liquid rocket gas generator question

[spiveycool]

I am building a liquid fueled rocket engine driven gas generator. I plan to use gasoline for fuel and oxygen gas at a rate of .022lb/sec and O2 at .055lb/min The chamber is 2.15 in long 1.15 in id with a .238 nozzle. The chamber pressure should be 300 psi with a thrust of 20 pounds. I calculate the thermal output at 2583 BTU. The velocity should be about 1mi/sec. I am trying to calculate the rate of water injection in lb/second required to reduce the temp/pressure of the output of the thrust stream to a high velocity1200 degree superheated steam suitable to drive a turbine. I am open to ideas on the location,angle and configuration of the water injector nozzles and the proper turbine A/R ratio required to optimize this this type of thrust. What thrust nozzle style would be appropriate for expansion of hot gas stream when expanding the thrust to turbine intake. What would be the proper cross section of the entrance to the turbine volute. Any thoughts on the matter would be welcome

[racket]

Hi

Those fuel/oxygen ratios sound a bit strange , could you elaborate .

This would be an extremely small turbine with extremely high rpm .

2583 BTUs ( per what time frame) will only raise ~2 lbs of water to superheated steam

Cheers
John

[spiveycool]

Sorry -should be l/b sec on both. Water injection intended to lower gas temp enough to prevent turbine from damage. I am intend to drive the turbine section of turbo charger with the thrust stream. I will need to change A/R of housing to use the high velocity gas from rocket engine. Turbo chargers are normally driven with low velocity,high mass exhaust not 3000 degree,1 mile/sec gas stream. Engine is copy of project motor described in ROCKETLABS book- How to design,build and test small rocket engines. I intend to use a GARRETT turbo and rework or cast and machine new exhaust side housing to suit application.

[racket]

OK , 0.022 plus 0.055 , that doesn't seem much oxygen :-(

Lets assume 0.077 lbs/sec of gases , only 4.62 lbs/min. ..........OH !!

Even adding on the 2 lbs/min of water thats still only ~6.6 lbs/min

The smallest Garrett turbo flows ~10 lbs/min and spins >200,000 rpm .

Theres no point feeding high pressure gases into the turbo turbine stage as its only designed for ~30psi max , at 300 psi it would need to be only a tenth the size .

You may need to rethink this one :-(

Cheers
John

[spiveycool]

Turbochargers are designed to operate with high volume/low thermal expansion exhaust gas in an application that is very limited in the back pressure that will allow the turbo to work correctly. Low A/R ratios drive the compressor with much less mass gas flow and higher back pressure. When this is done on a turbo installed on a motor low speed air flow is high but high back pressure prevents higher engine speeds.The turbine maps show that as you reduce the A/R ratio the mas flow required to compress the same lb/min of air to the same pressure ratio drops and back pressure goes up. I realize that I will have to fabricate a housing to suit the radically different flow characteristics involved. In the early 1960s Turbonique of Orlando fl marketed a mono propellant (N-Propyl-Nitrate) powered turbo. I have studied the engineering that went into this set up in detail and have a full set of plans for all their parts. This system burned .127 lb/sec of the n-p=n fuel. This fuel decomposes at 2000-2200 F. This was done at a chamber pressure of 300 psi and gave a gas velocity of 1 mi/sec. At 7420 BTU /lb this generated a thermal output of 942 BTU/sec and a gas flow of.127 lb/sec . Gasoline/O2 has a Is of 260 and at a 2.5/1 mix rate burns at 5750 F and 20,855 BTU/lb This gives a output of 4583 BTU/second @.022 lb/sec. 42.41 BTU=1hp is the conversion factor At 4583BTU/sec the thermal output works out to be 648 horsepower. I do not know how to figure how much kinetic energy is contained in the gas flow of 4.62 lb/min at 1 mile/sec and may need to adjust lb/sec to suit the task at hand. On the face of it there would appear to be more than enough energy to do the amount of work in question. As this was done 50 years ago with very crude turbine wheels,lower thermal value fuel etc, I know there is a way to harness this power to do this. I will find a way to bring this setup into the modern era.

