Yesterday Kristian von Bengtson of Copenhagen Suborbitals answered questions about the endeavour on Reddit. I took the opportunity to ask a bunch of questions about their custom built TM65 rocket engine.
I fired an email off at KvB’s invitation and when I woke up this morning I had a fantastic reply from Peter Madsen, Copenhagen Suborbitals’ propulsion specialist responsible for the genesis of the TM65 engine, answering all of my questions in detail.
[box_dark]Astronomy Aggregator – How exactly did you guys come up with the specs for the TM65? Rocket engine design isn’t very common in engineering courses. Is the TM65 more soviet or american engine inspired?[/box_dark]
Peter Madsen – The TM65 and the new TM65II A and B are very traditional bi propellant rocket engine designs. The engine is simple, cheap in construction and contains no exotic materials and requires no exotic fabrication techniques. The engine is heavily inspired by the North American (later Rocketdyne) A7 rocket engine used on the Redstone rocket. It represents roughly A4/V2 technology matured with the most stupid details of the V2 engine fixed. E.g. the injector of the A7 is far better and far simpler than the complex 15 burner cup design of the V2 engine, using a perforated plate instead of the cups.
[box_dark]AA – What kind of cycle does the TM65 operate on? Staged/expander/gas generator cycle? Is there a turbopump involved or do you simply use the pressure of the propellants to force them into the combustion chamber?[/box_dark]
PM – The TM65IIB is a turbine feed gas generator cycle engine. The first version TM65 was tested with pressure feed test stand tanks to separate pump trouble from engine / combustion trouble.
However the ultimate incarnation of the engine uses a turbine pump powered by decomposing Hydrogen Peroxide 80 % into steam and oxygen.
[box_dark]AA – Just how much of the TM65 was produced in-house and how much of it was bought off the shelf? Especially if it operated on a cycle needing a turbopump (those things are expensive!)[/box_dark]
PM – The turbines, combustion chamber, gas generator, all are inhouse designs based on the A7. The pump houses and impellers are fire pumps upgraded for higher rpm and higher head with LOX lubricated bearing and teflon seals.
The turbopump is again very much the Redstone A7’s.
[box_dark]AA – Have you considered using an aerospike instead of a bell nozzle exhaust?[/box_dark]
PM – No, we generally use very traditional designs – and are happy with high enough performance. The engines we build excel in low cost, and moderate performance, not novel design. After all to succeed we must build a functional bi fuel turbo pump feed rocket engine at a minute fraction of the normal cost. This design goal is not reached by looking for hot new alloys or unproven design concepts.
[box_dark]AA – Can you say conclusively that the failure of the TM65 on Dec 30th was primarily down to faults in geometry or material limitations? Would a more temperature resistant alloy have avoided the failure?[/box_dark]
PM – Yes we can. The engine nozzle skirt failed due the a structural problem – when it was still cold during the engine start up prestage. The test program aimed to look for week spots in the design by gradually increasing engine load during a series of static tests. The failure is quite easily fixed, and the failure was actually talked about in blog a week before it happened – we knew we had an interesting area – that was to be improved in the next design.
The plan is to test fire the improved version of the engine, TM65IIA, and fly it in a pressure feed cycle on the 2X rocket.
[box_dark]AA – Over the course of CS you’ve moved from hybrid rockets to liquid rockets. Do you plan on returning to hybrids at some point if the liquid rockets dont pan out? Or is there no way to get enough impulse to get to orbit using hybrid rockets at the scale of the TDS2 capsule? [/box_dark]
PM – Its not about performance. Both hybrids, solids and liquids can deliver the flight we are after. The reason for the shift from hybrid to liquid was the discovery of complex combustion induced oscillations in our LOX hybrids. LOX is a difficult oxidizer in a hybrid – N20 or H202 are far easier to make work. The reason is that a hybrid oxidizer must ideally be “energetic” that is able to self decompose exothermally. H202 and N20 does this – our small HATV hybrid ( 220 mm diameter ) works excellently and is very reliable on N20.
However working with a energetic oxidizer in very large amounts can be a problem – check for Scaled Composites explosion on google. So we want to use LOX for our big rockets and that means it has to be bi propellant liquid. [box_dark]AA – Where are you headed with the new engine design? Simply bigger is better? Taking cues from more advanced commercial/historical engines? [/box_dark]
PM – The last. Except for “advanced” We are not building a J2 or SSME inspired engine next, but rather a TM250 ( 250 kN thrust AA note: scaled up TM65, probably based on the TM65-IIB ) again based in the North American A7.
[box_dark]AA – Finally, do you have any recommended books for rocket engine design? I’m a mechanical engineering student looking for a final year project and a test bed rocket engine seems like a fantastic idea.[/box_dark]
PM – G.P. Sutton: Rocket Propulsion Elements, is an excellent text book to start with. If you can get hold of the 2nd ed ( 1956 ) and 3rd ed ( 1963 ) you are lucky – the latest 7th ed ( 2010 ) is all to modern for a rocket start up.
I’d like to thank Peter Madsen – for not only taking the time to comprehensively answer my questions, but also for doing so when he did; I received the email at about 1am my time, so about 2 o’clock in the morning in Denmark! – and Kristian von Bengtson for doing the Q&A session and continuing to provide an inspiration for everyone through the work they do at Copenhagen Suborbitals.
You can read more about Copenhagen Suborbital’s progress either here on Astronomy Aggregator, the Copenhagen Suborbitals website, or Kristian von Bengtson’s excellent blog series on Wired.com.
Image Credit: Thomas Pedersen / Copenhagen Suborbitals