Safe limit of twist in a shaft

[strantor]

I consider myself "smarter" and more "practical" than an engineer.

Which one? ... never mind, I'm sure there are several.

I might count myself among them. I'm not a very good communicator. It is a known deficiency. I try hard to express things clearly, and I thought I did a decent job in my first post, but it seems the majority read something between the lines that wasn't supposed to be there. Not totally sure but from context cues I guess it came across as "hi, my name is strantor and I'm an idiot. I've destroyed 15 PTO shafts in the past year and I can't figure out why." Just read the parts in bold, or I can summarize: I'm building a PTO Dynomometer.

I want to make a torqe-sensing PTO shaft for my tractor. I don't want a pony brake dyno or anything similar, I want to measure PTO torque and speed while running actual PTO attachments. To do this I think the simplest way is to measure twist of the PTO shaft. So contrary to the norm, I want the maximum safe amount of twist. This will make my measurement less challenging. I've calculated some values for hollow tube and solid rod, and for my target maximum torque over a 5ft shaft I can get 20+ degrees of twist, but I have no idea if that's safe, or one the verge of turning into a pretzel. Is there any way to figure the maximum safe amount of twist?

To be clear, I'm not wanting a way to measure whether or not my shaft has a permanent deformation of twist. I want my shaft to twist, (as much as possible without permanently deforming) so that I can measure the twist in operation, and turn it into a running torque value. I just want to know how much torque I can put on any given shaft without damaging it or someone or something.

Because.

I have unconventional interests. The torque sensing shaft project is an interesting enough project for me to pursue in and of itself. But, it is a means to an end. My tractor is the lowest HP variant (40HP) of a line of tractors that goes up to 55HP. They are all mechanically identical. Same heads, same pistons, same stroke, same fuel pump, same turbo. The only difference is varying levels of handicap in the ECU. I'm going to hack it and I need a way to quantify my results. Planning to use the PTO generator I'm building as a variable load to put behind this.

 

[pontiac428]

@strantor Have you looked into getting a torsional load cell transducer? You could mount it inline with your PTO shaft yoke, send the signal into a dataq, and even send the stream to a smartphone via bluetooth. I have bought load cells and transducers in the past as industrial surplus on eBay for cheap, less than 1/10th of new. Programming an arduino is easy, but working with lab software like WinDAQ gives you more flexibility. Check it out.

 

[strantor]

@strantor Have you looked into getting a torsional load cell transducer? You could mount it inline with your PTO shaft yoke, send the signal into a dataq, and even send the stream to a smartphone via bluetooth. I have bought load cells and transducers in the past as industrial surplus on eBay for cheap, less than 1/10th of new. Programming an arduino is easy, but working with lab software like WinDAQ gives you more flexibility. Check it out.

Thanks, I looked into them briefly but but was immediately soured by the cost. Even on eBay they were expensive, but deals do come up on ebay from time to time. I will keep my eye out.

Problem with ebay is, nearly all of the sellers are lazy and only put a part #. C'mon! List the friggin specs! I don't have time to track down a data sheet for every torsion sensor on ebay. They put that PN on there and list it for some unreasonable price, fishing for that one person searching by PN because they're only confident to replace a part with another one of the exact same PN. Probably a good tactic actually, I imagine it works out nicely every once in a while; often enough to matter, if you have a lot of surplus to sell.

 

[Weldingrod1]

We don't think you are an idiot! Most of us interpreted this as a "I WANT to build a torque transducer " and let me tell you, we are all about WANT to build something here! Lots of (fill in the blank) range and plenty of "why would anyone want to build that?" scope. Lots of us build things because we can or to obey the inscrutable exhortations of our souls, or ??? Myself HIGHLY included! I'm finishing a Magnabend scratch build right now. I could have bought one, but building it was sooo much more fun!

Sent from my SM-G892A using Tapatalk

 

[pontiac428]

Yes, torsion transducers are more expensive than pressure transducers, and there are many different styles as well. For a rotating shaft, you would also need a slip ring to commute the revolving wiring terminal to a static breakout. I think a slip ring could be shop made without too much heartache, but buying commercial might be dear. There are a number of surplus dealers out there, and some can be real gems. I think you could find the hardware with persistence. My last dataq project only cost me $60 or so for a used Honeywell NASA-grade load cell and a Hitachi linear transducer. The modest 8020 framing for the project cost more than the electronics and acquisition software combined.