Compression vs. extension springs and tempering...

[RAc]

Hi everybody,

 

I have devised a way to make cone springs needed in a lockdown design project (details to follow). I found a video (https://www.youtube.com/watch?v=ubQqYq5vmnc) in which it is suggested that any compression spring could be made by making the corresponding extension spring (meaning adjacent tight windings) and then pulling apart the spring. Now it is much easier making an extension cone spring than a compression cone spring (you basically lead the spring wire in between two washers the wire thickness apart and wind under tension). Pulling out the cone into a compression form afterwards works, but the spring is not durable - unless, as I think I have figured out, I temper the spring after pulling out (this appears to work fine using a normal cigarette lighter by holding the spring into the blue part of the flame until the spring turns copperish).

 

Is this an appropriate way to make lasting good springs, or can I expect the springs to weaken out after some time longer than the few minutes I had a chance to test? If the approach basically works, are there better ways of tempering which do not require professional type equipment? (I read somewhere that springs need rather high temperatures to temper; would a gas torch be sufficient)?

 

Any other thoughts on the subject?

 

Thank you for your responses!

 

[Tiposx]

I was under the impression that heating then quenching copper or brass anneals it  rather than softening it. I will have to look it up, as there must be an accepted way of hardening it.

 After experimenting with work-hardening on various kinds of materials including phosphor bronze I couldn’t get exactly the qualities I wanted. Now I make my own torsion springs for ‘tinas and melodeon from stainless steel spring wire. I don’t heat treat it at all.

Tiposx

[RAc]

Thanks for the reply, Tiposx!

 

Sorry for not making myself clear, I am using steel spring wire (fwiw, 0,8mm ordered from Juergen Suttner, but very likely any fitting guitar string will do likewise).

 

I understand that with this material, you don't need to temper if you wind the spring directly into its target shape, but as I explained, I am taking the extension first, then force into compression form factor strategy which so far I wouldn't have thought would work. I was just wondering about more thoughts on the subject.

 

Thanks again!

[Tiposx]

I would turn up a conical former from brass, then attempt to thread cut a thread into it. I would need suitable lathe change gears for the screw pitch required. It wouldn’t be a simple job.

I guess that doesn’t really help though!

[RAc]

I actually tried to go down that road first, but the dimensions are fairly small, so as you suggested, tools available to the everyday person wouldn't be able to build a mandril for that job. I then considered designing a mandril with FreeCAD, have it printed and use that one, but I'm quite sure you'd also need sophisticated tools to get that to work (I don't have a lathe to begin with, just a Makita rechargable battery powered drill).

 

It seems as if my technique works, I'd just be curious to hear other folk's opinions; very likely there's something I don't see or know (yet), and I'd hate to drive a truck down a dead end...

 

Thanks!

 

[Dana Johnson]

Generally hardenable steels are heated above their normalizing temperature (alloy dependent, but generally above well above 1200 degrees F) then quenched or quick cooled to lock the steel into a hard crystal structure, then reheated to a lower temperature to draw out some of the hardness to make it useful for something other than a file.  Spring steel wire is annealed as thick rolled wire, then drawn out which hardens it.  If it needs to be drawn out very thin, it may undergo a number of annealing stages between  drawing through dies which makes it hard again.  300 series stainless steel can only be work hardened.  But for practical purposes, all spring wire is work hardened by drawing through dies.  Trying to re-temper a spring first has to go through the normalizing temp. Quick cooled, then reheated to the correct  tempering temp again.  None of this is a very practical thing to do since it would need to be done in an inert atmosphere among other issues.  By choosing the right diameter wire, you can stretch out a tight spring and have it work, but you need the wire size that is strong enough to give you the right result.  Stretching a spring doesn’t deform the wire any differently than winding it does.  It is just bending the wire in a different direction.

Dana

[RAc]

10 hours ago, Dana Johnson said:

By choosing the right diameter wire, you can stretch out a tight spring and have it work, but you need the wire size that is strong enough to give you the right result.  Stretching a spring doesn’t deform the wire any differently than winding it does.  It is just bending the wire in a different direction.

Dana

 

First of all, thank you very much for this valuable information, Dana!

 

Could you elaborate a little bit on "right" in the two contexts you use the word ("right diameter wire" as well as "right result")? Do I read you correctly when paraphrasing this as "depending on the wire strength, you need exactly the amount of stretching that does the job but doesn't overstretch the spring" (which effectivelys reiterates Charly's 4:44 statement "don't stretch it too much, or it will ruin the spring")? Effectively this would translate to a trial-and-error process, sacrificing springs until the right amount is accomplished?

 

Thanks again!

