Community > Forum > Technology & Science > Rocket Pumps

Rocket Pumps

Posted by: WannabeSpaceCadet - Mon Jul 31, 2006 9:31 am
Post new topic Reply to topic
 [ 18 posts ] 
Rocket Pumps 
Author Message
Space Station Member
Space Station Member
User avatar
Joined: Wed Nov 10, 2004 6:12 am
Posts: 321
Location: Melbourne, Australia
Post Rocket Pumps   Posted on: Mon Jul 31, 2006 9:31 am
I've been thinking about alternatives to turbo pumps, at lower pressures.

How do you calculate the required shaft power for a pump (of moderate efficiency) at say 100 lt/s, 1000 psi output, 50 psi input, fluid density 1.1 g/ml?

How much would such a pump weigh (not including the motor to drive it)


Back to top
Profile
Moderator
Moderator
avatar
Joined: Thu Jun 03, 2004 11:23 am
Posts: 3745
Location: Hamburg, Germany
Post    Posted on: Sun Aug 19, 2007 3:53 pm
What about using intelligent materials for rocket punmps? Materials reacting to current, voltage etc. as well as to heat, pressure etc.? What possibilities and chances to replace mechanical parts and components by them?



Dipl.-Volkswirt (bdvb) Augustin (Political Economist)


Back to top
Profile
Space Station Commander
Space Station Commander
User avatar
Joined: Wed Dec 08, 2004 12:55 pm
Posts: 506
Location: Germany
Post    Posted on: Sun Aug 19, 2007 10:28 pm
In case I haven't made a mistake...

Shaft Power= 1/(Pump Efficiency) * Density * Flow Capacity * Differential Head

Pump Efficiency depends on the type of the pump, a radial pump has usually 50-80% efficiency, an axial pump 75-90%.

Differential Head = Discharge Head – Available Suction Head
Discharge Head = Discharge Pressure / (Liquid Density * Gravity)
Available Suction Head = Suction Head + Initial Tank Level Above Inlet
Suction Head = Tank Pressure / (Liquid Density * Gravity)

of course everything in SI units! ;)

_________________
"The hardest hurdle to space isn't the technicalities and money. But rather, the courage and the will to do it." - Burt Rutan.


Back to top
Profile WWW
Spaceflight Trainee
Spaceflight Trainee
avatar
Joined: Tue Jul 04, 2006 10:12 am
Posts: 34
Location: Vienna, Austria
Post    Posted on: Mon Aug 20, 2007 11:43 am
Quote:
Shaft Power= 1/(Pump Efficiency) * Density * Flow Capacity * Differential Head


I really find "Differential Head" quite non SI. We should simply use pressure differential. Then there is no need for density. This makes the formula for a given pressure rather than head, which depends on fluid density.

Example the SSME need ~200bar or 20MPa pressure. So the power for 1 liter a second is (Assume all tank pressure is "used" in pipe losses etc and we leave out the preburner pressures and cooling passage losses, which is dam high):
(1/eta) * Pressure * flow rate
(1/0.8 ) *20 000 000 * 0.001=25kW

I don't need to care weather this is LOX or Liquid Hydrogen. Also for a given pressure, the pump power scales with volume rate not density, which really hurts hydrogen.

There is no simple way to estimate weight without getting specific with the design. Like how many boost pumps and inducer configuration, where you get the power from is *critical* to weight estimates, and things like gear boxes also have to be considered. Tank pressure also can have a effect. ie more tank pressure, less inducers but more tank weight.


Back to top
Profile WWW
Spaceflight Participant
Spaceflight Participant
avatar
Joined: Mon Feb 26, 2007 7:28 am
Posts: 68
Post    Posted on: Mon Aug 20, 2007 1:07 pm
I think that free piston reciprocating pumps are really the best alternative to centrifugal pumps, especially for small/cost sensitive/lower pressure systems.

I'm currently building a small free piston pump which will feed propellants at ~550psi and at flow rates between 0 and 1Kg/sec (1Kg/sec is the upper limit of my valves and around that point efficiency would drop like a stone due to back flow and pressure drop in the check valves). It can feed from tanks with down to 0 psi ullage (gauge) as long as the pressure drop across the intake check valves at whatever flow rate it is operating at isn't high enough to cause cavitation, of course at any decent mass flow you'd want your tanks at around 25-40psi. My prototype is running off 100psi compressed air for simplicity however you could easily run it off hot gas from an expander/dual combustion/gas generator cycle. I plan to switch to decomposed hydrogen peroxide as the working fluid for my second generation of pump.

