Creamy Robot Goodness

Well it’s Friday. “Red Glare” (the big launch) started today. I’ll probably end up working on Jujubee all day tomorrow. I wish everything was done, but I do think I’m in ok shape for Sunday. In just a couple of days I designed and built my staging timer. It uses a 250g accelerometer, an AVR, and a mosfet. It lets you, using a serial port, set the acceleration threshold for launch detect, and to set the ignition delay in 100ms increments.

I decided to run copper tape down the side of the rocket to carry the current from the staging timer. I ran a number of tests and it seems very reliable in its current configuration. I can power it up, pull the reset jumper to start the safety lockout timer, and then (after waiting for the timer) a quick smack on the side of the board simulates the high acceleration needed to trigger ignition. After the ignition delay, the igniter burns beautifully. I did, however, end up having to make my own low-current igniters because a 9v alkaline battery wasn’t able to provide enough oomph. Below is a simple explanation of how I made them, for posterity:

Materials:
- 26 gauge shooter wire (aka zip cord, etc), the first 1/4″ or so stripped
- soldering iron, solder, paste flux
- 40 gauge nichrome wire
- Magnelite dip

The concept is that, unlike the pre-soldered igniter wires that often come with Magnelite (see http://www.publicmissiles.com / Webstore / Igniters), we only want a very small loop of nichrome to bridge between the two copper wires (rather than wrapping it multiple times around the insulated section of wire). Using a small piece of nichrome wire (1.5″ is plenty, but most will go to waste), wrap one of the copper conductors several times, very close to the insulation. Then make a bridge, looping halfway around the insulation, to the other copper wire. Apply some paste flux to the two places where the nichrome wraps around the copper. Put a blob of solder on the tip of your iron, and touch it to the flux and copper. This should coat all the nearby copper in solder instantly, and tack the nichrome down adequately. Trim up the wires, carefully bend them into shape if you need very small igniters, and dip in Magnelite. I double-dipped mine after a few minutes to increase the amount of pyrogen per igniter, but I haven’t tested these yet. [Update: I tested most of the double-dipped igniters and one of them had a pretty violent pop. Guess that answers that.]

So the biggest launch on the East coast in 2006 will be October 20-22nd in Price, Maryland. I’m planning (if all goes my way) to put Jujubee through the sound barrier then. I’ve got my carbon fiber all ready to go to make a super-light but super-strong airframe… I’ve got my 29mm “I” rocket engine and reloads. The old altimeter should work for this project. I’ve ordered a radio tracking system from http://www.theplanelocator.com so I can recover it. I still haven’t nailed down the total design though. In case I don’t come up with a good rocket ignition system I’ve ordered an RC-based system that should do the trick, but I really need to get on that. Anyway, I have a month and things look pretty good at the moment.

First things first, check out this great video of the cannon launch! Best one I’ve seen by far (thanks to Chuck Rudy at MDRA!), it follows the rocket really great:

http://www.mdrocketry.org/photos/esl100/ChuckRudy/Full/ESL100CanonLaunch.mov

Ok, since the cannon launch I’ve mostly been working, and taking it easy. I had decided to scrap the current altimeter (a MiniAlt WD) and build my own solution that could detect launch from the cannon, airstart the rocket engine, and handle dual deployment. I started designing it today, when I thought I’d check the current offerings of the G-Wiz guys. You may remember it was the MC2 that sort of let me down last year when it died after I [cough] slightly modified it to fit inside a 29mm rocket. Well… the LCX looks bloody perfect! It should fit VERY easily into the 29mm rocket, and it looks like it has all the features I need, and one version can even have a 200g accelerometer! Ooooohhh…. I’m excited. Now I just need to find somebody that CARRIES the 200g version.

What else is new? I’ve ordered an Aerotech 29/360 system, along with 3 I-200 reloads. Holy crap that thing will scream. RockSim is estimating 1300+ fps and 9500 ft altitude, without the help of the cannon.

Speaking of cannons, I’ve been thinking a lot about a future cannon. This one would pull out all the stops. Aluminum instead of PVC? A bigger better valve? Replace compressed air with something with a higher speed-of-sound like helium or even steam (those crazy launchers that fling fighter jets off of aircraft carriers use steam)? Use a sprung piston to compress the working fluid at firing time the way high-performance air rifles and the crazy gas guns do? Evacuate the barrel in front of the cannon? The rocket would have to be carbon fiber or something similar. But this kind of a monster could pack an incredible punch.

I added some pictures and text to the Jujubee page.

I’ve been having some trouble getting my altimeter to fire my electric matches. I don’t have any low-current e-matches and I can’t buy them except at the launch without a LEUP (low explosives user permit), which I probably couldn’t even get given where I live. Anyway. Matches aside, Jujubee was all ready this weekend for a motor-free test shot out of the cannon (no rocket engine, since my ignition circuit isn’t done yet). Gotta do something about that soon. If I can just find my nichrome wire I can try and scratch something… The steel wool bridges weren’t working on the altimeter this time.

