Air Turborocket for TAN

guest blogger john hare

The potential problems I see with developing the original TAN concept could be quite wrong. If I am, I owe several people apologies upon proof. Whether I am or not though, the concept itself is obviously valuable. I am not dismissing the people that disagree with my take, just throwing more concepts out using the basic principles of TAN and ATRs.

When looking at the force accounting on the TAN concept, you notice that much of the performance is from the reduction in over expansion losses. Anything that introduces more gas in the over expanded nozzle should see similar improvements. Whatever method is used to introduce this gas will have and Isp and thrust/weight improvement when all of the net forces are accounted for. So if the original TAN ‘booster section would have had an actual Isp of 150 with a T/W of 200(just guessing, substitute your own if you like), net Isp was 260 with total T/W of well over 300. An ATR in place of the TAN units should improve from an Isp of 1,000 and T/W of 25 to an effective net Isp of well over 1,500 and an effective  T/W approaching 40.

An expander deflector nozzle is very short and uses a center body to control the flow. At high back pressures it is supposed to direct all the exhaust against the nozzle walls with the center body creating an area in the center at ambient pressure. At low back pressures, the center fills with the expanding exhaust to increase the effective expansion ratio. I propose placing an ATR in place of the center body of the nozzle. This should be a simpler arrangement than the original TAN locations even with a pure rocket system as it allows more volume and residence time for the TAN propellants .

The gas generator cycle is easier on the equipment than the staged combustion cycles. It is not necessary to pump fluids to thousands of psi higher than the combustion chamber pressure to have enough pressure drop to drive the turbines. The down side of the gas generator cycle is that it doesn’t get very good thrust from the turbine exhaust gas which hurts Isp compared to the staged combustion. Since the gas generator exhaust doesn’t give very effective thrust compared to the main exhaust, developers must try for the most efficient pump system possible to minimize the turbine drive gas requirements.

 Suppose you instead go for a possibly less efficient system that is cheaper, and use the turbine exhaust from the gas generator to drive an ATR engine. Then the gas generator exhaust products become part of a much higher Isp engine instead of a liability. With the fuel rich turbine exhaust driving the ATR turbines and burning with air in the afterburner, it becomes desirable to use a much higher chamber pressure on the rocket engines even though it increases the turbine drive gas requirements. If 10% of the total propellant usage goes to the pump drives, that is OK because that 10% is operating a system with an Isp three to five times that of the primary rocket chambers.

TAN atr

Using multiple rocket chambers would seem to be easier than the normal expander deflector nozzle arrangement. It allows incremental development and throttling for either attitude control or acceleration limits. As individual chambers are shut down later in the flight, net expansion ratio improves again.

Tip turbines to drive the ATR compressor as used in the Japanese ATREX are lighter and more compact that the standard arrangements, and makes the plumbing much easier in this application. The downside is that they are not as highly developed as conventional turbines, and may be less efficient.

At launch, all rocket engines and the ATR are at maximum throttle with the ATR fulfilling the TAN function. When the ATR is shut down at altitude, inlet doors must be closed to hold the pressure of the rocket exhaust in. The gas generator exhaust then keeps the ATR area cool with no requirement for additional regenerative plumbing. If possible, LOX can be injected into the ATR afterburner area to get more performance from the gas generator exhaust. In landing mode, the ATR is the only engine operating with the nozzle operation in dual bell mode to assure clean flow separation from the main nozzle wall.

It seems possible that the rocket engines would be operated at 3,000 psi to get an Isp of over 300 at sea level. The ATR would use the gas generator exhaust to get an Isp of well over 1,000, with net being perhaps 1,500. My guess is system net Isp of 400 or so from the ground. At ATR shut down, the gas generator exhaust again becomes a liability, but if it can be after-burned with LOX, it seems possible to get a vacuum Isp of 250-300 from the gas generator products, with main chamber Isp of 350 or more. My guess is net Isp of over 325 vacuum. When in landing mode, the ATR Isp will likely drop under 1,000 with poor inlet possibilities.

