8
$\begingroup$

I have a steel coil spring that is used by compression. It is too strong, and I would like to reduce its strength by some fraction, ideally keeping its length.

Reducing it to roughly half would be nice. Uncontrolled heating and cooling in air would make it very unelastic, I think, and shorter after the first compression.

It can be manually handled. (The dimensions are: cylindrical with diameter 3.5 mm, length 60 mm, and round wire wire diameter 0.5 mm.)

I would be interested in the theory, even if it's something that I can not do, like precisely controlled heat treatment. I have only two of them, it can be very slow.

$\endgroup$
  • 2
    $\begingroup$ The Young's modulus (elastic modulus) of steel in remarkably consistent over various alloys and carbon contents. The difference that heat treatment makes is to raise or lower the bending stress that the steel can withstand. So anything you do with heat will just make it deform and stay deformed. $\endgroup$ – TimWescott Dec 6 '19 at 21:07
  • 1
    $\begingroup$ Have you thought about of, instead of manipulating the compression springs, adding some tension springs as “negative strength” springs? You could possibly run them inside the compression springs, to save space. $\endgroup$ – leftaroundabout Dec 7 '19 at 14:38
  • $\begingroup$ Short answer: Nope. $\endgroup$ – Hot Licks Dec 8 '19 at 23:46
  • $\begingroup$ @HotLicks That is a good summary, I'd give it a +1 as an answer! $\endgroup$ – Volker Siegel Dec 9 '19 at 2:54
  • $\begingroup$ Do note that if you have two springs, and there is no requirement that the length be maintained, you can just stack them into a single assembly twice as long. $\endgroup$ – Hot Licks Dec 9 '19 at 3:25
7
$\begingroup$

You could try a bath of concentrated acid. As long as you could maintain circulation so the concentration was fairly constant, a spring presents a very uniform cross-section and should be dissolved at a consistent rate.

I would recommend keeping one of the springs as a reference so that you can measure the result after a short period and adjust your time correspondingly to obtain the desired result.

| improve this answer | |
$\endgroup$
  • $\begingroup$ Brilliant! I would never ever have thought of changing the shape instead of the inner structure. Even if I would, I would immediately reject the idea because I explicitly keep the shape. Or so I thought. $\endgroup$ – Volker Siegel Dec 6 '19 at 14:43
  • 3
    $\begingroup$ This may be dangerous depending on the acid chosen. Some acids will leave the remaining steel very brittle, which would be bad for a spring. $\endgroup$ – dwizum Dec 6 '19 at 21:08
  • $\begingroup$ @dwizum Why does the steel change inside? Are these acids diffusing into the material? Or do they make hydrogen into it, and that causes hydrogen embrittlement? $\endgroup$ – Volker Siegel Dec 6 '19 at 21:58
  • 3
    $\begingroup$ The H ions/protons are attracted to locations of tensile stress where they may lower the inter-atomic forces between iron atoms in the lattice ., more of less . There are shelves of books on the subject in any worthwhile technical library. All acids, some organic poisons (S), hydrogen gas, electroplating , etc.etc. do it. I may be biased because hydrogen stress cracking provided a pretty good income for me. $\endgroup$ – blacksmith37 Dec 7 '19 at 0:11
  • 1
    $\begingroup$ Also, any minor surface irregularity could leave you with a very non uniform result - pitting and acid trapping .... best bet would probably be keeping the acid in constant motion. $\endgroup$ – rackandboneman Dec 7 '19 at 12:34
4
$\begingroup$

By "strength" do you mean stiffness/modulus? The modulus does not change with tensile or yield strengths. Make it with wire of 1/2 the diameter for 1/2 the modulus. Acid will hydrogen stress crack hardened steel ( This condition has many other names like "embrittlement".). Make it out of aluminum and get 1/3 the modulus or titanium and get 2/3 the modulus ( approximately). Doing your own heat-treatment could work ; simple enough to try : Heat to orange in a gas flame ( 1600 F +/-) hold time - a couple seconds , IMMEDIATELY quench into cold salt water. Temper at 300 to 400 F ,somewhere around an hour is traditional. Good for carbon and most low alloy steels. High alloy or tool steel -another story. The heat-treatment risk is the exceptionally small thickness which could partially decarburize in a few seconds at 1600 F. Commercially it would be heat-treated in a protective atmosphere or vacuum , possibly with induction heating.

