Rick Dickinson on April 17, 2012

Traditional model and high-power rocket construction relies on cardboard or fiberglass tubing for the “body” of the rocket.  At this scale, that’s going to be incredibly heavy, and possibly unworkable (not to mention, incredibly tough to transport to the launch site).

Instead, I’m thinking of building a more lightweight structural framework for the rocket body that can be bolted together on-site, and then “skinned” with a series of body panels.  Each panel can be an identical lightweight composite sheet, laid up and formed on a cylindrical mandrel, which can be simply screwed into place on the rocket’s frame once it’s assembled.


Rick Dickinson on March 5, 2012

I’ve been working off the assumption that the eventual full-scale launch of the Big Purple Crayon Rocket would be on a “Q” motor, as that seems to be about the right size of motor for the size of rocket that I’m contemplating.  To budget for this, I need, among other things, to start figuring out how much propellant I’ll need to make, and how much of the various chemicals I’ll need to buy.

As a first cut estimate, I’ll make a few simplifying assumptions:

  1. Performance (in terms of specific impulse) will be decent, but not great.  An Isp of around 200s seems like a reasonably attainable ballpark goal.
  2. I’ll work out my estimate based on the motor being at the upper end of the “Q” impulse range.  If the actual motor I build turns out to be smaller, then that will just be a savings in propellant.  The Q impulse range runs from 81,920.01 N-s to 163,840 N-s, so I’ll estimate that my motor will have 163,840 N-s of total impulse.
  3. I’ll calculate the AP200 fraction of my propellant at 75% of the total mass.  In reality, the formulas I’ve been experimenting with have all used about 60% AP200, and some additional quantity of other sizes of AP and other oxidizers.  But, those numbers may change as I refine the formula, and I’d rather over-estimate than under-estimate.


\frac{163,840 N-s}{200 s} = 819.2 N

We’ll need about 819.2 Newtons of propellant weight.  Let’s convert that to mass in Kg by dividing by the gravitational acceleration:

\frac{819.2 N}{9.98 \frac{m}{s^{2}}} = \frac{819.2 Kg\cdot \frac{m}{s^{2}}}{9.98 \frac{m}{s^{2}}} = 82 Kg

This means that our theoretical 100% Q motor, with a 200s Isp, will take about 82 Kg of propellant.  Assuming 75% of it is 200 micron Ammonium Perchlorate (AP200), we’ll need:

82 Kg \times 75\% = 61.5 Kg

For some reason, chemicals are typically sold by the pound, rather by than kilogram, in the USA.  (We’ve officially been “on the metric system” in this country since the mid-1970s, but you’d never be able to prove it in practice.)  Converting to pounds of mass (yes, I know… *sigh*), we get:

61.5 Kg \times 2.2 \frac{lb_{mass}}{Kg} = 135.3 lb_{mass}

So, if I order 150 lbs of AP200, I should have plenty for my Q motor, as well as a number of smaller test motors.

Tags: , , , , , , , ,

Rick Dickinson on October 24, 2011

The Putative Purple Propellant v1.1 that I cast this past weekend hardened up nicely, so it looks like the problem with the curative ratio has been fixed.

I cut and cored a few 38mm grains with 3/8″ cores, in preparation for a test burn in a Dr. Rocket (Aerotech-compatible) 2-grain 38mm motor casing.

I also finished the machining on an aluminum delay well plug that I had started a couple of years ago — it fits into the delay well with an o-ring seal, and then also has an aluminum plate that replaces the forward fiber washer that Aerotech uses to provide a sealing surface for the forward o-ring.

Assembled delay well plug and forward closure with o-rings


Firing the motor will need to wait, for now, as I need to buy a few of the nozzle end o-rings to finish putting the motor together.

Tags: , , ,

Rick Dickinson on October 22, 2011

For my next try, I’ll be making a few changes to the basic formula, in an effort to come closer to my “ideal” purple propellant.

For the record, here’s the formula for Putative Purple v1.1:

  • AP200 (60%)
  • AP90 (8%) – this is reduced from the 10% in v1.0, to make room for more Mg.
  • Sr(NO3)2 (10%) – this slight increase will hopefully add a bit more red to the flame when burned under pressure.
  • Mg 325 (5%) – this is an increase from the 3% in v1.0, in an attempt to get a less-transparent flame.
  • CuO (1%) – this is reduced from v1.0, to hopefully tone down the blue component of the flame when burned under pressure.
  • Tepanol (0.3%)
  • DOA (3.5%) – slightly increased to make it more “workable” when casting
  • HTPB R45-HTLO (10.2%) – slightly decreased to increase NCO:OH curative ratio
  • E744 (2.0%) – increased to increase NCO:OH curative ratio.

