Rollerthroat Pump

guest blogger john hare

[Saturday afternoon post change. Argh. While replying to Marcus, I realized that this post could be considered Monday Morning Quarterbacking at its’ worst considering the loss just experienced by Armadillos’ team. I respect that team and consider the members I have met as friends. This post is bad timing that I put out just a few hours before reading about the burn through. This rollerthroat idea is not ready for use or even serious consideration by any active rocket company. It is meant as a thought exercise for something that might be useful in the future.

My condolences to the whole Armadillo team. My apologies to anyone that might be offended by my bad taste in posting this on the same day. Congratulations to Armadillo on winning level one.]

Two of the problems facing rocket developers are protecting the hardware from excess heat and persuading the propellants to enter the combustion chamber. The throat and its’ approaches are about the toughest areas to keep cool. Supplying power to a fuel pump is another major issue that is frequently dismissed as too difficult in favor of a pressure fed design. There might be a way to make both tasks easier.

The throat of a rocket experiences thousands of degree gasses at high pressure moving at sonic speed just small fractions of an inch away on the other side of a boundary layer. Extreme care must be taken in coolant flow through this area to keep everything from melting. Frequently a serious amount of fuel is used as film cooling to assist the cooling channels in protecting the throat and chamber walls.

I suggest mounting rollers at the throat normal to the direction of flow and let the fast gasses spin them so that no section of the roller experiences the maximum thermal load for more than a  millisecond or so at a time during each revolution. A relatively small amount of fuel can flow through cooling channels in each roller to keep them cool, pump the fuel to a higher pressure so that it enters the chamber at substantially higher than tank pressure, and be a reverse flow film cooling blanket for the regular chamber walls.

Though there is a tremendous amount of power in the throat, it is not clear just how much of it could be applied to pumping fuel. 10% of the total fuel through the rollers would keep them cool through regenerative cooling in the impeller micro passages and heavy film cooling during the 50% of the time the roller is in the cooling cycle. 10% of the fuel pumped to much higher than tank pressure is not enough to be seriously beneficial. If there is enough power across the rollers  though, running a shaft to a serious fuel pump makes sense. Depending on power actually available, the oxidizer might also have pump power available with a  roller turbine. While making the surface of the roller rough would generate more power, it would also heat the roller more and restrict throat flow.

The relatively small amount of fuel pumped through the rollers will emerge on the full 360 degree perimeter of each roller. From the nozzle roller seal to the vertical chamber wall, the fluid will accumulate in the half volute and be directed in a reverse flow up the chamber walls providing some film cooling before the fuel boils out into the chamber to be burned. Film cooling from the upper injectors covers the balance of the walls before boiling out into the chamber . Since none of the film cooling fuel  exits the throat against the walls, more fuel than normal can be used for the purpose to lower the work required by the regenerative cooling passages.

25% of the time the fuel from the rotor will be directed into the throat with no oxygen to burn or down the nozzle with no real benefit for the engine. CFD and experimental work would have to be done to minimize the amount of fuel through the actual roller if Isp losses are excessive. The rollers may not need internal fuel flow for cooling. It would be simpler and cheaper to use a heavier rich flow from the injector perimeter to flow down the walls as film cooling to the bottom of the roller before returning to the main chamber with the roller, cooling it before being reinjected into the chamber. With the cyclic heating and thermal mass of the roller compared to normal throat cooling methods, fuel required would be minimal.

The rollers could be used for minor TVC by shifting in from one side to bias the flow down the expansion nozzle. This would not be something to depend on without strong experimental proof of effectiveness. The more useful function of roller movement would be as variable throat area control as in a pintle nozzle. During deep throttling, the rollers both move to the center to create a smaller throat with the same expansion nozzle increasing the expansion ratio. An expansion ratio of ten at liftoff could become a ratio of one hundred at altitude when the engine is throttled to 10%. This would allow a high expansion ratio engine in a short package that is relatively light. More practical would be a ratio of three or four. See Jons’ ‘Thrust Augmented Nozzles” post for another more professional method of variable throat area.