Such a design of pump can scale down further than centrifugal designs and can scale up to moderate sizes (probably even launch vehicle sizes) and I think could deliver close to the same performance as a standard turbopump at a fraction of the cost, development and complexity. Also my prototype design will double as a flow rate meter and will be able to tell in real time how much propellant has been pumped to within ~15mL, all intrinsic to the design (no extra weight or parts).


Back to top
Profile
Spaceflight Trainee
Spaceflight Trainee
avatar
Joined: Tue Jul 04, 2006 10:12 am
Posts: 34
Location: Vienna, Austria
Post    Posted on: Mon Aug 20, 2007 1:57 pm
I think the group that worked on the mocking bird did the analysis for piston vers centrifugal pumps. The results were that they are lighter for small engines and the break even is at about 5-10kg per second for dense propellants IIRC. They are also easer to throttle.

Centrifugal/turbo pumps will be very hard to beat at the higher flow rates. Remember with a design margin they can be very reliable (aka RD10 )


Back to top
Profile WWW
Spaceflight Trainee
Spaceflight Trainee
avatar
Joined: Sat Oct 28, 2006 4:53 pm
Posts: 36
Post    Posted on: Mon Aug 20, 2007 5:20 pm
Ekkehard Augustin wrote:
What about using intelligent materials for rocket punmps? Materials reacting to current, voltage etc. as well as to heat, pressure etc.? What possibilities and chances to replace mechanical parts and components by them?



Dipl.-Volkswirt (bdvb) Augustin (Political Economist)


Or how about ferrofluid rocket fuel, possible pump with no moving parts.
(YouTube video of the stuff :) http://www.youtube.com/watch?v=fpI4EiGACo8)


Back to top
Profile
Moderator
Moderator
avatar
Joined: Thu Jun 03, 2004 11:23 am
Posts: 3745
Location: Hamburg, Germany
Post    Posted on: Tue Aug 21, 2007 6:31 am
Sounds interesting...

Oh, I seem to have forgot the aspect I had in mind mainly - perhaps the pumps could be lighter suchways.



Dipl.-Volkswirt (bdvb) Augustin (Political Economist)


Back to top
Profile
Spaceflight Participant
Spaceflight Participant
avatar
Joined: Mon Feb 26, 2007 7:28 am
Posts: 68
Post    Posted on: Wed Aug 22, 2007 12:53 pm
delt0r wrote:
I think the group that worked on the mocking bird did the analysis for piston vers centrifugal pumps. The results were that they are lighter for small engines and the break even is at about 5-10kg per second for dense propellants IIRC. They are also easer to throttle.

Centrifugal/turbo pumps will be very hard to beat at the higher flow rates. Remember with a design margin they can be very reliable (aka RD10 )


Would you have a link to this? The pump I'm making will be part of a thesis and that sounds like some interesting research/reference material.


Back to top
Profile
Spaceflight Trainee
Spaceflight Trainee
avatar
Joined: Tue Jul 04, 2006 10:12 am
Posts: 34
Location: Vienna, Austria
Post    Posted on: Thu Aug 23, 2007 9:10 am
Jordin Kare posted a bit about it on sci.space.tech (The group seems dead now) and still posts on the other space forums. It was a small, 1500kg SSTO design which as i understand never got to bending metal.

Google archive:

http://groups.google.at/group/sci.space.tech/search?group=sci.space.tech&q=Mockingbird&qt_g=Search+this+group

Note there are a few links to piston pumps as well. A university built one for H2N2 monoprop engine for mars ascent a while back too. All in All for small rockets they seem pretty good.

I was personally working on a all rotary positive displacement pump back in NZ. Using a scroll expander. You can remove the vibrations from pistons easer that way. But I really needed a CNC mill set which i didn't get around too.

If i get back to NZ i will start up again with a mockingbird class type experimental rocket. Not SSTO or anything.. but 200Km altitude should be fun (with guidance and a RCS).


Back to top
Profile WWW
Spaceflight Participant
Spaceflight Participant
avatar
Joined: Mon Feb 26, 2007 7:28 am
Posts: 68
Post    Posted on: Thu Aug 23, 2007 9:39 am
Sounds like a very interesting project. What interests me most about the piston pump over a rotary style pump is the simplicity in construction. You can use soft seals even for decomposed working fluid systems so you don't need any precision ground metal on metal seals.

You said the people working on mockingbird found that piston pumps broke even up until around 20kN (based on their technology which can vary significantly), wouldn't that be enough for a decently sized module in a modular launch vehicle design? Of course the primary aim of a modular rocket is to be as cheap and simple to build as possible, but a piston pump is technologically easier to build than an internal combustion engine.


