path: root/ProjectLV2AirframeFishClamps.mdwn
blob: a35b73265d814e3a5114f45156b8b2f37eecbff1 (plain)
# <a name="Preliminary design drawings for"></a><a name="Preliminary design drawings for "></a> Preliminary design drawings for a detachable launch clamp

**Advantages of detachable clamps**

- Better aerodynamics
- Higher vehicle mass ratio
- No internal space used in the vehicle

**Disadvantages of detachable clamps**

- Construction more difficult v.s. screw-in lug
- Careful and time-consuming testing required
- Possibility of fin damage as the clamp leaves the rocket
- Possible premature detachment and incorrect flight path
- Possible failure to detach
- Small kick given to airframe at detachment might effect path
- Additional personnel hazard due to flying clamps
- Additional personnel hazard during assembly of sprung elements
- Longer assembly time at launch site

Despite the considerable difficulties, launch clamps may be worth pursuing if some of the design challenges can be met.

The background information is basically this. Curently LV2 uses two screw-on launch lugs. These are simple plastic bobbins that slide in the launch rail. They are attached to the airframe skin by approximately 2 inch long stainless steel screws threaded into the aluminum underframe. The bobbins are about 1 inch long and 3/4 inches in diameter.

The primary disadvantages of the launch lugs are additional aerodynamic drag, and the necessity of another hole in the airframe. Though no detailed caclulations have been offered, the added drag is probably at least proportional to the increase in projected frontal area. The un-adorned frame has an area of about ( pi \* 5.25"^2 / 4 ) = 21.6 in^2 . The lugs have an area around 3/4 in^2 , or around 3.5% of the whole frame. Probably this is a little optimistic. Even so, the increased drag is not overwhelming, so it only makes sense to use a detachable clamp if a high confidence in its reliability is obtained.

What we're proposing is a clamp of minimum mass, low mechanical complexity, and high reliability. The clamps consist of a rail-riding mechanism employing two bobbins almost identical to those used in the screw-lugs. The rail-rider bolts to a 9 in^2 curved pad which is held to the airframe by a 250 pound-test nylon monofilament line. As the rocket clears the pad the clamp passes a fixed razor blade which cuts the monofilament line. The airframe continues upward while the clamp is deflected from the flight path by a curved rail extension.

This idea is still in the early stages, however, a preliminary drawing is attached below. The [[DXF format drawing|FishClamp-0.0.dxf]] is the most useful, but requires a mechanical drawing program to be read. (A free and open-source drawing program is avaliable at [;product=qcad]( )<br /> As an alternative the drawing is avaliable as two PDF files. [[FishClamp-0.0.pdf]] is a print of the DXF file with all the layers turned on. [[FishClamp-Hidden-0.0.pdf]] is an alternative print with some layers removed, and most of the hidden lines eliminated. It should be more suitable for casual examination.

I must have gotten Lucky or something, as I found this the day it was posted. I recall Tim mentioning somthing of this nature, but intuitively I thought "The area of that one strand of fishline against that slick rocket skin... I have doubts as to whether it will hold onto it." Otherwise, it has the following advantages: 1) All machining operations are lathe, drillpress, and perhaps deburring. I could do the entirety of this setup at home in about a week or so. 2) if the rocket does rip clear of the clamps, i.e the Nylon stretches and looses tension, it won't be a problem, as the sharp edges of the fins will simply cut it. Our only worry is if this happens so early that there is significant deviation from the vertical. 3) If I cannot get that other clamp back from Jimmy, this setup will have to do. 4) Testing is just a matter of seeing that the assembly slides smoothly and that the cutter (better use something stouter than a razor blade!) cuts the line. If it passes those tests, then run it. But my concerns mean nothing. Run the thing. [[DennisYoung]]-18 Feb 2003