Assuming a 0.01% moisture content in the final propellant, this gives an NCO:OH index ratio of about 1.04:1, which should give a nice solid propellant with a complete cure.  (We’ll see — I hope to cast this weekend.)

More to come….


Tags: , , ,

Rick Dickinson on October 18, 2011

Last weekend, I finally had a chance to test some of my Putative Purple v1.0 propellant.

My first impression upon examining the cast propellant was that it was rather soft, as though it was a slightly incomplete cure.  This could be due to residual moisture in the chemicals due to the humidity in the air; when I work on v1.1 of the propellant, I’ll want to up the curative quantity to compensate.

I cut several 29mm grains to 1.75″ long, which is the same as the standard length of Aerotech’s H128 motor grains.  I drilled 1/4″ cores in three of the grains (by hand, holding the drill bit in my fingers), and then burned off the cored scraps in a stainless steel bowl in the middle of my driveway to assess the color at 1atm pressure.

Success!  The flame color was a very pinkish purple.

Buoyed by this initial success, I decided to do a burn test in a motor casing.

Due to the soft nature of this propellant batch, I worried about runaway Kn increases due to erosive burning.  So, I set up a Dr. Rocket 29/180 casing with the phenolic nozzle from an Aerotech H128 reload (0.181″ throat), and only two of the three grains.  I plugged the forward closure to avoid needing to worry about a delay grain.  I put tape over the forward end of the topmost grain, and then inserted the two grains and a grain-length piece of empty casting tube into the liner.  (This is the same technique Aerotech uses for their D & E reloads in the 29/40-120 motor casing.)  The motor was assembled otherwise according to the usual Aerotech instructions.

I pounded a dowel into the ground in the back yard to make a motor-sized hole in the dirt, then inserted the motor, nozzle-up.  In the event of a casing failure, this should be somewhat safer than an open-air motor test, as it’s likely to contain (or at least significantly slow) any pieces that are thrown out horizontally, and only allow pieces to be thrown straight up in the air.

The burn was fast — similar to the burn rate of Aerotech Blue Thunder propellant.  The flame color was mostly blue, with a reddish tint visible mostly around the edges.  The flame was quite clear, as well, making it even more difficult to view the color.

All in all, the test was a success, and pointed out a number of potential changes to make in the next version.  I’ll probably save the rest of the cast propellant for the fire pit at ROCstock, as I’m not too happy with the hardness or the flame color when burned under pressure.

Tags: , , ,

Rick Dickinson on September 27, 2011

After consulting with a few of my more-experienced propellant-head friends, I’ve come up with a formula that should give me a decent purple flame color.  Hence, “Putative Purple Propellant”.

I mixed up a 300g batch Sunday, and cast some in 29mm and 38mm sizes for preliminary testing.  Once they’ve had a few days to ensure that they’re completely cured, I’ll cut them into grains, and drill the cores to give me a relatively neutral “BATES” grain burn profile.  I’ll do my initial testing at a fairly conservative Kn ratio, as I’m not interested in absolute performance numbers, yet — which is good, since I haven’t built a test stand for data capture, yet.

Instead, I’m mainly looking to get a good purple flame color and a smooth burn.

Read the rest of this entry »

Tags: , , , ,

Rick Dickinson on September 26, 2011

It’s not sensible to tackle a project of this magnitude all at once.  The only way to succeed is to break it up into more-manageable chunks, and work my way through them, one at a time.

Let’s start by stating the goals of the project:

I will design, build, fly, and successfully recover a very large (roughly 30′ tall) crayon-shaped rocket, using a rocket motor of my own design and construction, at the BALLS launch in 2012 or 2013.  The rocket will be designed to resemble as closely as possible an actual wax crayon, except MUCH larger.  The crayon will be purple, and the propellant flame color should be purple, as well.

Given these goals, there are actually a number of development projects to undertake:

Tags: , ,

Rick Dickinson on August 25, 2011

What’s the largest crayon you’ve ever seen?

This project is going to dwarf that.  How does a crayon roughly 30 feet tall grab you?

Now — to make it fly!  I’m leaning towards either a cluster of “P” motors, or a single “Q” or “R” motor, depending on the eventual weight estimate.  I’ll want to keep it down below Mach 1, but I’d really like to get some serious altitude out of this.

Tentatively, I’m planning to fly this at BALLS 2012, which will probably be in late September or early October, 2012.  That gives me roughly a year to plan and build.  As a fallback, BALLS 2013 is also a possibility.

Construction ideas to come in future posts….

Tags: , , , ,