Re: Multiple engines (was RE: Liquox)

At 09:16 PM 6/10/2004 -0700, you wrote:
>On Thu, 2004-06-10 at 19:47, John Carmack wrote:
>
> > Doing it with cooling passeges sounds pretty challenging to me, but we are
> > planning on fabricating a palte with 16 x 1" nozzles, which should be a
> > direct performance comparison with our current 1 x 4" nozzles, if I keep
> > the same expansion ratios.  I don't expect it to have a significant
> > performance loss, and we might even get some aeruospike expansion 
> effect at
> > high altitudes if we arrange the nozzles correctly.
>
>         You just cool the two plates -- making a flat-plate heat exchanger
>isn't terribly challenging, and use a material with a nice high heat
>conductivity to conduct heat away from the throats. If that's not good
>enough, just bore straight holes through from one side to the other and
>arrange the flow paths appropriately.

You would have to be a smaller than 1" throat nozzles (several inches tall) 
to heat sink all the way to the top and bottom, but 1/2" nozzles would 
probably work.  That would be a pretty heavy plate, but you might be able 
to live with aluminum.

>         On the performance side, I will be very curious to see your results.
>There are four types of losses that I see affecting an array of square
>nozzles much more than a single nozzle. One, the square nozzles aren't
>as efficient as round ones, due to different flow behavior in the
>corners. Two, since each nozzle is smaller, the boundary layers take a
>relatively bigger bite than they do in a larger single nozzle. Three,
>the oddly shaped combustion chamber probably will not be as efficient.
>Four, since each flow passage is smaller, manufacturing defects of the
>same absolute size have a larger effect.
>
>         -p

Square nozzles?  I was talking about getting 100 little nozzles run off on 
a CNC lathe.  They would have integral shoulders so they fit into holes cut 
with an abrasive waterjet in stainless plate, then you weld them in.  The 
catalyst / combustion chamber would be exactly the same as on our normal 
engines.

John Carmack