Variable Pitch Conical Spring

[chirpa12]

I am trying to design a linear conical spring. I've done some research and found that the way to make a conical spring linear is by making it a variable pitch. I found a post from a few years ago by member "Isrealkk" stating this. I have scoured my literature searching for a way to effectively design it but was unsuccessful.

Does anyone have any information on variable pitch conical spring design? (Equations, reference material, etc...)

 

[MikeHalloran]

The traditional way to design a conical spring is one coil at a time, which gives a piecewise approximation of an exact solution. You could probably extend that to 1/2 or 1/4 coil at a time, and use a spreadsheet to deal with the math.

I'm not convinced that variable pitch alone will do what you want; you may need tapered wire. It's been done, but you have to buy a lot of springs to justify it.



Mike Halloran
Pembroke Pines, FL, USA

 

[israelkk]

Variable pitch linear conical spring with constant rate is calculated exactly as a compression spring with the average mean diameter of the conical spring D = (D_top + D_bottom)/2. Therefore, the pitch between the coils is designed such that that all coils will touch each other at the same load.

The stresses should be checked for the small (top) coil and the large (bottom) coil. The wahl factor is larger for the small (top) coil therefore, the top coil may yield first.

What MikeHalloran refers too is the procedure to calculate a variable rate (non-linear) conical spring which have an equal pitch between the coils. In this case the large (bottom) coil which deflects more than the next coil will touch the next coil before all other coils comes to touch. When this happens the number of coils is reduced by one coil. Therefore, the spring rate will increase. Next the second large coil will touch the third coil and the rate will increase again. This process continues until all coils touch each other.

http://israelkk.googlepages.com/home

 

[MikeHalloran]

Ah. Thanks, Israelkk.



Mike Halloran
Pembroke Pines, FL, USA

 

[chirpa12]

Thanks a bunch, to both of you. A few more question though. How is the pitch calculated? How do I know the distance between each successive coil? Is there literature I can cite for this information?

 

[desertfox]

Hi chirpa12

The book Spring Design by W.R.Berry as the formula's for
conical springs.

regards
desertfox

 

[israelkk]

The deflection of a compression spring is a function of the power of three of the coil mean diameter (D^3). Therefore, the pitch for each coil will change as a function of D^3. For example, if the large coil diameter equals 1.1 of the next coil than the pitch of the large coil will be 1.1^3=1.331 larger than the pitch of the next coil. Usually, this task is left to the spring manufacturer to establish the pitches between the coils such that at the maximum designed load the coils will not touch each other. This will assure a linear behavior of the spring.

 

[chirpa12]

Thanks for the help. Here's another one for you. Is it possible to hold a compression spring at solid for 5 years or so without failure. I know it will set but will spring continue to relax over that entire time? Or does it relax only due to the intial set?

 

[israelkk]

Yes, it will continue to relax even after 25 years. The relaxation rate decreases exponentially over time but it accumulates. If relaxation is an issue then simple calculations of the spring are no longer simple and this is a whole new game. In the aerospace field there are products that can sit 20-25 years before they are used and this issue is crucial.

When you design the spring you need to make it stronger such that after the 5 years period the relaxed spring will still give the required performance. If it is possible, design the spring with the lowest stresses as possible which will require a larger and heavier spring and more space for the spring.

To my best knowledge there are no published formulations for spring relaxation over very long periods. The longest published test was conducted over approximately 11 month and it was a creep test under constant force (relaxation is under constant deflection). The test was conducted only on two spring materials and one wire diameter. There are dozens of different spring materials and add to this that each wire size has different properties.

Data or formulations to estimate relaxation over long period is a propriety information.

http://israelkk.googlepages.com/home

 

[sbozy25]

This is something that my company deals with all the time. One of our products are brake safety springs that go into big rigs. They set in the compressed state for years on end and may never get used, however, if the truck's brakes would ever go out, they would need to kick into action and lock the tires up. That being said, we have conducted many internal studies that show how our springs yield over time and we have an engineer here that has developed his own formulas that show this. I wish I could be more help, but I was just assigned not to long ago to learn that product line, so more than 1 person is knowledgeable.

If I were you, I would look into safety brake springs and read about them. This should give you a good idea of what you are looking for. Also, if you are planning on making the print for your spring, make sure that you are very stringent on what you expect. This spring will have to be virtually perfect when you receive it.