The ignition circuit has almost been completely designed. Basically the only thing holding it up is my laziness because I don’t have all the parts in my library already and I’m getting tired of making them myself. But it’s really close. Once it’s designed it should be a fairly simple matter to etch the board on my mill using pcb-gcode (which is fantastic, btw). But then I still have to design and make the wireless arming box, and THEN write the firmware for BOTH devices. Yeah. Less than 2 weeks left to do all that in my free time? I’m a bit skeptical myself.

Quick update on Jujubee. The new scheduled launch date is August the 26th, a Saturday. This will take place in Price, Maryland, with the Maryland Delaware Rocketry Association (www.mdra.org). I’m going to try hard to have this rocket ready for a low-power test run in the next week or two, possibly even the SoJARS launch next Saturday.

I’ve decided that launching on the small H is probably too unsafe to start with, and on such a small field the chances of losing it are just too high. Even on a G motor, RockSim is currently predicting a mile-high flight, so I’ll probably start with an F40 or something (sim is estimating about 3,000 feet — still pretty darned high for such a tiny rocket and a small field). It’s really going to be good to have a radio transceiver onboard the rocket, hopefully that’ll help me find it again. (I’m hoping I can make the ground station antenna directional by, say, recessing it inside a coffee can. Got a better idea? Let me know.)

The altimeter bay is almost done. I need to re-cut the 29mm phenolic tube to have squarer ends, machine some parts to double as electric match mount points and shock cord mount points, and install some nuts, and I think the bay is done. The couplers were machined out of PVC and look (and work) great. I love PVC, btw, it’s so great to machine. I’m really happy with this altimeter bay design. I’d like to write an article about Jujubee after the (successful) launch, possibly to submit to a magazine. At that time I will provide pictures and some more explanation.

Ordered a Magnalite kit. I think I’m gonna make my own igniters using the steel wool bridge I’ve had such good luck with. (The idea is to use an extremely short length of steel wool. The resistance is really low so it pulls tons of current, but it heats up so fast that it pops instantly. I’ve had good luck using a suitably-beefy capacitor to give it some oomph, usually when fed from a 12v N-cell battery.) Dipping this in Magnalite should provide a decent way for my custom ignition circuit to fire up the rocket engine upon exiting the cannon barrel.

The last major hurdles remaining are almost all related to electronics. I need to build and mount a circuit on the rocket to detect cannon launch and ignite the rocket engine, and I need to build a control box that will send the authorization code to the rocket to enable the ignition circuit. (Safety first, kids.) It should also help me locate the rocket after it’s come back to earth. The circuit that resides inside the rocket is to contain one N-cell (it’s got to be light since it’s behind the CG!), a beefy capacitor, an accelerometer to detect launch, a radio to talk to the control box, a power MOSFET to ignite the rocket engine, and an AVR microcontroller to tie it all together and work the magic. Oh man, this is gonna be so cool.

I’m still alive! Ok, quick update… I have moved Jujubee back to the top of my priority list. I’ve completely redesigned the entire rocket, sacrificing some strength for accuracy, reliability, and performance. The design is still in progress so it’s bound to get heavier, but currently RockSim is estimating that it will go about twice as high as it estimated for the last Jujubee (it’s currently saying 7200 feet up, and 1020 feet per second, WITHOUT the cannon). Very, very impressive. I’m a bit nervous because RockSim’s model reports that the rocket is stable, but Barrowman says it’s way unstable. Fortunately this rocket is small enough that the ol’ string test is still doable.

Tonight I CNC’d the fins out of copper-clad 1/16″ fiberglass. This CNC mill is kicking butt for this type of work, and it’s fantastic for slotting the airframe accurately as well. I plan to fiberglass all the fin roots together so this thing will be just about bulletproof. Not as tough as the last rocket, but it will be much easier to set up and it’ll have much less chance of cracking upon launch. One interesting thing about the fins is that I designed them myself more or less through trial and error in RockSim. I don’t know if this is some artifact of its simulation mechanism that is incorrect, but it claims that this particular fin design is extremely efficient — very very short and stubby, all the way at the back of the rocket, but there are several of them (5, currently). I hope RockSim isn’t lying.

This rocket (and the entire system) is going to be pretty hardcore. I’ve got a radio link so I can arm the rocket remotely (after it’s been loaded into the cannon) as well as to help me find it again… I’ve got an awesome switch cover for the “Arm” button on the wireless arming circuit… a keypad for entering in the “launch codes”… It’s going to be great if I can pull it all off. Currently, the launch date is scheduled for Sunday the 27th of August. Gotta be ready!!!