Imagine a soft spring in series with a stiff spring. Work it out from that.
You haven't got many variables to play with. You can either taper the wire or change the coil diameter along the spring, or both. You can also adjust the pitch of the coils so that some short out at high load.
Sometimes we use an auxiliary spring in parallel with the main spring (a jounce bumper).
You can buy SAE paper 2004-01-1557 and/or 1925-01-01 (probably much better)
Cheers
Greg Locock
New here? Try reading these, they might help FAQ731-376
Elastic deformation of metals is linear. What you need is a mechanism is that compresses the spring at different velocities depending on position. I designed a descending rate spring that acted around a circle. The spring was linear. The spring rate of the the rotating shaft was not.
The method to design anything should probably BEGIN with good description of what's up.
Pictures, sketches, and more than one sentence.
"Define the problem."
Run the spring off a cam. The slope of the cam at a point will be the spring rate at that point. Put in a couple of detents if you want. Or reverse the cam slope to let the spring drive for awhile. But remember the clutch pedal assist spring in the old Chevrolets. It was just a simple over-center spring, but if you put it in the wrong hole, the clutch pedal would slam to the floor after depressing it just a little.
I think the OP went on vacation while we solved the problem. But this brings back memories of wedging a small block of wood behind that spring that would move it just enough to feather a clutch to death within a hundred miles.