[racket]

Hi

OK , a modern Turbonique using a bottle of compressed oxygen, and gasoline at several hundred psi injection pressure, feeding a rocket motor .

The rocket motor exhaust is too hot for the turb wheel so you need to cool it down...............why?? ...............how long will the rocket motor burn for ............3, 4 or 5 seconds at a time .................longer??

You mention a 0.238" nozzle , is that throat diameter , or exit diameter ??

I wouldn't be thinking of turning the "cooling" water into superheated steam , theres too many BTUs required in the phase change , I'd try and keep gas temps and velocities high to maximise kinetic energy whilst using "liquid" water to cool the blading during the very limited duty cycle.

A "normal" turbo can experience gas velocities of ~2,000 ft/sec at the turbine tips , with tip speeds ~1700 ft/sec max .

I'm assuming your rocket velocity of 5280ft/sec ( 1mi/sec) is at atmospheric pressure at the exhaust, if so then there'll be "problems" if fed into a standard scroll as theres the requirement for a pressure drop across the single convergent nozzle within the scroll , you'd need to find a scroll whose throat was the same area as your rocket exit nozzle and plumb your gases directly to the scroll throat without any cross sectional changes of area .

The turb wheel inducer to exducer area ratio would also need to be looked at , an exducer area much larger than the inducer would be required due to the lowered gas velocities at the exit , you'd be looking at all impulse energy transfer to the wheel's inducer tips and no reaction at the exit .

Yep , it could be done :-)

Cheers
John

[ernie wrenn]

I am not sure who won... but I agree. Remember, its only Rocket Science.

ernie

[racket]

Hi Ernie

LOL.......this is a complicated one , but I think we'll find a good solution :-)

Cheers
John

[spiveycool]

The nozzle minor dia is.238 exit is 15 degree and .4555 at the exit producing exit flow at 1 mi/sec . I have studied all the tech info I can find on the net on all types of radial inflow turbines and it appears that an A/R ratio of .30 to .60 should work. Would I be better off using a over or under expanding nozzle? I believe that the use of the rocket motor as a flow inducer in a bell mouth air intake. The mixing of the rocket thrust stream along with very fine mist of water should drastically lower over all temperature at turbine intake and increase mass flow. The Sr 71 at cruse had only a small part of the flow thru the engine,most flow was bypass air. very different set up but the same idea. I have a feeling that this will have to be determined by trial and error. I would like to be in the basic range with the prototype and any input would be helpful.

[racket]

I'm still not certain a radial inflow turbo turbine is going to be your best bet as there needs to be a pressure drop as the gases spiral inwards whereas a pure impulse axial wheel wouldn't have that problem , an under expanding nozzle on the rocket motor would provide enough pressure ratio for the radial inflow , maybe a 3:1 area expansion ratio instead of the ~3.6:1 ratio .

I don't think trying to cool the rocket exhaust would work by surrounding it with a fine water mist, you'd end up with the same very hot core surrounded by a shallow ring of steam probably .

Perhaps a tube centralised in the rocket exhaust through which water is pumped onto the turbine blade face , the rocket exhaust would splatter the water and cool both blade and surrounds .

With a 0.30 A/R housing and an "A" of just 0.1629sq ins ( 0.4555 dia ) the "R" would be minute at 0.543 inches , the turbine wheel would be less than an inch in diameter , .........you can't use A/R for this engine .

The cartridge starters like the Hamilton Sundstrand ..............http://www.govliquidation.com/auction/view?auctionId=5099947
only blows hot gases onto a couple of blades of the axial wheel at a time , you need to think similarly , by having the turbine wheel running within a close fitting shroud around the inducer with your rocket exhaust feeding onto one blade only , I'd suggest using a Cummins VT-ST 50 turbo to experiment with .

