Need to make an odd spring or two

  • I have an idea that requires a rather unusual spring or two. I would probably need to bend this from wire, as I don’t think there is anything already available that would work. I have no experience in this so would like to talk to someone who could brainstorm with me and give me some ideas of how I might go about doing this. Let me know if you have any experience to offer.

    Thanks a lot.


  • There are many people here that could potentially help. Can you provide some sketches or dimensions and specs about what you trying to do? Also make sure to take a look in our box of springs. We have taken lots of strange springs out of printers and other things we have hacked.

  • I have some music wire in a few sizes you could use. I can show you how to set up the lathe to wind springs depending on your specifications.

    You should probably look at this article:

    It outlines the parameters you need to watch to get a given result.

  • Arasbm:

    Okay, I was avoiding this because I don’t know the nomenclature, but have since looked it up. This is approximate, as I am not sure what will be needed in the end, and it will depend on experimentation to be sure, but for now, in broad strokes, I need:

    A linear compression spring with closed square ends (the latter not so important).

    Spring measurements:
    Free length: 500 cm
    compressed length: 45 cm
    inside diameter/rod diameter: 20mm
    material size: 1mm

    For now, the direction of wind does not matter. Later, I may end up needing two, the second one being of the opposite direction of wind to the first.

    … I told you it was an ODD spring! With what I had in mind I did a little lol when you suggested that I look through the box of printer springs…



  • LOL! We have a lot of odd springs in the hack room, but yes I do not think we have any 5m long springs just laying around.

    I like @pierre’s suggestion to use the lathe, but make sure you get him to help. Although I dont think the sherline lathe can do a 45cm long spring (~18"), right @pierre ?

    @Chainmaildave frequently makes springs with a drill and then cuts them into rings of course. He has a pretty good technique for doing that and probably would have some tips for you.

    Sounds like an interesting project :smile:

  • @pierre

    Thanks for your input. I did have a look at the article at your link. It was helpful in understanding the basics of how a spring is turned. I think you can see from my reply to Arasbm that a small lathe won’t do. Probably a pretty big one, but maybe something could be modified to take a large rod to do what I need. Music wire definitely would not do… and this brings me to the question of what kind of wire might work for the dimensions that I am thinking of?

    I saw this and it interested me:

    It is similar to the one used in part two of your article. Later in the article he makes a spring from

    .055" wire, which is 1.4mm, so stout enough for my purposes.

    I am also concerned that it would seem that the wire needs to be heat treated after being bent in order to relieve the stress. As you can imagine, this seems a bit difficult given my five meter finished length.

  • @arasbm

    Thank you so much for your interest and quick reply!

    From your comment about bending a 45 cm spring, can I take it that it could be bent into shape in its compressed form and then stretched later? This would certainly make manufacturing easier!

    I pictured trying to turn a 5 meter long spring on a lathe…


  • @Lawrence I have never done this before, so I am not sure if this plan could work or not. I definitely recommend doing a few trial runs in smaller lengths first.

    I was thinking you could turn it to its compressed length, and then stretch it to the full length and secure it. Then you would have to heat treat it (harden and temper). You would have to do some calculations to get the right diameter.

    Like I said, I have never tried this, but that is what I would try.

    I dont know anything about your application, but I am assuming you couldn’t just get away with stacking a bunch of smaller compression springs together, right?

  • Arasbm,

    Smaller trials for sure. I don’t think jumping right into trying to do the full length would be prudent. I have read that bending spring wire is dangerous for the newb.

    Okay, so how about heat treating it by running current through it? This would avoid the problem of having to build an oven or something that is 5 m long.

    You are right in assuming that we couldn’t get away with stacking, unless, of course, we want to weld them or crimp them somehow together end-to-end. The finished product needs to be one big spring. I suppose the length could be shorter than 5 m, but I’m not sure until something is built and tested.

    For the prototype at least, springs could be welded together. It would not have to be too pretty.

  • Fastenal carries music wire in 0.041 (which is pretty close to 1mm) stainless:|categoryl1:"600930 Raw Materials"|~ ~|categoryl2:"611723 Wire Products"|~ ~|categoryl3:"600946 Wire Coils and Spools"|~ ~|sattr01:^0.0907"$|~

    and Carbon Steel:|categoryl1:"600930 Raw Materials"|~ ~|categoryl2:"611723 Wire Products"|~ ~|categoryl3:"600946 Wire Coils and Spools"|~ ~|sattr07:^Spring$|~

    The stainless will be stiffer, but it is terrible to work with…

    Their next size up is 0.091:|categoryl1:"600930 Raw Materials"|~ ~|categoryl2:"611723 Wire Products"|~ ~|categoryl3:"600946 Wire Coils and Spools"|~ ~|sattr07:^Spring$|~

    You are certainly not going to get a 5m spring out of my little Sherline… I’ve only got a 12" bed on it. :) If we can find a rod or pipe to form the mandrel, we can probably rig some sort of feed mechanism to lay up the coils. You really want to form the spring using the free length, not lay up the coils tight and stretch it for a compression spring. The geometry will not come out right and it’ll suffer metal fatigue in way less cycles if you try stretching it.