The rollers naturally fit a modular installation method with a simple chamber build. A throatless chamber could be built with the roller throat assembly installed and removed as needed for modification, maintenance, or inspection. The heavy film cooling could reduce or eliminate regenerative cooling requirements in some conditions.

The following two tabs change content below.
johnhare

johnhare

I do construction for a living and aerospace as an occasional hobby. I am an inventor and a bit of an entrepreneur. I've been self employed since the 1980s and working in concrete since the 1970s. When I grow up, I want to work with rockets and spacecraft. I did a stupid rocket trick a few decades back and decided not to try another hot fire without adult supervision. Haven't located much of that as we are all big kids when working with our passions.
johnhare

Latest posts by johnhare (see all)

This entry was posted in Space Policy, Uncategorized. Bookmark the permalink.

4 Responses to Rollerthroat Pump

  1. Marcus Zottl says:

    I see several problems with that idea (although I’m no expert and could be wrong):

    1st) How do you intend to keep the rollers in place? In a cylindrical chamber there would need to be many small rollers to achieve a uniform reverse filmcooling around the chamber. So how do you intend to mount them there?

    2nd) How do you get the fuel into lots of small, rotating parts inside the chamber? This adds a whole lot of complexity to the engine and possible failures! (If only one, or a few of the rollers fail, then cooling on that side is reduced or maybe gone at all which could lead to a burn through of the chamber.)

    3rd) Since the rollers have to operate inside a very fast expanding gas you probably end up with really, really high RPM values. How do you intend to cool the relatively small (not much heat capacity) bearings for the rollers?

  2. john hare says:

    Marcus,

    I see several problems with that idea (although I’m no expert and could be wrong):

    Just several problems, come on, you can’t find more than that? 🙂 I’m not an expert either.

    1st) How do you intend to keep the rollers in place? In a cylindrical chamber there would need to be many small rollers to achieve a uniform reverse filmcooling around the chamber. So how do you intend to mount them there?

    Four rollers for a square throat. The axles inside corner housings. Fan out the volute channels to spread the flow before vertical distribution.

    2nd) How do you get the fuel into lots of small, rotating parts inside the chamber? This adds a whole lot of complexity to the engine and possible failures! (If only one, or a few of the rollers fail, then cooling on that side is reduced or maybe gone at all which could lead to a burn through of the chamber.)

    The fuel would enter the rollers through the axle. With the axle as the fuel line, there is no housing as on a conventional impeller. It adds 50-100 parts and complexity. It adds failure modes. It also adds the possibility of running an inexpensive engine very hot without throat burn throughs.

    3rd) Since the rollers have to operate inside a very fast expanding gas you probably end up with really, really high RPM values. How do you intend to cool the relatively small (not much heat capacity) bearings for the rollers?

    Hydrostatic bearings are basically bushings with heavy lubricant flow. The full flow of kerosene lubricates the bearings as it enters the roller passages. I should have addressed this. Dry bearings at that rpm would last what fraction of a second? The really, really high RPM would mandate an axial flow pump if enough power were harvested to make it desirable.

    There would have to be some detailed mock up and partial equipment tests to answer the points you raised. This concept certainly shouldn’t be considered for an engine being designed now. Good points.

  3. Eric Collins says:

    While I do not claim to be an expert in rocket design either, there are a few concerns I would have about a design like this.

    – My first thought is that you would be adding considerable complexity to the chamber design for not alot of gain in terms of performance. Indeed, it seems as though you are explicitly trying to extract energy and momentum from the flow to pump the fuel. Surely this can only lead to a less than optimal use of your propellant resources.

    – Most rocket engines are built with very robust structural margins so that they will be able to withstand a wide range of thermal, acoustic, and mechanical loads. The exhaust itself is an unsteady, turbulent, chemically reacting flow which couples to the thermal, acoustic and mechanical modes in non-trivial ways. These interactions can (and often do) lead to rapid unscheduled disassembly of the engine. Adding moving parts to the thrust chamber is not something to be suggested lightly. For one thing, they would, more than likely, provide less structural integrity in a region where it is needed most. If one were to add these parts, they would have to be designed to withstand these forces, and I don’t really see how that can be accomplished for less trouble and expense than would be incurred by just using an external pump.