| improve this answer | |
$\endgroup$
  • $\begingroup$ (note that a coil spring in compression is everywhere locally in torsion, so your effective modulus for the structure scales with the square of the wire diameter, not the diameter itself) $\endgroup$ – RLH Dec 7 '19 at 21:19
3
$\begingroup$

Probably just another hypotetical idea: If you squeeze the wire from circular into a square shape, it will have about 54% of the original rate. If you squeeze it slightly more into a rectangle, it will at some point become 50%. And 1/2 the diameter of the wire does not gie you 1/2 the rate. It would give you 1/16 of the rate. 85% of the diameter gives you 1/2 the rate.

| improve this answer | |
$\endgroup$
  • 1
    $\begingroup$ And you only need a press to squeeze the wire into shape... cheap then... $\endgroup$ – Solar Mike Dec 6 '19 at 21:25
3
$\begingroup$

Another hypothetical way to reduce the amount of material in the spring would be to fix it in a jig stretched to double its normal length.

Then, heat it sufficiently to anneal it and remove the tension in it.

Then, re-temper it and cut it in half, back to its original length.

But just buying a new spring would be a much simpler method.

| improve this answer | |
$\endgroup$
  • $\begingroup$ This would make it stiffer. $\endgroup$ – Hot Licks Dec 8 '19 at 2:42
2
$\begingroup$

I would consider perhaps the easiest way would be to change the leverage...

Failing that, I would get a new spring made with the details necessary and spare one 10% above and below your "new" value.

| improve this answer | |
$\endgroup$
  • $\begingroup$ I can not change the leverage because it is in an fully confined space. But it's a good idea. $\endgroup$ – Volker Siegel Dec 6 '19 at 14:44
  • 1
    $\begingroup$ Regarding getting a new spring, I did not check whether it is a standard size and available in a less strong variant. But building it myself is hard, because after heating and bending the wire to a coil, I would need to spring harden it. I think that is one of the most complicated methods of heat treating, including multiple heating and cooling cycles to specific temperatures, keeping it hot for a specific time and most importantly, controlling how fast it is cooled. I'm pretty sure the values even depend on the kind of steel. $\endgroup$ – Volker Siegel Dec 6 '19 at 14:54
  • 1
    $\begingroup$ If this is for product design, then look into getting it custom made -- springs are a challenge for the casual metalworker, but no problem for a shop; as a consequence, custom-made springs abound (they certainly do for cars). If it's for a one-off, see if you can find a machine shop or retired machinist who will do it. $\endgroup$ – TimWescott Dec 6 '19 at 21:11
  • $\begingroup$ @TimWescott That makes sense. I only want to adjust two springs. They are used in a clamping device. It's almost perfect - except that the clamping power is too strong and make it useless. $\endgroup$ – Volker Siegel Dec 6 '19 at 21:52
  • $\begingroup$ @VolkerSiegel if it's a normal coil spring you should be able to buy low quantities from specialist suppliers. I bought 4 of each of 3 lengths/strengths for a design from a UK firm called Springmasters (as your profile says Germany, you can probably find a local stockist of an equally large range of sizes) $\endgroup$ – Chris H Dec 8 '19 at 17:32
2
$\begingroup$

While I think the concentrated acid solution is brilliant, I'd like to suggest a refinement.

Rather than use just any old acid, you should specifically use a chemical etchant.

Typically, a chemical etchant's nucleophile strength does not come from Hydrogen, so it should have the benefit of much lower -if not entirely eliminated- hydrogen enbrittlement. (At least theoretically, anyway.)

A good etchant for this might be Ferric Chloride, [FeCl3]. Copper(II) Chloride [CuCl2] is another possibility.