With gas generator engines at high pressure, T/W of 125 is possible with high development effort with T/W of 100 attainable by Newspace under these conditions. The ATR, optimised for T/W more than Isp, and sharing the expansion nozzle with the rockets, a T/W of 35 seems reasonable. A system T/W of around 60 seems reasonable.

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.

7 Responses to Air Turborocket for TAN

  1. gravityloss says:

    So, another way to think this would be a shaped afterburner? In some sense an ATR is a motorjet. (Before WW2 they had to use piston engines to run the compressor, so they sucked back then…)

    Jet engines are air cooled, which makes them pretty simple in that plumbing regard. I don’t know how turborockets etc would handle, I don’t know if any have even flown.

    Yet another way to think is to think it as an aerospike engine where the spike gas is powered by a jet engine.

  2. jsuros says:

    It’s nice to see the return of the notch-bell nozzle! The whole design seems attractive from a development cost point of view, and I like the altitude compensation parts of it. Loss of performance when you hit vacuum makes me think of this as a first stage engine, though. if you only used it up to ATR maximum altitude you could forego the engineering to seal the compressor.

    What application did you have in mind for this engine?

  3. johnhare john hare says:

    It would be similar to a shaped afterburner. According to John Bossard, no turborockets have ever flown. Operationally it more resembles an expander deflector with a jet powered spike, though what name it goes under is not that important IMO.

    While I had read of the motorjet before, I had never noticed that they put the burners outside the chamber. I wonder what they could have done with proper afterburners.

    It was not supposed to be a nothbell concept, rather an expander deflector varient. My financial situation somewhat improved from when I posted about the notchbell, I have had some thoughts of utubing a video of that and a few other concepts after I get this @$&^ pump to produce proper results.

    The loss of performance in vacuum should be a bit less than on most gas generator engines with the afterburning volume of the ATR to work with. If I were specing this for dinospace, it would be about halfway between staged combustion and gas generator performance in vacuum, with sea level considerably higher. T/W would be 25% higher than this article as well. Of course I would probably spec hydrogen instead of kerosene for another 100 Isp points also. I’m quite certain that the RS68 vac performance could be beat.

    I picture this as an SSTO or parallel stage RLV engine. As a pure booster, I would think in terms of massive pre-cooling injection for the ATR on the pad to maximize sea level T/W.

  4. Tim says:

    Could this be simplified by speccing a tapping cycle instead of a gas generator cycle? You would only need one combustion chamber -the main one(s)- and; you could run some of the tapped gas straight to the ATR turbine instead of through the pump turbine, to decouple the pump flow from the compressor flow (although you could probably do this with a dedicated combustor too).

    Given that the fuel source is emerging from the tip turbines in a ring around the air flow, do you think there would be problems in achieving adequate mixing, or are those arrows below the compressor on the sketch supposed to represent flame holders or some such?

    Sorry to go OT, but I’ve been thinking a bit about the notchbell concept and I noticed that Eric Collins said that if he could find time he’d run a simulation; do you know if he ever did?

    P.S I know I said I’d send you an idea related to the ejector, but i now have some serious doubts about the concept, so I’m holding off until I figure some things out.

  5. johnhare john hare says:

    Using the tapoff cycle seems a little chancy to me, but it would provide a lot more available power down low where you need it. I was suggesting the gas generator cycle to make a virtue of necessity. The tap off does seem to offer some advantages of cooling is not an issue.

    The ATREX experiments seemed to indicate that mixing from tip turbines was not a problem. They even eliminated the flameholders as not necessary.

    I never wrote up enough information for him to work with on the notchbell. When I’ve had the time and inclination to blog, I’ve been trying to throw out new ideas rather than push older ones.

    I have been wondering if people think I should flesh out some of the earlier concepts using both the commenter input and some more serious numbers. Considering my reaction to some, not all, people that continously try to push one or two concepts regardless of reception, I’m not sure that would be a good idea. OTOH, some of the ideas could be quite useful and might benefit from more discussion. The notchbell might be one.

    P.S. If I didn’t have serious doubts about many of my ideas working, I would be GM.

  6. Would this concept be best for VTHL, HOTOL, or VTVL?

    Perhaps it would work for all three?

  7. johnhare john hare says:

    While I was thinking of VTVL, it should work for the others as well.

Leave a Reply

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