Yes, I do recall the linear valve, I was there while the prototype threads were everywhere
. It was an interesting design, but to get a decent flow the T would have to be massive. The hole had to not reduce the integrity of the pipe (5/8 could only really safely go to 7/16" on 1/2" PVC (OD 5/8" = 10/16"). In addition, it requires the purchase of parts from McMaster-Carr, which proves difficult as I am from Canada and they don't like shipping there to new customers.
In addition, Schedule 40 PVC does not exist in sizes larger than 1/4" at local home improvement stores (Rona and the Canadian branch of Home depot), but is expensive in specialized plumbing stores.
My mechanical engineering course matter got me thinking about this again for some reason, and made me want to resume an old project. This is for a homemade, but I'm sure it could be adapted to be a replacement.
Now, on to the issue at hand... I'll make a drawing of what I was considering and upload it when I can. I'd make it with rubber washers instead of O-rings, as I don't really like the seal conditions for O-rings.
Threads would be pointless, unless you make it a longer valve, in which case it'd be a bit more difficult to build, and you'd want to leak proof it, using Teflon tape or something of the sorts, making it difficult to open. Unless it's screwed together, it needs to have a really solid construction, strong springs and a good seal on the first build. Rubber washers would work in place of an O-ring there, so long as there is a washer and nuts holding it in place. It would also not require much in terms of strength dealing with pressure either, as it is only water going through it, it does not need to withstand pressure in excess of the pressure of water, if I'm not mistaken. The Super Soaker ones are over-engineered to prevent any possible failure, and have tons of screws or are just epoxied together. (Correct me if I'm wrong in assuming the CPS 2000 was the first with that trigger system) The only issue with screws is you need to make it watertight, and thus need some form of O-ring or something somewhere wrapped around the entire circumference being held between the halves, creating a watertight seal.
A possible material for the actuator is a bolt, either threaded all the way or just part of the way(Threaded partway would seem ideal as rubber washers could then be used to create the seal). There would be one internal spring to maintain the seal and close the valve after trigger release, and there would also already be something to pull on, being the head of the bolt. I'm not sure if including a trigger torque arm as they do in their design is necessary, but if so that would be figured out while prototyping.
The output would likely need creation of linear flow, so straws and maybe screens to hold them in place would help. Somewhere on SSC(dating myself there) there's an article about a stream laminator, of which the design would be similar for the nozzle.
For a prototype, I was considering 3" PVC for space to work with things, with two NPTs on one side, and a 3/4" CPVC threaded adaptor, as that is what I have laying around. I was designing this for a homemade CP system (Which once again is lost somewhere in SSC, something about a sideways CPH) as they did in the old days, as I have some LRT left over from when I repaired my CPS 2000's bladder. Although now that I think about it, the size of washer would be ridiculous for a 3/4" hole, thus requiring me to think smaller (3/8"-1/2") pipe would be ideal, or simply installing a NPT right in the valve itself. I'm not sure, but it'd require lots of production cares such as making sure that the PVC for the output is flush with the inside of the endcap such as to improve the flow. That and smoothing the output intake, such as to once again increase the flow rate. (Curse you fluid mechanics... making me think of this scientifically)
For future creations though, I have seen some waterproof water conduits, which would work wonderfully with vinyl tubes (Not that there's anything wrong with NPT connections, these just look cool and adapt to different sizes, clamping around them). I was also considering figuring out the pressure that makes the SS PRVs work, as it would help to prevent bladder bursting issues, and could be tuned to different situations.
The only real downside to the design really is the dead space and not exactly linear flow of the water, but it could work very well if executed properly.
A friend of mine is printing parts of a RepRap (3D printer) for me when he gets his working, so I could possibly be drawing something up in CAD and fabricating a prototype of sorts at a later date.
[EDIT 6/18/2012] Attached the file with an explanation to the end of this post. Black large parts are rubber washers. Same style of seal is created both ways, this is just an alternative. Sorry about the small size, it's legitimately drawn on a 2x2" sheet of paper.