Initial tension necessary to overcome "sag"/"bow" horizontally

I am the originator of this question and will gladly give additional info about this problem in order to help get the answer.

the spring will form a catenary curve under tension and since it is a spring the tensile strength is just the spring rating...pull any harder and it gets longer.

Typically I have seen this solved with a tube and a rod that slides inside the springs that fit easily into each other and telescope with the spring action and they have a support that keeps them rigid and capable of holding the weight of the spring.

The tube is attached to one side and the rod to the other and they have to allow for the full range of motion

Sounds a bit like a car shock-absorber on its side?

Similar, the inner sections of a car hydraulis shock absorber contain a fluid that is pumped through an aperture to drain energy from spring oscillations. Variable rate shock vary the aperture. In this case the concentric tube and rod will not have a hydraulic function, but the concept is valid.

You can only approach "straight", never quite getting there. More tension usually requires larger diameter wire which is more weight which in turn creates sag. Aurizon's solution is the best that I can think of.

You cannot eliminate spring sag in a horizontal unsupported spring, but perhaps you could mount the spring remotely, and use a cable, Kevlar cord, etc to fit though the small space?

The "movable" block is lead screw driven, however the "fixed" block end actually moves the same distance that the movable block moves. The extension spring does not compress or extend under "ideal" conditions. The "artificial" increase in "initial tension" to keep the coils tightly together appears achievable only through a cable, (adjustable for tension), thick enough to fill the I.D. of the spring.

I thank each of you for your input.

Respectfully,

BluThundr

Physics and the law of gravitation guarantee that there will be deformation of an horizontal, end-supported, uniform cross section, member in a catenary fashion no matter the size or material so long as it has mass and we are speaking in Newtonian terms. The real question probably is, "How much deflection can be tolerated?"

The physical arrangement described by #2 is almost certainly the most practical solution. Another one does come to mind, however.

With computer it is possible to design springs in ways that were hitherto impossible--for instance a spring in which the increment of force required to extend or compress does not change with the physical length change. That is, a constant linear relationship between force applied and distance moved.

It should also be possible to design a spring with a k in two axis. Imagine a screen door spring with a built in arch such that it approaches dead level under its own weight when oriented properly and fixed there. Then what happens under working loads and what torsional effects might be observed? Such simple little ideas can generate a lot of complexity.

Stick with #2 I think.

The spring is obviously attached to a fixed object at one end and a moving object at the other. There must be some sort of guide that the moving object runs on so how about having a shaft run through the middle of the spring like this.

The idea is that the shaft is attached to a fixed object and the spring is then placed over the shaft. The moveable slide is then installed so the shaft runs through its middle. As the slide moves back and forth on the slide the spring can either compress or expand and will not sag since it is supported by the shaft