If you are looking for a quick estimate. I was once told a rule of thumb for springs in constant compression.

1 year in the compressed state is roughly the equivalent of 1 million cycles from free to solid state. Having that will allow you to get a quick and dirty calculation using fatigue formulas. But I must warn you, that is not an exact ratio, merely an estimate...

As far as material, I would recommend a chrome silicon alloy with brake spring quality. That is the material we use most often and it tends to have the best quality.

 

[israelkk]

ShawnBozarth

"1 year in the compressed state is roughly the equivalent of 1 million cycles from free to solid state...". What do you mean by that? what is to fatigue analysis to do with relaxation analysis? Classic fatigue analysis doesn't calculate relaxation but the life to crack/break.

Chrome silicon alloy is an excellent material. However, Chrome vanadium valve spring material maybe even better. There are many more excellent materials. chirpa12 didn't give much information. It may be that a stainless steel is needed due to corrosive or marine environment.

 

[sbozy25]

hahah

well perhaps my wording was not as eloquent as it could have been. I suppose "fatigue" was effecting me. I don't think I read the statements above correctly. Some how I got fatigue stuck in my head. whoops...

Any way... As far as the spring in constant compression, yeah it will continue to relax over time. As you have said, it depends on how you initially calculate your stresses.

As far as the Chrome Vanadium valve quality, yeah that is a good quality material. But I have never really seen much of a difference in performance as compared to chrome sil. you might get a slightly better spring, but today's market makes chrome van. material rather expensive at times. Not to mention you have to most times buy a minimum order of around 5-10k lbs. I suppose if the $$ is worth the slightly better performance, then by all means go for it. Also, I am not sure if you would be able to get SST to last that long... you would have to have one beefy spring... I don't think we have even thought to give it a try. That and it is expensive.

To answer the initial question though. If I were you chirpa I would contact a spring company that is used to making springs for this application. Tell them exactly what you want, and ask them what they recommend. I think you will be pleasantly surprised with what you find out.

 

[israelkk]

ShawnBozarth

All I wanted to say is that there is no point going into specific material when there is no information on the application. As to stainless steel group it includes, 17-7PH, Custom 455, Custom 465, Elgiloy, etc. which are new superior materials far beyond the classic AISI 301,302 or 304.

As to your offer to contact a spring making company you may be surprised to find that although many can manufacture springs according to drawings not always they can correctly design springs. Only few may have the knowledge and time to correctly solve problems in fields such as aerospace. Most of them do not have the expertise and know how including the spec requirements.

Most of them have tools and experience to check and validate a spring design but they do not posses the wide view and expertise necessary to correctly tackle the issues.

Instead, I suggest chirpa12 to find a spring design expert or an experienced engineer with a proven experience in the specific field (aerospace, military, automotive etc.). The spring manufacturer should be consulted for the aspects of manufacturing.

 

[chirpa12]

Thanks for your insights. That is the problem I'm facing Israelkk. The spring manufacturer that my company is dealing with doesn't appear to have much engineering background. They are more like salesmen who just punch numbers into a DOS-based spring calculation program.

The spring I'm attempting to design is for using a medical device that may be shipped to any part of the world so we've decided to go with 302 Stainless for corrosion resistance purposes. I'm trying to optimize it to be as small as possible because the device height depends on how small I can make this spring. I've used the suggestions you've all mentioned and I'm afraid I can't find a solution due to high stresses. I have learned a lot from the help you've all given. Thanks again

 

[israelkk]

optimizing a spring to be as small as possible and at the same time keeping the relaxation within a specified limit is exactly my expertise see

http://israelkk.googlepages.com/home.

I am pretty sure 302 is not your only option. A spring design program either DOS based or Windows is not the issue. There are many excellent DOS based program that no Windows based program can match them. I myself use many DOS based program for Solenoid and electromagnets design, Gear design and dimensioning, Spring Design, etc. However, programs are merely tools. What you do with the tool, knowledge and experience are what matters.

see

http://israelkk.googlepages.com/home

 

[sbozy25]

Ok, I see what you are getting at IS. I suppose that I have been spoiled since I went from an automotive company to a spring company. I suppose our company is strange, we have 4 engineers who are degreed and well qualified in spring design working here. One of these guys specializes in brake springs which are very high stresses and long periods of compression. That is what I was saying, perhaps Chirpa should contact a brake spring manufacturing company and get a hold of what calcs they use. Perhaps it could help?