Regeneratively Cooled Engine 


Update 2007-08-12

Big overpressurisation. Worked like a bomb:

Video (slow motion)

The sugar starter grain:

The nos and petrol valves opening:

A millisecond later:

Any theories?

I favour the "The sugar propellant pressure went higher then the nos tank pressure and pushed into the nos tank decomposing the nos and causing and explotion" theory. There was still quite a bit of suger propellant in the combustion chamber when the nos valve was opened.



Update 2007-08-10

I made a new like impinging injector for the fuel. This should improve combustion efficency quite a bit:

It used to spray like this:



Update 2007-07-23

We tested the regeneratively cooled engine on Sunday. It was a success. But we need to tweak the injectors and nozzle a bit. Video

Frans' Observations from the video:

Studying Zechy's video footage my observations are as follows...(time in seconds)

T + 00.00 : Beautiful and peacefull sunday morning somewhere in the wild.

T + 01.00 : Ignitor smoke is seen emerging from the nozzle.
 There is a slight breeze from the right of the video.

T + 02.00 : The sugar grain is starting to burn to full capacity.

T + 03.25 : The engine is rocking under the strain of the thrust (sugar)

T + 04:25 : Let rip...
 Approx. 1/2 meter length of flame emerges from the nozzle.
 The valve lever actuated in a small puff of smoke (left bottom)

 Shock waves are seen just above the nozzle exit
 The engine is pulling to the right under thrust.
 - possibly due to engine pressing the support brick into the sand.

 Some mach diamonds are briefly visible - appears like six in total (?)

 The flame is almost always surrounded by a dark cloud
 ...implying incomplete combustion. Possibly fuel/oil rich.

 I am sure the oil added for lubrication could help here, however
 normally two stroke oil burns with a white smoke (go-kart)

 There regularly appears a large cloud of plasma at the top of the flame
 Interestingly sometimes it is down-wind (left)
 - probably excess fuel/oil mixing with the surrounding air.

T + 05:00 : The video camera is also having some fun here...
 Even when l o n g flames emerge from the nozzle there is a cloud 
 of plasma at the top - excess fuel/oil mixing with air ..?
T + 06.00 : The flame dies (no fuel) - there is still vapour(N2O) coming from the nozzle
 The engine moves to the left returning to the 'no thrust' position

T + 09.00 : The vapour begins to subside - N2O too is finished. 

T + 10.00 : Peace and quite returns to the area.

T + 11.00 : Video ends.

Summary : Assuming each frame is 40mS (PAL TV @ 25 f.p.s.)

1. Good job.
2. Sugar grain took a second to get fully ignited. Can be improved.
3. Sugar took just over another second to get to 'full' thrust (see 1)
4. Another second later the 'liquid engine' was activated.
5. Engine ignited without a problem - 'full' thrust within 1/4 of a second.
6. The mixture was extremely fuel/oil rich - therefore the short burn &
7. There was another 3 seconds of N2O left (needing fuel)
8. Fuel/oil lasted just longer than (only) a second. Fuel injector way too BIG.
 This can also be to pressure losses in N2O plumbing..!
9. Nozzle way under-expanded. We knew this.

*** N2O regen' cooling can work - no melting & no chamber discoloration at all..! ***

My suggestion(s)

We use exactly the same setup to retest, but

1. 1st everybody writes their comments and observations (can also have a meeting)
 - may see/understand something the others did/do not.
2. We obtain additional video footage from Mr. John to review (maybe faster frames)
3. Fuel injector 1/2 the current size (maybe even slightly smaller)
4. Throat can also be slightly smaller - better combustion stability (can enlarge later)
5. Better nozzle expansion. Not just a throat..! - we need to do this anyway.
6. We should be able to get a proper 4 second burn..?


1. Do not introduce additional variable(s) un-neccessarily. One step at a time.
2. We use little N2O to re-test & fix if neccessary.
3. Save time & effort by re-use of existing system(s).
4. Can do bigger tanks later when it works better.

The setup


The fuel injector size is wrong becasue we ran out of petrol after 1.4 seconds and then run on nos only for 3 seconds. The nozzle was sub-optimal because we simply increase the throat size of an old nozzle. Frans will make a new one for the next test.

We need to do some test to check the coefficient of discharge for the nos and fuel injectors and fix the sizing and corresponding O/F ratio


Update 2007-07-19

I have increased the thrust of the engine to 1800N. We plan to test on Sunday. We need the increased thrust for a 10km challenge issued by


Update 2007-04-26

I have repaired this motor and will test it soon (DV)


Update 2006-12-18

Never use cardboard to bind the starter grain to. The cardboard blocked the nozzle and the motor overpressurised:


Update 2006-12-16

Here it is - the regeneratively cooled nos and petrol bi-propellant motor.

The motor is cooled with nos. Liquid nos is fed in at the bottom of the combustion chamber between the combustion chamber wall and the outside wall. There it gets vaporised (due to temperature and pressure drop) and flows up to the injectors. There are eight nos injector holes which have been sized for nos vapour (2.2mm). These eight nos streams impinge each other and the fuel stream coming straight down the center of the combustion chamber.

Here you can see three of the nos holes and right at the bottom of the cup, the 1mm fuel injector hole.


With the combustion chamber wall in place, it is easier to see how the nos will flow past the wall (down in this case) and into the injector holes.


Here the combustion chamber outer sleeve is shown in position.


The nozzle with a graphite insert.