Intuition and Rocket Concepts

guest blogger john hare

One of the engineering teachers at the community college I briefly attended had an expression, “Engineering is an exact science, and intuition.”   In both my day job and rocket hobby, I take that to mean that if you have to be able to see a bit more than shows on the surface. I have lost considerable amounts of money in business by ignoring the reaction that something is wrong when everything seems to be going right.

In concrete work, an instinctive feel for potential problems is an asset. You look and everything is in place but doesn’t ‘feel’ right, so you look closer and find that someone forgot to nail a brace in one critical area that was hard to get to. One nail saves hundreds or even thousands of dollars. Unfortunately, some people don’t have that intuitive feel for the job and either waste thousands in labor preventing non-problems, or miss true ones costing more thousands. Worse is the supervisor that thinks he has the  instincts, but doesn’t, that multiplies problems wherever he goes.

The same general principle applies in rockets and spaceflight. The better intuitive feel an engineer  has for the problem he is working on, the less time he spends on blind alleys. When there is no intuitive feel for the problem, then billions must be spent investigating every possible problem and varient  in order to get anything worthwhile. And then sometimes it takes so long and costs so much that the end product is worthless.

The intuitive feel for the problem must be based on knowledge and backed up with research. A concept purely based on intuition by someone that has never studied the subject is almost certain to be wrong. I never paid any attention to solid rockets until the Aries series. My first impression was that it should be a quick program to slap an upper stage on a booster with a long track record. It still feels like it should be possible if the right people with the right incentives were working on the problem. It is safe to say that my intuition is wrong in this case. That is where the backed up by research comes in. When your intuition points you in a direction, you get hard numbers before spending hard money going in that direction.

In Selenian Boondocks I have done several posts on various concepts based more on intuitive design than serious research. I can justify that based on the fact that we are brainstorming here more than calling for a serious effort. Sometimes though, I do need to back up my shooting from the hip with something more solid. In the last post on TAN modules, I threw out an intuitive concept without seriously showing how I got there. Several people raised serious objections that I didn’t have time to research and justify last week.  I need to show how I reached the conclusions I did, so that we can start addressing the real concerns raised. I think this is a serious enough issue to warrant a separate post. While I am not putting really hard numbers together here, if a concept shows promise I’ll put a few together.

In comment 3,  Jusuros suggested interconnecting solid gas generators to keep the one nozzle to orbit concept intact. My first thought is that you probably want to get rid of the mass of the solid  casings as quickly as possible and it would be almost impossible anyway. Actually it is possible and might even be desirable to keep  the whole casing. Something the size and strength of an SRB casing would make a dandy orbital assembly garage. For interconnecting the gas generators, use a segmented solid with ablative seals between segments to spread the burn time. Use different burn profiles in the upper segments to throttle down.   This would take a few weeks to nail down some numbers if, and only if, someone doesn’t shoot it down with something obvious that Jsuros and I missed. At this point it is strictly an intuitive concept  that would need hard number verification in stages to avoid expensive pitfalls.

In comments 5 and 6, Tim and Habitat Hermit expressed concern over cooling the TAN modules with hot gasses on both sides. The main nozzle is already regenerative cooled by the core engine, so that shouldn’t be an issue. The modules have a lot of propellant flowing into the modules  compared to the surface involved. At worst, the middle module shell could have a structure with regenerative cooling channels on both the inner and outer walls. The exterior modules suggested by HH would leave a cleaner nozzle flow.   

In comment 6, Habitat Hermit questioned the fluid dynamics of venturis and supersonic flows. One assumption I made was that most people were more familiar with venturi effect and ejector operation than I was. The venturi effect is that subsonic flows speed up through a restriction and have lower pressure. The narrow passage, where the flow is fastest,  has lower pressure than the wider  flows both upstream and downstream that are running slower. This is the same effect used by many paint and insecticide sprayers and seems a straight forward application to getting more propellant in the chamber as in crazy 319. The lower chamber would have to be at half or less the pressure of the upper to keep the flow velocity required to make this work.  The supersonic flow injecting is something I lifted direct from the ejector ramjets that compress air by entraining it in the supersonic exhaust. In this case, we are pressurizing liquids instead of gasses, and vaporizing and mixing them in the process. The supersonic ejector in the TAN module is an ejector ramjet turned inside out. The supersonic gasses from the outside pressurize fluids from the inside instead of vice versa. Supersonic flows slow down through restrictions and increase in pressure, just the opposite of subsonic flows. That is why the injection point needs to be as far upstream as possible in these modules.

