A pipelauncher is based on simple concept- a pipe with one end capped. It's to provide a launching platform for rockets.
The size of this pipe is hundreds of feet long/tall and tens of feet in diameter. The capped end is up and the open end is
down, if this large pipe is lowered into the water, water will enter the open end, the air inside the pipe will be trapped
and will pressurize until the pressure is enough to stop the pipe from sinking. The rocket would go on top of the capped end.

Now, a pipelauncher could be a simple pipe, but pipes of this large size are not manufactured, the pieces of pipe this big
are, of course, are manufactured, but they need to fabricated and are normally fabricated on site. Recently I was reading
article on 4' diameter pipe was delivered to the job site in two halves and welded together lengthwise to form the pipe,
they did it this way because it was cheaper to make and cheaper to transport split in half than shipping whole pipes. I
thought this was interesting because these were high pressure pipe, requiring a weld as strong or stronger than the pipe
itself, they were doing in the field, and fast enough to make economical rather then having whole pipes shipped.
Anyway, I don't envision making a pipelauncher in the middle of the Pacific, instead a drydock with all the skilled labor
available and infrastructure to make fabrication rather easy.
Another thing I noticed recently on the internet was that X70 steel was commonly used and easily welded, and there many
technical papers on using X100 steel and welding this. So the impression I got was X100 or higher was state of the art
in big projects like pipelines, X80 was high quality standard use, and X70 was decades old technology.
Here's something on this:
"Oil and Gas Pipelines in 2003: An Overview":
...
"Petroleum Institute (API) grades as 5L-X65 or 5L-X70 (65,000 psi or 70,000 psi yield strength steel,respectively) were
generally the highest-strength materials used. Steel manufacturers have now produced steels of 100,000 psi and even 120,000psi
yield strength, known as X100 and X120, and these are being widely considered for future use."
http://66.102.7.104/search?q=cache:fBSFUY69wdoJ:www.world-petroleum.org/isc2004/File%2520018/112-119.pdf+X100+%2B+steel+%2B+psi&hl=en

In my other post about pipelaunchers I had calculated for 40,000 psi yield strength steel because I was unsure how easy it
was to weld these higher grade steels.
Also I believe I read somewhere that the new natural gas pipeline in Afganistan will be 56" and X100. But I'm not going to
count on X100 or X120, I think that X70 or X80 would be a reasonable grade to use for pipelauncher.

Thing that makes these pipelauncher work is big diameter, the ultimate pipelauncher would hundreds of feet in diameter
and thousands of feet tall, and these would massive structures costing tens to hundreds of millions of dollars.

The first time I proposed a pipelauncher I picked about 800' in length/height. The length of a pipelauncher is important if
one want to accelerate the rocket to a high speed. I had also thought that accelerating a rocket at 3 gees was reasonable-
but after some thought I decided that 3 gees might be too much load to place on a typical rocket which was fully loaded
with fuel.
A pipelauncher could be desighed to accelerate a rocket at any accelerate and could done at a constant smooth acceleration,
the problem is the rocket being able to withstand the additional load. So since I didn't know if a rocket would hold together
if it was made 4 times it's weight, I assumed that it's more reasonable that a rocket could safely withstand being twice it's
weight [1 gee]. Though a safer assumption would be to accelerate a rocket at the same acceleration that it's engines
provide when it's fully loaded with fuel [.5-.7 gees seems typical].
Numbers:
Distance = acceleration 1/2 times Time squared.
4 second acceleration at 29.4 meter/sec/sec requires 235.2 meters [just over 700'] and give a velocity of 117.6 meters/sec
[423 kpm/ 262 mph], Whereas 7 sec acceleration at 9.8 meters/sec [1 gee] requires 240 meters and gives velocity of
68.6 meters/sec [247 kph/ 153 mph]. Or say 10 seconds of acceleration at 9.8 meters/sec requires 490 meters and gives 98
meter/sec. So, if you want a pipelauncher to accelerate a rocket to high speeds you will need to use rockets that can
withstand high gees- such as say, 6 gees: 3 second acceleration at 58.8 meter/sec requires 264 meter and gives 176.4
meters/sec [635 kph/394 mph]. Or with one gee or less: very, very big pipelaunchers.

Instead a more modest proposal (with a bonus wild idea thrown in):
A pipelauncher which is 500' tall and about 40' in diameter.
Starting from bottom:

16 4' diameter steel pipes 1/4" in wall thickness separated by spacer which are 2' wide and 6' wide, making 8 pairs of pipe
forming a circumference of 128' slightly larger than 40' diameter. The spacers are 1/4" steel plate. This section of
fabricated pipe is 200' long.
Below this is single walled pipe 1/4" plate steel which is 44' in diameter and 50' long.
Above this is 8 4' diameter pipes 1/4" in wall thickness separated by spacers 3/8th" plate steel 12' wide. This is 100'
long section of fabricated pipe. On top of this is a collar which tapers to the diameter of next section.
This next secion is a single walled pipe 36' in diameter with wall thickness of 3/8" which is 50' long with curved dome/cap.
In this section there will be a burner and plumbing that allows cryogenic liquids to be sprayed. This is the top of the
pipelauncher, everything below this area is pulled and everything above it is pushed. The above sections are essential
spacers, they are pressurize at a constant pressure and are are passive, whereas the sections below "breathes" and part of
pipelauncher's "engine".
The next section up is 36' in diameter with 1/4" plate steel walls, it is sealed and is filled with a few atmospheres of
Helium. It is 50' long and on top of it is another collar which expand out to 40' in diameter. The next section is a
cylindrical high pressure balloon filled with couple atmospheres of Helium. It is 50' long section

4' diameter 1/4" plate 100' length has 28.4658 cu ft of steel

16 pipes 200' has 910.9 cu ft of steel

Weighs about 220 tons

Spacers 64' by 200'
weighs 64.48 tons

Skirt: 44' diameter 50 long:
34.82 tons

Pipes in next section: 55.1 tons
Spacers: 72.54 tons

Collar: ?

36' diameter pipe 3/8" thick: 42.73 tons Dome: about 10 tons

36' diameter pipe 1/4" thick: 28.49 tons

Totaling 508.16 tons. Make a round number of 600 tons to include everything other than the rocket itself.

To fiqure out it's buoyancy. The 300' section 16 & 8 steel 4'diameter pipes and spacers will providing the lift, being 40'
in diameter each foot in length has 20sq times pi which about 1257 cubic ft. And since sea water has density of about 64 lb
per cu', each foot in length has about 80,448 lbs [40.224 tons] of displacement force.
So to float the 600 ton pipelauncher requires 14.9 feet of pipe length, or said differently the water inside the pipe needs
to be pushed 14.9' below the waterline and the pressure inside the pipe needed to do this is about 7 psig. To accelerate the
pipelauncher upward at 1 gee, requires twice this pressure. As the pipelaucher goes up it increases in total volume,
therefore more gas and/or hotter gas needs to be added [1257 additional cu' is added per foot it goes up]. Then adding a
rocket which say weighs about 600 ton, double the psi needed [requires about 30 psi to accelerate at 1 gee, this forces the
water inside the pipelauncher about 60' below the waterline].
The easiest way to add this air is use liquid air which has expansion ratio of around 1-700 [one cu' liquid expands into 700' of air at 1 atm].