RCU Forums - View Single Post - What size engine - A poll
View Single Post
Old 12-12-2009 | 04:01 PM
  #32  
erazz
Senior Member
 
Joined: May 2003
Posts: 398
Likes: 0
Received 0 Likes on 0 Posts
From: NA, NJ
Default RE: What size engine - A poll


ORIGINAL: Turbotronic

A lot of reasearch has been done in the last 5 years which has turned conventional wisdom re compressor difussor design on its head. Some of it was known decades ago in the soviet block and has found its way into small missile engines. There is no reason why a reworked design with the technology we already use cannot produce 240N in a KJ size engine with close to half the fuel consumption. Its a question of know how. Add ceramics and youre talking 600N...
...
I've been tinkering with different diffuser designs and reading a lot of the published literature in the past couple of years.
I agree that there are some really nice innovations in impeller/diffuser designs.

The problem (for us) is the use of ceramics. The current trend for turbines is to use Silicone Nitride and hot press it under high pressure (Hot Isostatic Pressing). While the process itself is not terribly expensive when compared to vacuum casting Inconel the bare materials are still very expensive. Plus any FOD that enters the engine will instantly turn that expensive turbine to dust. All in all the technology has not yet matured to the point we can reliably use ceramic turbines. I'm tinkering with this on-and-off. I promise to tell if I hit on anything

As to the thrust figures - we're probably very close to the limit of what you can get in terms of thrust. 240N of thrust from a 70mm impeller will require almost 100% efficiency of the compressor stage. I doubt it's possible but I'd love to be proven wrong. Going to ceramics can increase thrust twofold. 600N from a Kj sized engine is probably not possible.


The truth is current compressors barely works as it is. CFD tools available to the public are not good enough to hint at this. Their flow predictions are just way wrong.
That's simply not true. I've seen results from OpenFOAM that rivals anything ANSYS can dish out. The real problem is setting up the problem correctly and analyzing the results correctly.