and radio on a table and walked down a long driveway while I had someone click on the command buttons on the PC application. At 700' the system became intermittent if the receiving radio box was on the ground, but worked well if I held it at waist height. That was using only the built in 2" wire antenna on one radio, and a 7" 1/4 wave antenna on the other radio. I do have a spare high gain antenna I could use if I need much more range, but for now it seems the range is adequate for the job at hand. The radios in the remote launch control system are MaxStream 9XCite 4 milli watt units.
Here's a new drogue chute I made for this flight.
I considered using a single 36" drogue, but that would have put the descent rate at about 99 fps, a touch fast I thought. This new drogue is 48" in diameter and should give me a descent rate under the drogue of about 74.6 fps and under the main about 27.7 fps. I beefed up this drogue as well, since I'm not as cramped for space this time around. I doubled the canopy edge with hem tape and then a layer of 5/8" nylon ribbon. The 6 shroud lines are 3 lengths of 12.5' of 9/16" tubular nylon sewn from the apex, down to the shroud attachment loop, then back up the other side to the apex again. The tubular nylon is triple stitched to the canopy and well bar tacked to the edge hem materials. I did order some new 1/2" tubular Kevlar as well. The 5/8" tubular Kevlar I have is a little short for the nose cone deployment. I'll keep the 5/8" tubular Kevlar on the main chute, and add this new 36' length of 1/2" tubular Kevlar to the nose cone and drogue chute.
I had already gone through over $60 worth of resin, not mention the plastic pellets and a week and half. So I decided to start all over again. It looks like the resin won't crack if the HDPE pellet content is around 50%, so I decided to cast the grain full length with the 50/50 mix of resin and pellets. I used about 45% of the recommended hardener to resin ratio, the resin still heated up quite a bit but didn't crack at all. I used about 2.25 quarts of resin for the grain, and it ended up about 2" longer than needed. I cut the grain to length with a power miter box saw.
On the left is the first grain cast that cracked, I did some test runs with my router for practice. On the right is the final good grain. For this engine the required grain length is 22.6". It's hard to see, but the top of the grain is routed out to allow the injector plate to fit into the recess.
Here's the router bit I used. To get the proper depth of .75" I made four passes, adjusting the router deeper with each pass.
Here's the grain in the combustion chamber. This case has a thinner wall than the previous 5" cases, so I was able to glue a layer of .045" EPDM rubber to the casting tube. I used 3M #77 super spray adhesive. The finished fuel grain weighs 9.8 pounds.
I also did a leak pressure test of the N2O tank this week. Under fairly low pressure I did detect a couple of bubbles at each bulkhead. I tried re-honing inside the tank, installed new o-rings and lubed the wall with Krytox grease, but I still had a few small bubbles when I sprayed the outside with a soap/water mix. I really think those will seal up when more pressure is applied, but both coupler ends are vented, and I don't see any real problem even if they don't seal up completely.
One last big job I have is to make the valve actuation mechanism for the launch tower. That's something I need to get the launch tower trailer outside to do, but the weather has been so poor I haven't been able to get at it. With luck things will dry out and I'll be able to get started
I spent about 9 hours on Saturday building/installing the mechanism to open the flight N2O valve. John came over mid-afternoon to help with final construction and installation. It really took a second pair of hands to hold and adjust things.
It's really a fairly simple design. A shaft goes through a tube to allow it to pivot. On the end of the shaft is a slotted bar that mates over a singe straight bar on the rocket, that straight bar on the rocket connects to a socket which turns the valve inside the casing. When the valve is opened, the rocket moves up the rail and slides out of the slotted bar. Once the slotted bar is cleared a spring pushes the socket/straight bar out and leaves it at the launch pad. The mechanism is activated by a spring and released by a PIRM2.
I took advantage of a dry, albeit windy day to cold flow test the new design. I used .75" (.285 grams IIRC) of my soda straw measure of Red Dot gun powder on both the line cutter and the PIRM2 on the valve release.
I didn't do a full load of N2O, but certainly enough to send up a nice white cloud.
Click Here for a short video of the cold flow test.
This is a video capture of the valve opening mechanism just after it activated.
This cold flow test gave me a chance not only to test the valve and set up of the rocket, but was a good opportunity to test out the new wireless control system in an "almost launch" situation. Everything worked exactly a planned, the pyro circuits all worked, continuity reporting worked and the system reported N2O fill pressure at the PC application.
After the test on Sunday I decided to apply the finish coats of paint to the nose cone and fin can. This is 3 coats of Dupli-Color candy apple red followed by 2 coats of Dupli-Color clear acrylic enamel.
That pretty much concludes this build page. With everything working fine I think I'm ready for next months launch. Again, it amazes me how much time (and money) goes into a large project like this. It isn't as easy as simply "scaling up" a small rocket. Unless you've built one, it's hard fathom what goes into it. With luck the launch will go well and be worth all the effort. Of course this rocket is just another stepping stone to bigger and better rockets to come.