Home / News

News Archive

Fuel experiments, Engine tests

Fuel Experiments

April 29 and May 3, 2003 Meeting Notes


Fuel Experiments


We did a good set of experiments with different fuel / catalyst mixtures this week, working towards possible bipropellant options with 50% peroxide while we are waiting for more 90%.  A few things are looking good on the 90% front, but we are still pursuing backup plans.


We did another experiment with the mixed-monopropellant scheme, dissolving 8% by mass glycerine with 50% concentration peroxide, in hopes that permanganate catalyst solution would auto-ignite the mix based on the energetic permanganate / glycerin reaction.  No luck.  We have seen a few references to these types of mixed monopropellants in use long ago for torpedo drives, but we haven’t been able to get them to do anything.  We haven’t been too disappointed, because mixing up large quantities of an oxidizer / fuel mixture would be a bit nervy.  With high concentration peroxide, that is a sure detonation risk, but 50% water is supposed to desensitize it sufficiently for operational use.


The only thing we have seen auto-ignite with 50% peroxide is red phosphorous, which is generally nasty stuff that probably wouldn’t dissolve in anything anyway (and we don’t intend to find out).


We have heard from a couple places that 50% concentration peroxide burning with fuels is not self-sustaining without a catalyst of some kind, and our silver screen catalyst packs weren’t even close to decomposing all the 50% peroxide we tried to flow through it, so the three possible options are:


50% peroxide mixed with a fuel, plus an aqueous catalyst

50% peroxide, plus a fuel / catalyst mixture

50% peroxide, plus a fuel, plus an aqueous catalyst


A triprop with only a 160-180 Isp would be rather depressing (and adding more water wouldn’t be helpful), and the peroxide / fuel mixtures are scary, so we would really prefer to find a good fuel / catalyst mixture.


Our previous tests with dissolved catalysts in fuels had used manganese acetate dissolved in ethanol with the addition of some epoxy hardener to act as a promoter.  I don’t care for the three-part mix, so we tried dissolving potassium permanganate in several different fuels:



Isopropal Alcohol

Furfural Alcohol




The only one that lets any dissolve is acetone, which holds about as much in solution as water does.  It is hard to tell how much is in solution with the dark purple color, so we ran the solution through filter paper to make sure we didn’t have any suspended solids.  We also have some sodium permanganate on the way to try later, but the acetone / permanganate mixture does seem to be stable and workable.   There is still a notable concern with leaving deposits in the plumbing when the acetone evaporates away, and it does generally make a mess.


One other experiment of note:  if you drop a few grains of potassium permanganate into a larger quantity of peroxide, it will catalyze the peroxide for a while, then completely vanish.  This consumption of the catalyst probably bodes very poorly for the longevity of ceramic catalysts with baked-on permanganates that are often proposed for 98% peroxide use.


The more we work with other propulsion options, the more we appreciate the benefits of 90% peroxide / kerosene with catalyst pack based auto-ignition.



Engine Experiments


We finished the boring and tapping of the distribution manifolds for our big cooling jacketed aluminum motor.chamber.  This was built for a 1000 lbf regeneratively cooled 90% peroxide / kerosene engine, but we figured we might as well use it for our 50% peroxide engine tests.  The cooling jacket is more restrictive than it probably should be: there is a 0.010” gap between the inner and outer walls (4.5” diameter), which gives 0.143 square inches, which is only mildly less than the feed line, but all the wetted surface area adds a lot more drag.  We will probably turn the inner section down another 0.010” or so in the future, but we are just using it connected to a water hose for now, and the 50% engine won’t make all that much heat, so it isn’t a priority.


I added data collection options to my program that sequences all the ignitor valves and spark plugs, so we have a nice little test bed.  I also added key commands to manually actuate each of the controlled solenoids for testing.  I installed a pressure transducer on the engine with a standoff line and a snubber, which should keep it from cooking.


We have previously noticed that it isn’t that hard to get a fairly big bang when the XCOR torch igniter lights, if the chamber has any residual fuel vapor in it.  Working with volatile acetone based fuels today made it a lot worse, so we finally got tired of the bangs and installed a fairly high flow nitrogen purge.  Yet another valve to control.


The igniter was delivered from XCOR with 24VDC solenoids, because they typically work with airplane electrical systems, but we will be swapping the coils for 12VDC to make our lives easier.  The solenoid company http://www.snap-tite.com/divisions/sv/index.html has a wide range of solenoid offerings, which we may start using some of in the future.  The NOS nitrous solenoids that we typically use have a pretty big price markup, and they are overdriving the coils quite a bit to make them operate at higher pressure / flow rates, which makes them burn out if left on too long.


We had hit-or-miss results with the testing.  Sometimes we got it to light, and sometimes we just had a gush of liquid out the bottom of the engine (we are definitely doing all this testing vertically, to avoid any chance of pooled propellants in the chamber).  We believe that we need to get the torch igniter operating at a higher pressure for a longer torch flame, but I need to find a high-pressure oxygen regulator to do that.


We also should measure the actual flow rate through our two feed systems.  The spray nozzles are sized to give the correct ratio, but the solenoids and plumbing may be skewing the results.



© 2001-2011 Armadillo Aerospace, LLC. All rights reserved.