In several later comments, Roderick and Axel questioned why I wanted to do something different and more complicated than the original TAN. That is what got me on this post. I didn’t have good answers until I went back and reviewed how I reached my original conclusions. My intuitive feel for the TAN concept was that it was a great idea but was missing something in execution. I never bothered to check on how I got there until challenged by people I respect. Using the paper they both linked to, I can point out some of the things that got my attention.

On page 4, they say that augmentation is virtually unlimited. To me this is a red flag since an excess flow would result in a nozzle that was under expanded at some point. Also much of the flow would seem to exit subsonic if it was not properly burned and expanded, which could hurt Isp far more than the augmentation thrust gain would help. In the test firing picture and chart shown on page 5, the maximum  augmentation is 40%. The highest they have tried is 77% mentioned a bit lower down. The 260 seconds Isp is quite good. That much of the measured Isp is caused by the reduction in pe/pa losses suggests that the isp will drop considerably as augmentation increases or as back pressure decreases.

The thing that subconsciously bothered me the most, and that I missed on the first two rereading of the document, is hinted at in recommendations on page 9. They suggest a hot fire demonstration with multi-stage liquid injection. This strongly suggests that the demonstrations were done with gas injection. Liquid injection requires considerably more residence time than gas injection to complete the burn. If the injection must be gas-gas, then it will take a considerable bit of hardware to change phase on the propellants prior to injection. If they can make liquid injection work, then they have a real winner on their hands.

My intuitive conclusions on the TAN concept may be steering me wrong. After going back over the original, my current conclusion are similar to what I had before, only with a few more details.

TAN is a great and potentially valuable concept. It will be used in some form in some new engines, and maybe a few old ones.

As augmentation increases at sea level, Isp will drop radically as the nozzle pressure matches ambient. This is because much of the measured Isp  is by reducing over expansion losses.

As the vehicle gains altitude and back pressure reduces, Isp will drop for the same reasons as above. Careful attention  will have to be paid to the trade offs. Low Isp may be acceptable if it leads to even lower costs.

The injection process is questionable if it must be gas-gas. That could add a lot of hardware. Liquid liquid may have some serious performance issues.

A small subsonic combustion area would benefit the concept by giving more residence time for vaporizing and combustion as well as supplying a throat for initial gas acceleration to supersonic.

Better results might be available from multiple combustion chambers sharing a common expansion nozzle. The classic 7 cluster could shut down subsonic chambers in pairs to increase expansion ratio while throttling down. Expansion ratio could go 7 chamber er=20.    5 chamber er=28.   3 chamber er=47.   1 chamber er=140.

The following two tabs change content below.


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.

Latest posts by johnhare (see all)

This entry was posted in Uncategorized. Bookmark the permalink.

2 Responses to Intuition and Rocket Concepts

  1. jsuros says:


    Is it possible you are overthinking problems with TAN here? On the gas/gas injection question, for instance. In a TAN design, You have to route the augmenting fuel and oxidizer that you want to inject past the main combustion chamber and throat to some downstream location on the nozzle. Since you need that plumbing to pass over hot portions of the nozzle in any case, why not put the two in contact and transfer some energy to vaporise your propellants?

    My understanding was that in a TAN you inject the denser propellants into the nozzle and use the higher performance reaction in the core engine. So you don’t need a heat exchanger to vaporize H2 just before injection into the nozzle, or whatever you were getting at.

  2. johnhare john hare says:

    It is very possible that I am overthinking the problems. While I obviously don’t think so, data to the contrary could change my mind fast.

    It is the denser fuels injected into the nozzle I am refering to. The Lox and kerosine both need to be vaporized before TAN injection if I am right. You can’t route the TAN propellants through the cooling jacket of the main chamber because the primary chamber propellants must do that job after TAN shuts off. Changing cooling systems in mid flight would probably be a very bad idea.

    Vaporizing the kerosine before injection is going to be a real trick. The hydrogen uses whatever normal injection into the main chamber as an unaugmented engine, and it plays no part in the TAN injection.

Leave a Reply

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