Getting an even and precise etch will also be up to idiosyncrasy's of the exact setup you try and run. Chemical concentration, bath agitation, cleanliness of the work, temperature, ...


Another, related, option is to use electrochemistry instead of direct chemistry.

Specifically, use the reverse of electrowinning, electroetching.

One of the pros of this method is that, with controlled current + time, you could fairly easily calculate and control the exact mass of the material removed by the relationship between the unit of electric charge (coulombs), and the unit of molar mass (moles.) This extrapolates well to controlling the exact change in spring's mechanical property.

With direct chemistry, concentration + temperature + time = mass of material removed. Even if you precisely dial in the last two variables, the first one is going to be a ever changing target, as the work will dilute/poison the bath.

Probably the biggest con of this (aside from requiring a semi-sophisticated electronic power supply) is that surface irregularities will become much more of a problem than with chemical methods. In particular, any insulator will be practically inert to etching with the electrochemical method.

Conversely, an insulator might still be attacked by a pure chemical method. e.g. Any oil on the surface might become esterfied to something soluble, be freed from the surface, thus allowing the etchant to get at the actual work piece.

| improve this answer | |
$\endgroup$
1
$\begingroup$

Another way to decrese the stiffness without making the spring longer:

Making it longer would obviously decrese the stiffness. And adding other springs in series would do also. You could add springs in series and put them one in another. The inner springs should be shorted and connected end-to-end. You could have three springs in total. The second connected to then end of the first, going back inside. At some point (say half of the outer springs lengt) you append another spring, which is again inside the second. It should be long enough to reach out of the first one. Well, you may have to cut the first one a little bit.

I don't expect this will be useful, but it really depends on factors as your packaging, available of springs, what you really want to achieve...

| improve this answer | |
$\endgroup$
  • $\begingroup$ Wow, that one is impressive! I think it is the first solution that is solved by plain, clean engineering, a opposed to all kinds of brilliant hacks. The first solution with thinning the wire blew just blew me away, because in my mindset, I had the explicit requirement that the geometry of the spring stays the same. I had missed that the cross section of the wire can vary, even in area. Right, this does not solve my practical problem, there is just no space for it. But to me, adding to the body of structured knowledge of SE is as important as getting an answer, so this is very relevant. $\endgroup$ – Volker Siegel Dec 14 '19 at 18:02
0
$\begingroup$

A possible way you could fix this, if you have access to something like a blowtorch, or better yet, a blacksmithing forge: you can heat the steel up until it is glowing orange turning yellow, and then find a place to set the spring down while it cools without anything touching it. If you allow it to cool slowly enough, the spring will become annealed, softer and more malleable and flexible.

| improve this answer | |
$\endgroup$
  • $\begingroup$ I could do that, and it would become annealed. But then, the material is no longer spring-like. It would be malleable so that compressing the spring would give much easier - that's good. But it works only one time, because it stays short, and does not spring back. There is a heat treatment to make it a spring again, and even a weaker one as I want. But that heat treatment is very complicated, with multiple heating, keeping temperatures, and controlled time for cooling. $\endgroup$ – Volker Siegel Dec 8 '19 at 2:00
  • $\begingroup$ You could try that annealing of a spring yourself to feel the difference, if you find a small spring that is no longer needed. If it's small enough, a candle could be enough heat. You do not need to make all wire glow at the same time. $\endgroup$ – Volker Siegel Dec 8 '19 at 2:04
  • $\begingroup$ I just tried it! I'm the kind of person who has a box of old springs in the living room. And a blowtorch. And, of course, a fire extinguisher. So, the spring was easy to compress, and it did not get harder along the way. And it did not go back to the long spring shape at all, it just stayed short. The different was greater than I thought myself, it's worth trying. I cooled it slowly by not simply taking it out of the flame, but taking it further away in the hot gases of the flame at first. Not sure that made a difference. ) $\endgroup$ – Volker Siegel Dec 8 '19 at 2:20

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.