Back to top
Profile
Spaceflight Trainee
Spaceflight Trainee
avatar
Joined: Tue Jul 04, 2006 10:12 am
Posts: 34
Location: Vienna, Austria
Post    Posted on: Thu Aug 23, 2007 10:45 am
found the source:
http://groups.google.at/group/sci.space.tech/msg/4f886cf1cece5cde

This is where i got the "data" from. The whole thread is probably worth a read.

Quote:
but a piston pump is technologically easier to build than an internal combustion engine.

Well it depends on what drives the pistons. If its hot gas there are problems with seals and high temperatures. Also they lack thermodynamic efficiency from lots of "free" expansion. There also need to be values, either check values or otherwise. They would be in the same ball park as a normal piston engine. Note that lots of folk make there own IC engines for fun, so its probably "easier" than a turbo pump. That would need CFD simulations and CNC mills (multiaxis) and the very least probably. But then again there are hobby built jet engines.....


Back to top
Profile WWW
Spaceflight Participant
Spaceflight Participant
avatar
Joined: Mon Feb 26, 2007 7:28 am
Posts: 68
Post    Posted on: Thu Aug 23, 2007 11:04 am
By free expansion do you mean when you dump the high pressure gas used to drive the pump to atmosphere? Because (at least in my design which is a little different to the LLNL design) if you use hot gas in the pump with an area differential between the gas and fluid pistons you can vent the driving gas either directly into the combustion chamber or into a secondary small combustion chamber to finish burning. As I understand the design I've seen in a few reports by LLNL, they can only recycle their pump gases to a chamber with around the same or lower pressure as their feed tanks.

This also means you could run on an expander or other minimal loss cycle. Of course the lower pressure differential across the pump means you either have to make bulkier pumps or cycle one the same size as an atmospheric dump pump many times faster. This also means the percentage of your propellants that go to powering the pump increases, however as long as you're using them afterwards it shouldn't be a problem until you try to work truly epic output pressures from the pump.

If you take for example an expander cycle using a piston pump, there will be a limit in output pressure where all of your pumped propellants are gasified and used to power the pump and you can't go beyond that point without further heating the propellants (and thus the chamber and pump). But I think in most cases you'd never need output pressures that high.


Back to top
Profile
Spaceflight Trainee
Spaceflight Trainee
avatar
Joined: Tue Jul 04, 2006 10:12 am
Posts: 34
Location: Vienna, Austria
Post    Posted on: Thu Aug 23, 2007 11:36 am
If my piston pump needs to output then with a piston pump the drive gas also must have *constant* pressure because they are mechinacly linked. Then say i use 1MPa gas in the pump, at the end of the pump stroke I need to vent the 1MPa gas so i can do a return stroke. If i vent to 0.5MPa chamber I need to push the piston back against that pressure to suck more propellants into the pump. Generally venting it to ambient is the best idea. But then you have a volume of gas at 1MPa thats vented to about 100KPa without doing any useful work. Note that venting to the combustion chamber still gives you the same "free expansion". The energy shows up as heat in the gas, and not useful heat either.


Back to top
Profile WWW
Spaceflight Participant
Spaceflight Participant
avatar
Joined: Mon Feb 26, 2007 7:28 am
Posts: 68
Post    Posted on: Thu Aug 23, 2007 12:16 pm
In this case I'd be talking about decomposed hydrogen peroxide as the pump drive gas, which contains free oxygen and pumping that into the combustion chamber would add to the overall efficiency of the system. Another example would be powering the pump by boiling a cryogenic fuel like methane in a heat exchanger, it would help your efficiency to funnel the hot gas into the combustion chamber to be burnt instead of dumping it overboard.

I haven't run the numbers or done the testing to see if the performance advantage in recycling the pump driving fluid outweighs the disadvantage of pumping against the chamber pressure of the rocket motor (that would be one element of my thesis) however I think the possibility is there.

I suspect that in the peroxide system there is no advantage in using a larger pump and more pressurant mass flow to recycle what is basically a lot of fairly cold steam and a little free oxygen into the combustion chamber. But if you're using some sort of expander cycle with pure fuel gas as the pressurant there may well be an advantage in recycling it.


Back to top
Profile
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 18 posts ] 
 

Who is online 

Users browsing this forum: No registered users and 20 guests


cron
© 2014 The International Space Fellowship, developed by Gabitasoft Interactive. All Rights Reserved.  Privacy Policy | Terms of Use