SpaceX in 2016

http://www.fool.com/investing/genera...-its-game.aspx

Taking a key paragraph from the article:

Land rockets 70% of the time. For the rest of the year, Musk believes SpaceX
can land the Falcon 9 70% of the time.

I'd love to see that happen. It'll be a major shift in the paradigm if they
can do even 50%.

In article ,
says...

http://www.fool.com/investing/genera...-its-game.aspx

Taking a key paragraph from the article:

Land rockets 70% of the time. For the rest of the year, Musk believes SpaceX
can land the Falcon 9 70% of the time.

I'd love to see that happen. It'll be a major shift in the paradigm if they
can do even 50%.

They're also increasing production of engines, and etc. Couple that
with hardware leftover from last year (because Falcon 9 was grounded for
much of the year), and they should have enough hardware for an increased
flight rate even if the recovered stages are not used on a paying
customer's flight.

In 2016, I'd expect to see more inspections and ground tests of any
recovered cores. That may be followed by test flights which don't risk
a customer's payload.

Jeff
--
All opinions posted by me on Usenet News are mine, and mine alone.
These posts do not reflect the opinions of my family, friends,
employer, or any organization that I am a member of.

On Monday, February 8, 2016 at 5:47:58 AM UTC+13, Jeff Findley wrote:
In article ,
says...

http://www.fool.com/investing/genera...-its-game.aspx

Taking a key paragraph from the article:

Land rockets 70% of the time. For the rest of the year, Musk believes SpaceX
can land the Falcon 9 70% of the time.

I'd love to see that happen. It'll be a major shift in the paradigm if they
can do even 50%.

They're also increasing production of engines, and etc. Couple that
with hardware leftover from last year (because Falcon 9 was grounded for
much of the year), and they should have enough hardware for an increased
flight rate even if the recovered stages are not used on a paying
customer's flight.

In 2016, I'd expect to see more inspections and ground tests of any
recovered cores. That may be followed by test flights which don't risk
a customer's payload.

Jeff
--
All opinions posted by me on Usenet News are mine, and mine alone.
These posts do not reflect the opinions of my family, friends,
employer, or any organization that I am a member of.


The important detail is that they're using state-of-the-art techniques of

(1) computer modelling/simulation test,
(2) 3D Cad, virtual reality, and
(3) 3D Printing,

to improve reliability and reduce production and operating costs while allowing them to extend capabilities with far greater confidence and at far less cost than legacy producers.

* * *

From my experience, one major area of improvement that will radically change things is the production of large arrays of micro-scale engines made with automated production systems. This will achieve higher thrust to weight and lower cost per unit thrust while improving safety and reliability.

Because thrust scales with the square of the engine dimension whilst mass scales with the cube of engine dimension arrays of tiny engines have vastly higher thrust to weight than a single large engine.

Also, as engines become smaller, explosive energy scales with volume, whilst strength scales with area. This means vast arrays of tiny engines are vastly stronger and safer than a single large engine of equivalent thrust.

So, while large numbers of engines virtually guarantee high probability of failure of at least one engine per flight sycle, small robust engines mean an array will not be harmed by such failures. When arrays are built to 110% nominal value operators will always find the engine array producing at 100% of expected value over its useful life!

The assembly process is rather simple once you go through it a few times.

Consider this manufacturer of micro formed fully dense metal parts;

http://www.suron.com/electroforming/

And this product, the RL-10 engine;

http://alternatewars.com/BBOW/Space_...0A-3-1_DWG.gif

converted to a 3D CAD model - far more detail than this.

http://www.3dcadbrowser.com/download.aspx?3dmodel=14607

Cut into layers with software ...

http://www.123dapp.com/make

assembled and built along the lines described in this 11 year old document;

http://cap.ee.ic.ac.uk/~pdm97/powerm...53_Epstein.pdf

So an RL10 engine scaled down to an engine that has a 1 inch diameter and is 1.75 inches long. This engine has 16.7 lbs force of thrust and weighs only 0.56 ounces!

Made of precisely shaped layers of foil ultrasonically bonded to each other - a very fine engine is built easily by being very patient and careful.

Each layer consists of a 1 inch diameter circular build area. On a 22" x 24" sheet, with each circle hexagonally close packed to its neighbours, 636 circular areas are packed in the larger sheet area. Most of the centers of these circular areas are empty, and fit other key items in them as well, which are assembled when completed with the care of a watchmaker.

http://goo.gl/pL7JU2

Turbines, impellers, valves, actuators, are built inside the rocket bell!