    I’m wondering if they make a 3/4" OD EMT Conduit? It’d be around 19mm, leaving room for the spring to expand to 20mm when you remove the coil. That’d probably work for the mandrel.

    How many turns are you planning to put on?

    Cooking out the coil will likely not be a huge issue… We’d put it in a pipe, filled with oil, and heat the oil bath to whatever temperature we need. I’ve done this for tempering large blades before. We could also use scrap drywall to rig a little electric oven, and use a PID controller (or the guts of an oven thermostat…) to do temperature control.

  • Pierre,

    Wow, posts running fast and furious!

    Thanks for your input.

    The Fastenal source is great–didn’t even know they existed. Eventually, stainless will be needed but for now, carbon steel is fine. I like your idea of rigging a pipe into a mandrel. I understand your concern about forming the spring at free length, but, for now, there will be virtually no cycles to worry about. It will be tested at its free length–the need to compress it will come much later, and probably after a lot of design changes are made. The geometry not coming out may be more of an issue, but maybe some distortion would be acceptable given that we will be tweaking it so much anyway.

    And to make the first test easier to arrive at, and since compression and cycles will not be an issue for now, maybe we could bypass cooking it altogether until we come at something closer to what we need?

    Ditto for the distinction between 19 and 20mm–not a problem right now at all. Approximate is fine. It can be wider but not too much narrower.

    Number of turns–no idea. “enough” but in this case, “more is better”, as long as it fits to the 45 or so cm when compressed (ie when the coils touch each other, if that is possible). So, with, say a 450mm compressed length, and a 1mm dia wire, we can ballpark a maximum of 450 turns. This gives us about an 11mm pitch. I like this because the ratio of the pitch to the diameter seems about right.

  • @pierre

    I found this:

    I am not sure if the 3/4 inches refers to ID or OD. I believe the OD may be the .92 mentioned, which is 23mm.

    If that is the case, then maybe the 1/2 inch EMT Conduit here:

    Would work. It would have an OD of 0.71 inches, 18 mm, which is pretty close. It would work.

    In both cases the conduit is only 10 ft long, 305 cm, not the 500 cm I was looking for, but it would do.

    Any thoughts?

    Again, thank you so much for your input.


  • Just the musings of an old man ☺ I think you will find that a 5m long spring made from 1mm wire isn’t going to function like you might expect. The sticktion between the wire and the rod it rides on will be greater than the spring resistance itself. If the rod is vertical I would take a wild guess and say that 60% of the spring will end up being fully compressed under its own weight alone. Are you looking for a linear or progressive resistance along the length? There might be a better/easier option that just needs to be thought up🙄


  • That’s a good point @Bishop … When I need a long-throw and relatively constant force, I usually use a dashpot or gas-strut setup. I think 5 meters would exceed any such setup you would normally find.

    If the forces are going to be entirely vertical and not moving, a pulley and counterweight is my go-to option.

    If I need it to work in any orientation, a stronger spring driving cams or levers might get me the throw I need.

    I think @Lawrence, we might need a little more info about what the spring is trying to do.

  • Sorry I have been busy with year-end things, and have not got back as promptly as I would like.
    There are a few things I would like to do in January.
    First, sign up as a makerspace member and I have some sewing to do and a few small things.
    Second, I have a Jobmate to donate that I will bring by.
    I would also like to bend a prototype spring, as we have discussed.

    I have inquired with fastenal about ordering some wire. Pierre, would you be willing to help me to bend some of this around the home depot conduit mentioned above? Also, the bit about “cooking” it with a pipe and oil is great, but I would need at least another brain to help me with that part too.
    Hope you all had a great holiday and best wishes for the new year.



  • @Bishop Sorry I did not reply to this earlier, but as I mentioned, I have been busy with year end/beginning things. The spring will not be riding on any rod. It will not be vertical, and so not much affected by its own weight. I am sorry that I am not well versed in engineering principles and had to take more time to look up what you mean by linear/progressive resistance. It makes no difference, but I assume that linear resistance will be the result of a simple spring.

    Thank you again for your insightful thoughts.


  • @pierre Thanks, Pierre. I understand that this is difficult to imagine given that you don’t know what it is for. I promise to come clean as soon as I can. A pulley, cams, etc, will not work, needs to be a big spring. When you see it you will understand, and I hope that that will be soon. The spring does not really need to do much work at all, except, with little or no resistance save air and a bit of friction, expand from its compressed length to something close to its full length. For now, it does not need to be durable in cycles, ie it does not need to perform for many cycles at all. For initial testing it does not need to perform any cycles–ie it will be tested at or near its full, uncompressed length.

    Thank you again for all your thoughts.


Log in to reply