Sizing Calculations

[Anonymous]

I am aware this is not exactly the place to post about this but I have nowhere else to turn that I know of. 

We manufacture valves and a few of them are operated with a threaded stainless steel round bar shaft. Basically I am trying to figure out how to mathematically select the appropriate diameter of this shaft for the application. Until this point, we have been using a certain diameter for a selected range of sizes. We always boost the diameter for safety, and during the engineers review, we always pass. The engineers aren't exactly going to tell us how to make our design better so I have been trying to create a spreadsheet to type in certain parameters and hopefully pop out the correct diameter shaft required. 

All I know at this point is the calculated thrust and torque required to operate our valve. I do not know how to translate this into what I need.

Sorry for wasting your time Inventor users but I figured there has to be someone that uses inventor that has the mechanical background to give me a hand with this! Any help is greatly appreciated.

Thanks

Sandro

[graemev]

First off, get an engineer involved! Third party will do if you can't afford to hire one on full time.

For a very basic rule of thumb, use the minor diameter of the thread for stress calculations, not the major or mean diameters. The true problem is somewhat more complicated than that, but it's a good start.

Also, get an engineer involved!!!! The cost of engineering is miniscule compared to damages paid after a finding of fault in a court case. I cannot emphasise this enough! Legal liability is a real thing and you won't be exempt because you "didn't know better." Education looks expensive, right up until you've taken ignorance out for a spin.

[Anonymous]

I truly appreciate your response graemev. I know enough to use the minor diameter of the thread for these calculations although I wouldn't even be able to tell you if the threads would shear off or the stainless steel would fail in tension or compression. 

I totally agree with you as per your engineer comments. We use third party engineers to look at our work most of the time and use that information for similar applications. This however does not mean we are using the most economical choice, it just simply works. I am not sure if I could find an engineer willing to create a spreadsheet for me to calculate this ourselves in the future. If I can't, I am stuck figuring this one out on my own. 

I somewhat feel like given the right parameters, this could be a very simple calculation so having to run it by an engineer every single time is an unnecessary expense. 

Thanks for the advise anyway! If you could point me in the direction that would be great!

Thanks

Sandro

[DRoam]

Hi @Anonymous, here's how I would go about this:

1. Size the rod diameter so it can withstand the maximum moment it could potentially undergo. Give it a reasonable safety factor (at least 1.2) against YIELD (not failure) at the maximum expected moment.
2. Determine the thread engagement length necessary to avoid thread strip-out (i.e., such that the threads are not the weak link in the rod's strength)

Here's some additional details on those:

Step 1:

Assuming there is no axial force (i.e. pulling or "stretching") on the rod, and no torsional (i.e. twisting) force on the rod, your stress will simply be from the bending moment from someone applying force to the end of the rod. The bending moment is the greatest at the base (where the rod connects to what it's threaded to). Mathematically, the stress due to bending is:

s_b = M*c/I

where M is your moment, c is the distance from the neutral axis to the point in question, and I is the area moment of inertia of the rod's cross section. The safety factor at a given stress is calculated as:

N = S_y / s_b

where S_y is the yield stress for the material. 

Doing some substitution and re-arranging, you can solve for the required diameter to achieve a specified safety factor. In Excel terms:

D = (N * (32 * F * L) / (PI() * S_y)) ^ (1/3)

where...

 N = your safety factor

 F = the expected applied force on the end of the rod

 L = the length from the base of the rod (where it meets what it threads into) to the end

 S_y = your material's yield strength (36,300 psi for steel)

Step 2:

I always use this reference for my thread strength calculations: Calculating Thread Strength

If you're using a thread size not listed on that page, you can find just about any thread's dimensions here: UNC Bolt Calculator. Down at the bottom you can also choose UNF or even non-standard UNS (special) threads.

With a little effort you could put together an Excel spreadsheet that will calculate the required thread engagement length for a given diameter, thread spec, and material. But it might be easier to just do it by hand each time.

=========================

That's just a quick look at how you might go about this. Like @graemev said, it would be a good idea to hire an engineer to do this for you to reduce the dangers of liability (speaking of which, everything in this post is for reference only, and I am in no way implying that following these recommendations will result in a safe design).

... But it's a good place to start and the theory is sound. The only oddity is the fact that you're not putting the threads in pure axial tension, you're subjecting the threaded length to a moment, which may require you to take some other stresses/calculations into account. But I would think that following the guideline I linked to for thread length would be sufficient, especially if the lever is only being operated by hand.

Hope that helps!

[DRoam]

Oh and also, for future reference, the following forum is a great place to go with any future generic engineering questions: Eng-Tips.com Engineering Forum. It's free and has an active, helpful community.

Best of luck!

[TheCADWhisperer]

---

@Anonymous wrote:

...I somewhat feel like given the right parameters, this could be a very simple calculation so having to run it by an engineer every single time is an unnecessary expense. ...

---

Once you have a digital model set up that correctly represents the physical model - as long as you are not significantly changing the model - it should be a once-and-done problem that is then just a matter of changing parameters (within limits).

[Anonymous]

Thank you very much for the detailed response DRoam. I am getting incredible side tracked from this with active work so I cant give this my full attention at the moment. After a quick scan of your response i feel it may be a good start to solving my problem however, I may be wrong but I do not feel like it is a moment force acting on the shaft. 

The valve operates strictly with vertical motion. a threaded shaft will require force to pull the valve open (tension, pulling) and then force to close the valve (compression). This leads me to believe it is indeed axial force. I am unsure if this changes things but if it does I would love some advice. 

Also thank you very much for the resources! I will definitely be checking them out when I have some time.

Ill try to explain the actual application a little better to help understand...at floor level there will be an actuator which holds a threaded brass nut. Through this nut is a stainless steel round bar shaft which has a threaded portion just long enough to allow full valve travel. Down below, the shaft is bolted to the door of the valve which is wedged into its seal to stop water flow. So, when the actuator kicks on, the nut inside turns and pulls the shaft and valve door upwards to open, and in reverse to close. I am strictly concerned with sizing the diameter of the shaft required to do the job whether that is dependant on the yield strength of the stainless steel, the threads  on the brass nut, or the threads on the shaft, I have no idea. 

I really appreciate the help though!

Thanks

Sandro

[DRoam]

Ah ok, I misunderstood your application. I was picturing a rod being used as a lever to twist a valve open, rather than using the threads on the shaft as a sort of screw jack.

In that case, the threads themselves will definitely be a more primary concern. And the brass nut is most likely going to be the weak link there since brass is much, much softer than stainless steel.

When you get the chance, could you post a picture/drawing of how the bottom of the rod attaches to the valve door? That's the only part I'm not clear on. Once I know that, I may be able to help you now that I understand your design a little better. Sorry for the misunderstanding!

[Anonymous]

please don't apologize! I am very thankful you are even attempting to help me resolve this. Basically there is a clip welded to the end of the shaft with a couple holes that will bolt up to the door. For all intensive purposes just imagine that it is welded to the door. If you still need a picture then I could create one and send it to you in a private message.

Thanks again!

Sandro

[graemev]

---

@Anonymous wrote:

I truly appreciate your response graemev. I know enough to use the minor diameter of the thread for these calculations although I wouldn't even be able to tell you if the threads would shear off or the stainless steel would fail in tension or compression. 

---

As a rule of thumb, the length of thread engagement should be at least as long as the thickness of a standard nut for the same size screw. Bolts usually fail in tension prior to stripping out of a nut.