Cheers
John

[racket]

Have thought considered /looked at , using a fuel/air starter www.youtube.com/watch?v=EOEXRPfgm7M , this uses compressed air at 350psi rather than your proposed oxygen , and burns the fuel in a very small combustor similar to a turbine engine's so that exhaust temps are low enough for the turbine stage to cope with.

Theres no need for the complexity of water "cooling" or the need for the more expensive oxygen , just high pressure compressed industrial air bottles , the aircraft stored a total of 1,300 cubic inches of compressed air at 3,000 psi and was enough for 2 starts .

Cheers
John

[spiveycool]

How will using a flow inducer bell to guide the air around the motor influence temp/mass of gas. I was thinking the A in A/R in this case would be the cross section of the bell plus rocket nozzle area minus the area occupied by the rocket motor. Is there any way to figure the total mass/temp correction with the induced air. How much cooler would gas be if A/F ran very rich? VT/ST turbo is a good idea. I hope I can find compressor/turbine maps. Perhaps it combined with modified/fabricated volute housing may work. The original Turbonique turbine blade was a bucket design with a vane guide. Perhaps an up date of that wheel/volute might be the answer. The desire to use as many off the shelf parts as possible is the main reason to explore turbocharger mixed flow turbine wheels. Starter set up is cool but will not scale up to the horsepower level I need.

[racket]

A high velocity gas stream is virtually "solid", its difficult to get any mixing or even entrainment of ambiant air , you'd need probably 30 nozzle diameters of axial distance to start seeing much entrainment .

An overly rich mixture would bring temps down but probably no where near enough for the turbine wheel to survive for any length of time .

You most likely will not find even a comp map for a VT-ST50 , turb maps are non existent except for the Garrett GT range .

What horsepower are you designing for ??

Cheers
John

[racket]

Hi

In one of your earlier emails you enquired about kinetic energy for 4.62 lbs/min at 5280 ft/sec , kinetic energy is half mass times velocity squared .

4.62 lbs/min = 0.077 lbs/sec ,so 0.5 X 0.077 X 5280 X 5280 , divided by 32.2 = 33,333 ft lbs/sec or 60 hp as theres 550 ft lbs/sec /hp .

A rocket engine flowing 4.62 lbs/min - 0.077 lbs/sec with an exhaust velocity of 5280 ft/sec will only have a thrust of a bit over 12 lbs , not much :-(

With our turbine engines we get roughly a shaft horsepower per pound of pure jet thrust when the gases are fed through a freepower turbine wheel , your 12 pounds of thrust will be doing pretty good if it gets 50 hp .

Something isn't adding up here :-( .....................are you sure about your flow figures ??

What was the Turbonique flowing exactly to get their claimed 1,000 hp ??

Cheers
John

[spiveycool]

My objective is to generate a high energy gas flow that is appropriate in flow/temperature/pressure to drive a Garrett GT6041 turbine to its maximum output. Perhaps a subsonic nozzle mounted in a bell mouth tube acting as a flow inducer with a afterburner chamber with water atomizers to generate steam at high temp/velocity would do the trick. I am open to any and all ideas on the best way to solve this problem. The largest Garrett turbo the GT 6041 needs 80 lb/min to drive it using a 1.25 A/R housing. Looking thru the turbo maps I see the GT 5533 that is available with 5 different A/R housings from .90@55lb/min to the 1.40@70lb/min. When used as designed .90A/R gives great low rpm power but has excessive exhaust back pressure for high rpm when used on internal combustion engines. It looks as though as the A/R ratio goes down the mass flow requirement goes down at the expense of increased back pressure. I am not driving the turbine with exhaust gas and am not concerned with back pressure. It would stand to reason that a much lower A/R [.30-.45] would reduce the mass flow requirement albeit at much greater back pressure. I would fabricate/modify a scroll housing if necessary but would prefer to use off the shelf parts if possible. What would be the optimum combination of drive gas pressure/temp/scroll A/R? I am not concerned with fuel consumption but would like to minimize O2 use. All suggestions are welcome.