A 1.75" tall engine made of 636 STL layers has a layer thickness of 0.00275" - so a single sheet, made at a cost of about $100 - has sufficient parts layers on it to build a completed micro engine!

So, doing this as a hobbyist, I have the sheets made for me, and assemble the foil parts manually (actually have watchmakers build them for me).

Under clean room conditions, each layer of foil is carefully removed from the carrier sheet onto which it is wet etched, and carefully placed in a jig to hold it precisely in place. Each foil has notches to fit to the jig, with the precision of a watch maker.

An ultrasonic weld head is then applied to the foil to attach it to the foil beneath. The foils are held with a magnetic clamp, underneath a plastic film so they don't move during weld. The work is then inspected to assure a secure weld after each layer.

Here is a test bar made of foils. Such a bar is made and tested to determine structural strength of the process.

http://fabrisonic.com/fabrication/wp...7/IMG-9189.jpg

Completed parts are fully dense metal as strong as a cast piece.

The assemble, weld, inspect process takes about 1.5 minutes per layer, and in 15.9 hours, I have an engine. I build two to three engines in a week depending.

Now, paying a person $1,000 per week, and allowing $500 per week for facilities and tooling, this is a cost of $800 per engine at the production rate of two engines. Dividing by the number of pounds of thrust generated by the engines, about $50 per pound of thrust. A person hour per pound of thrust is another way to think of it.

A 25,000 lbf engine array of 1,497 engines, costs using this method, $1.25 million and 25,000 person hours. About 30% of an RL-10 in terms of cost. With 2,000 hours per year per employee, 15 people are needed to complete one RL10 equivalent engine in a year including management. Sold at $1600 per engine, a million per year is made after tax. 30 engines a week must be sold to support an operation this size.

125 people are needed to build a J2 (or two Merlins) equivalent 250,000 lbs force using 14,970 engines in a year using this process. The cost is $12.5 million at these scales. This is about 2x the cost of the Merlin, 1/3 the cost of the J2 in today's dollars.

http://www.nasa.gov/centers/marshall..._Engine_fs.pdf

Of course building larger arrays tend to favour more sophisticated tooling for the production of engine arrays automatically which radically reduce cost and labour.

Its clear a serious production run involving more than a handful of engines would not use the same processes I use just playing around with the technology! lol. Still, even as a hobbyist/researcher, I'm creating thrust at far less cost, and equivalent specific impulse, to P&W at the RL10 scale! Though I'm not competitive at all at this scale with SpaceX. Selling engines to hobbyists and researchers for $2,000 each, provides a handy income without the need to find buyers for 30 engines a week!

For the spacecraft designer, the lift per unit area of propulsive surface is 2,775 lbs force per square foot. Far larger lift capacity than a wing! The engines mass 6.4 lbs per square foot! A little less than the structure that holds them! A propulsive skin less than 2 inches thick! Easily integrated into advanced airframes!

Directional control can use gimballing of the engine or array.

https://www.youtube.com/watch?v=i2gU4CVDV6Y

http://www.instructables.com/id/Ardu...cket-Guidance/

using off the shelf Stewart platforms used to mount an array to the airframe to control thrust, it is far more efficient given the engine's high thrust to weight, to put engines in groups of three - a triad - and vary thrust in each to get a controlled resultant thrust vector!

Here, each engine is oriented at most perpendicular to the other two. Like the 3 axes in a cartesian coordinate system - x,y,z. With right angles between the three engines, I add another engine r, oriented along the resultant vector, the sum of the three in the triad. The resultant is pointed normal to the propulsive surface and permits far higher propulsive efficiencies than possible with the triad assuming normal thrust is desired in the particular airframe design.

The four engines assembled in this way form a quad - that is capable of producing thrust tangential as well as normal to the surface. An array of quads built into a surface gives very precise control. The density of quads and the surface area determines the total thrust and performance.

With each engine firmly fixed in an unmoving structure, and the structure part of a space frame array, I have a very lightweight, very firm, highly reliable propulsive system that is very capable.

Jet engines are also possible.

https://www.youtube.com/watch?v=W6A4-AKICQU

Small arrays of engines, used in rocket belts with off-the-shelf electronics Arduino controllers and gyros...

Jet belts like the JB-9 and it cousins are possible

http://jetpackaviation.com/the-jumpjet/jb-9/

https://www.youtube.com/watch?v=x2sT9KoII_M

We're working now on VTOL and transition to high efficiency wing, with heads up 3D display in the helmet, with gesture recognition and interactive control gloves on virtual controls... very nice. A vast improvement on arching one's back and putting one's head down ... and necessary too for take off and landing in a totally independent vehicle.