 

Edited February 9, 2021 by RAc

[Dana Johnson]

Spring wire like any wire will return to its original shape unless bent past its elastic limit.  Beyond that point it will not return all the way.  The wire isn’t damaged at this point.  Bending beyond a certain amount will start to create tension and compression failures which do weaken the wire.  Coils that are too small for the wire diameter will do this and you can see it under the microscope.  Annealing lowers the elastic limit, work hardening increases it. Spring wire needs a high elastic limit to be useful.

   In this context, “right” is dependent on the alloy and its properties after drawing.  As long as the spring wire is not bent to the point of initial failure,  you can use the new shape as a starting point.  The only way for you to check this is to try it out and see if your newly stretched spring retains its shape when you use it.  The actual force the spring can exert or resist, depends on how much it is flexed in use, the wire diameter, and its length.  Coil springs turn long wires into short spaces, so for a given pitch, adding coils = adding length and making that spring less forceful for a given deflection.  If you want a stronger spring, you can reduce the coils, make the wire thicker, or make the coils smaller so they don’t contain as much wire.  
   Conical springs are very handy for compression, since they can be squashed flat since the coils nest.   You may find that the wire you got from Juergen just isn’t thick enough for your type of spring.  Going up a gage or two is something to try.  One other thing, tension springs are generally wound with a negative degree of pitch which makes the coils very tight to each other, pre loading them.  By stretching these springs, you both add new deflection, and overcome the negative deflection that is built in for a total deflection that is larger than it appears.

   Sorry if this is a bit confusing.   There are a lot of resources on the web for spring design info.

Dana

[RAc]

14 hours ago, Dana Johnson said:

...which do weaken the wire. Coils that are too small for the wire diameter will do this and you can see it under the microscope. 

...

 You may find that the wire you got from Juergen just isn’t thick enough for your type of spring.  Going up a gage or two is something to try. 

 

Once more, thanks for the very comprehensive and useful answer, Dana!

 

One last thing: The two sentences I extracted to me appear to contradict each other, but that probably means there is an error in my thinking. The first sentence to me reads that for each wire, there is some minimum coil size under which the spring can not be wound without affecting the elasticity (the absolut size of this coil diameter of course depending on the wire thickness and other factors, so looking at the wire size alone will not suffice to determine the minimum). Applied to cone springs, this implies that the small end of the spring (which has the tightest coil) in turn determines which material can or can not be used. For example, if the small end of the coil must be 3mms, I must pick a wire that tolerates coils as tight as 3mm diameter, right?

 

Now if I choose a wire thicker than the one I have - given in theory all other factors are equal -, shouldn't this thicker wire require a larger minimum coil diameter, meaning the 3mm requirement might not be held anymore? Shouldn't I pick a thinner wire instead?

 

Thanks again and sincere apologies if this error in thinking should be prove me foolish (which it will likely do).

 

Edited February 10, 2021 by RAc

[ttonon]

RAc when you bend a wire having a circular cross section, the tighter the bend, the higher will be the compressive stresses in the side that is crimped and the higher will be the tension stresses on the opposite side, which is extended. The maximum tensile and compressive stresses developed in bending depends upon the ratio of the wire diameter to the bend (coil) diameter. The higher this ratio, the higher the stresses.

You can understand this intuitively by drawing a little picture of the coil and note that the geometry is defined by two lengths, and that the scale of the whole drawing is determined only by the ratio of those lengths. Thus, the scale of your picture is defined only by that ratio, and all combinations of wire and coil diameters that have the same ratio can be superimposed on each other by manipulating the magnification (scale) of the drawing.

Incidentally, this is the same situation with the bending of a beam.

Said simply, the only thing the material knows or "feels" about its physical state is defined by that ratio of diameters, and this explains why thicker wire diameters have more stress for the same bend radiuses.

Another suggestion I'd make is with the terminology, which can cause confusion. Broadly, there are "extension" springs and "compression springs. Using "tension springs" to describe "extension springs" confuses the fact that there is tension in both extension springs and compression springs. It's impossible to impart bending stresses in a wire without inducing both tension and compression stress.

I apologize for being so exacting with all this terminology. That's just what I do. I find that many people get confused by either not fully understanding the terminology they use, or they confuse others by not using the best definitions in their explanations.

Regards,
Tom
www.bluesbox.biz

[RAc]

Hi everybody -

 

for a follow-up, I present my solution for making small cone springs. Though it looks fairly organic and straightforward, the price for this construction were a perceived kilometer of wasted spring wire, about a dozen attempted solutions for the bin on the way and many many scratches, scars and dents in my hands (as you can guess from some of the close ups in the video).

 

It's here.

 

I'm quite satisfied with this solution (which is important because the springs are in the critical path for the project). Hopefully someone else can draw useful hints from the video. All suggestions and feedback welcome!

 

Thanks Ruediger

[ttonon]

Ruediger, congrats for coming up with a workable solution.  Good luck with it.

 

Tom