    – The rollers are going to adversely affect the flow through the throat above and beyond simply reducing the ISP. If you expect to use the rollers to extract energy/work from the exhaust gases, then I’d say there is a fair chance that such an interaction will trip up the boundary layer flow. In the best case scenario, this may just cause the boundary layer to increase in thickness a little yet still remain attached. However, it is probably more likely that these interactions would give rise to fully turbulent flow in the throat, and I can’t imagine that being a good thing.

    – The main purpose of the throat is to accelerate the flow to the speed of sound prior to it leaving the combustion chamber. Other than the aerospike engines, I’m not aware of any rocket which has ever been made to work reliably with anything other than a circular throat profile. The square profile necessitated by this design is going to give rise to uneven heating and compression of the flow along the length of the rollers. I’m not entirely certain that one could properly design a chamber with a square outlet that would generate the compression profile required to create a transonic flow in the throat.

  4. john hare says:

    Eric,

    While I do not claim to be an expert in rocket design either, there are a few concerns I would have about a design like this.

    – My first thought is that you would be adding considerable complexity to the chamber design for not alot of gain in terms of performance. Indeed, it seems as though you are explicitly trying to extract energy and momentum from the flow to pump the fuel. Surely this can only lead to a less than optimal use of your propellant resources.

    The question is how much energy is transferred to the throat walls by friction with the gasses in the sonic throat. If it leads to a very slight reduction in total friction, then the propellant resources might be slightly more optimal than standard.

    – Most rocket engines are built with very robust structural margins so that they will be able to withstand a wide range of thermal, acoustic, and mechanical loads. The exhaust itself is an unsteady, turbulent, chemically reacting flow which couples to the thermal, acoustic and mechanical modes in non-trivial ways. These interactions can (and often do) lead to rapid unscheduled disassembly of the engine. Adding moving parts to the thrust chamber is not something to be suggested lightly. For one thing, they would, more than likely, provide less structural integrity in a region where it is needed most. If one were to add these parts, they would have to be designed to withstand these forces, and I don’t really see how that can be accomplished for less trouble and expense than would be incurred by just using an external pump.

    I’m thinking there will be some moderate pumping action. The flow through the throat is relatively steady by rocket standards and the heat flux is extreme. The robust structural margins in rockets are by spaceflight standards. A margin of 2 is considered very high in this field. Burn throughs are still common from what I am told, and very much a concern in every design. These rollers would have more in common with the heat sink throats of solid rockets than the regenerative cooled liquids.

    – The rollers are going to adversely affect the flow through the throat above and beyond simply reducing the ISP. If you expect to use the rollers to extract energy/work from the exhaust gases, then I’d say there is a fair chance that such an interaction will trip up the boundary layer flow. In the best case scenario, this may just cause the boundary layer to increase in thickness a little yet still remain attached. However, it is probably more likely that these interactions would give rise to fully turbulent flow in the throat, and I can’t imagine that being a good thing.

    I see the rollers being in the same place as a standard throat hardware with the difference being that they are allowed to move with the flow. Energy extracted is the friction energy that had been heating the throat area. If the gasses of a particular rocket move through the throat at 2,000 ft/sec, and rollers in the same location are rolling at 1,000 ft/sec, then it seems to me that friction will be less. I am concerned about tripping the boundary layer with shock waves at the roller/nozzle seal.

    – The main purpose of the throat is to accelerate the flow to the speed of sound prior to it leaving the combustion chamber. Other than the aerospike engines, I’m not aware of any rocket which has ever been made to work reliably with anything other than a circular throat profile. The square profile necessitated by this design is going to give rise to uneven heating and compression of the flow along the length of the rollers. I’m not entirely certain that one could properly design a chamber with a square outlet that would generate the compression profile required to create a transonic flow in the throat.

    I’m going by one usenet exchange with a reliable source 🙂 years ago that square throats are not a problem. I’m not sure enough of my ground to argue the point though. It would be fun to set up a pool on where my ideas disconnect from reality.

Leave a Reply

Your email address will not be published. Required fields are marked *