With large arrays, of engines, the ability to control the mass flow rate through each engine independently across a surface using off the shelf electronics to paint thrust vectors across a surface, gives me very precise control. Furthermore, existing consumer control electronics that control tens of millions of pixels in a plasma UHDTV screen, is adapted directly to control the thrust painting process, updating thrust vectors at a rate of 120x per second. A four colour UHDTV controller used to extend the colour palette of TV screens, available from any variety of manufacturers is perfectly suited for this use.

13 tanks each 29 inches in diameter, carrying 155 lbs of LOX/LH2 propellant, with a small engine array, will put a person on orbit, and this is an interesting next step to this sort of technology.

Even so, some day, I envision that through automation production rates are radically improved and costs are radically reduced - by a factor of 100 I figure!

At this price 1 million engines costing $10 million producing 16.7 million pounds of thrust overall, covering 6,018 square feet of propulsive surface area massing only 38,500 lbs - an area a little larger than the wing area of a Boeing 747! 20x the lift!

Using only 25% of the area of an 87.5 ft diameter sphere covered with an engine array costing $40 million carries 7.7 million pounds of LOX/LH2 propellant, 154,000 lbs of engines, and 235,000 lbs of structure! A SSTO like this can lift 696,700 lbs into LEO. LOX/LH2 made from water using electrolysis with electricity costing $0.28 per kWh, costs $5.9 million for a fill up.. With 10,000 flight cycles possible with the system, and $100 million construction cost - we're costing around $10 per pound to orbit.

The payload in LEO is a smaller version of the SSTO. A 39.3 ft diameter sphere, capable of 9.2 km/sec delta vee. So, carried to orbit this sphere is capable of putting 62,900 lbs on the moon and bringing it back to Earth without refueling. Sending a similarly large payload to Mars and bringing it back to Earth.

Of course, I will improve performance with some attention paid to shape.

https://i.ytimg.com/vi/9NnhEYQJMmk/maxresdefault.jpg

http://www.astrosurf.com/luxorion/Do...ft-project.jpg

http://www.laesieworks.com/ifo/lib/t...htcraft-03.jpg

While this craft uses a laser heated air for propulsion, I have looked at the airframe and the data available for it, for a hydrogen/oxygen fuelled rocket array.

Using hydrogen fuelled ram-rocket from Mach 0.8 to Mach 5.0 increases payload to orbit for an 88 ft diameter ship carrying 44 ft diameter stage - to 1,000,000 lbs and 100,000 lbs respectively. The smaller stage is released from the tail of the larger stage once on orbit, and the payload exits the second stage from the base of the stage.

A TWO STAGE TO ORBIT (TSTO) system, with exceptional T/W and very low mass structure, a TSTO system puts up 731,000 lbs - a significant increase, but at a significant increase in complexity! Of course, if we move from hydrogen to methane, Isp is reduced, and the advantages of TSTO make it recommended. Ditto for Two Stage deep space stage. For lunar operations a Lunar Free Return Trajectory, and for Mars operations, a Mars Free Return (2 year orbit) - increases payloads to match or exceed capacities of single stage varieties using LOX/LH2.

A million pounds to orbit carrying 20 kW per pound thin film solar power beaming system, puts up 20 GW per launch. That changes the nature of the deep space stage to use laser sustained propulsion, and even raises the potential of modifying the aforementioned LOX/LH2 SSTO to use laser propulsion using the same shape.

A 20 GW laser beam propelling exhaust from a laser heated engine at 9.2 km/sec (the most efficient for orbital operations) produce at most 975,380 lbs force of thrust! So, we need 8 power satellites operating in parallel, to produce sufficient thrust to replace/augment the LOX/LH2 engine array. Using this airframe, the 7.7 million lift off mass can put 2.37 million pounds into LEO and 730,000 lbs on the lunar surface or Mars surface using lasers in the deep space stage as well! Basically, the two stage rocket is replaced by the booster stage, using laser energy to energise propellant.

The secondary stage form a fleet of smaller craft, once modified, or boost into the outer solar system.

A solar power satellite at L1 and L2 beaming energy to a moon base, beams energy to bring landing spacecraft to rest on the Moon.

The use of photonic thrusters further improves performance, in zero gee. At these power levels, we're better off than ion rocket engines.