Compressive Pre-Load (bolted member) 3

[MechEngineerNT]

Hello Everyone,

I am setting up a test for a hydraulic press in which I need to measure the force acting upon the face of the object I am pressing so I can get a pressure reading. So what I have done is inserted a 'force load cell washer' within the press piston (2 piece) to get a force reading. However, I need to secure it, so I am fastening the two peices of the press together, which therefore produces a compressive preload already on the load cell. So my question is, In order to get an accurate measurement of the force acting upon the product, can i do the following:

Force on object = Force load cell reading (Newtons) - Force load cell initial reading (Newtons)

Thank you,

 

[rb1957]

that would seem to make sense ...

another day in paradise, or is paradise one day closer ?

 

[EngineerTex]

No. If the load cell setup is not orders of magnitude stiffer than the bolt, then you have to take into account the following:

1) The elasticity of the bolt​

2) The elasticity of the load cell setup​

First, I think you are saying that there is a hydraulic cylinder rod that has a threaded blind hole on the end into which you have a washer-style load cell, which is retained by a bolt that is holding the cell in place. You tighten the bolt (do you fully torque the bolt, or do you simply tighten it a little so that the load cell is retained?) and then you press the ram against the part. So is the head of the bolt contacting the part that is being pressed or do you have a setup that has the head of the bolt going into a recess so that the load cell is in contact with the end of the ram on one side and the part being pressed on the other?

If you have a lot of preload and the load cell directly contacts the part:

1) When the force is less than the preload:​

Force on object is statically indeterminant. Without knowing exactly how the system is set up, we can't say for certain what the exact relationship is.​

2) When the force is equal to or greater than the preload:​

Force on object = Force load cell reading​

If I understand the problem correctly, the way you have it set up is that you will see almost no change in your reading as long as you are working below your preload. Once the load is sufficient that the bolt has been relaxed as a result of the load cell deflecting enough not to be experiencing any of the preload anymore, the load cell will be reading solely the load from the press because the bolt is not in contact with the load cell anymore.

Engineering is not the science behind building. It is the science behind not building.

 

[BrianE22]

Can you submit a sketch? I believe EnginnerTex is thinking like I am - you might have a relative stiffness (2 load paths) problem.

 

[MechEngineerNT]

Thank you gentlemen. I have changed the setup to a floating piston now to make only one load path thru the force washer. So with only the weight of the piston, my setup will follow the following formula? Or does this follow the same

Total Force Reading on Washer = Weight of piston + Press force = X. So to get a more accurate reading I can subtract the weight of the piston, or does the initial weight of the piston act as sort of a preload, that when the force becomes greater than the weight, that is the actual reading.

Thank you again for the help!

 

[EngineerTex]

It would really help if you had a sketch or a photo of this setup. I am currently thinking of the type of press where the cylinder is attached (loosely or otherwise) to the top of the frame and the rod of the cylinder or the outer body of the cylinder either one is pointed downwards.

So: Is the extending rod pointing up or down?

With no friction in the seals, you have to look at which part is below the fluid. If the rod is pointed down, you have to include the weight of the rod. If the rod is pointed upwards, then it will be the weight of just the barrel, which is the entire cylinder assembly, minus the weight of the rod.

Imagine it this way: Let's say you ignore the friction of the seals (this is not actually insignificant, so I'm not sure that you will get the answer you're looking for and I think that you can only determine the friction of the seals through experimentation, remembering that it's going to be very difficult to determine whether you're dealing with static or kinetic friction, since the seal may or may not be moving as the rod extends and retracts over micro-inch distances). But think about an exaggeration of this setup where you have a rod that weighs a thousand pounds. If it is pointed downwards so that the rod is contacting the part, the weight of the rod (because of seal friction) may register anywhere between zero and a thousand pounds while the pressure inside the hydraulic cylinder would remain at zero the entire time. Only after the entire weight of the rod has been transferred to the load cell and the friction of the seals has been overcome can you subtract the weight of the rod. If, on the other hand, the rod is pointed upwards and the barrel of the cylinder is contacting the part, you would have to take the weight of the barrel, minus the weight of the rod into account.

All that said, I would think that it would be easier to have the cylinder underneath the whole setup, put your part on top of it and then put the load cell on top of that and have the other side of the load cell contacting the rigid frame. That way, you don't have to take into account the weight of any part of the cylinder.


Engineering is not the science behind building. It is the science behind not building.

 

[EngineerTex]

Hang on - I want to revise my last answer.

If you are only looking at the readout of the load cell, then the weight of the cylinder, rod, barrel, etc. would not play a factor into the calculation. It should just be the readout of the cell to determine the load on the part.

But that is only if I understand how you have it set up. Try to put up a sketch or a photo.

Engineering is not the science behind building. It is the science behind not building.

 

[rb1957]

have we come full circle, back to his original suggestion ?

pls a sketch or schematic would help us understand.

another day in paradise, or is paradise one day closer ?

 

[EngineerTex]

Perhaps. But I think his original suggestion had a bolt preloading the load cell.

I also think a sketch would help. Lots.

Engineering is not the science behind building. It is the science behind not building.

 

[MechEngineerNT]

ok. thank you here is a picture. It is a floating piston with an adaptor and the thing in between is the load cell. so i believe i would take into account the weight of the piston, even though this may be a negligible value. please update me!

thanks!

 

[MechEngineerNT]

ALso, please ignore symbols, these are just NX mating symbols.
THanks,

 

[tbuelna]

From your sketch, assuming the part being pressed is at the bottom and the hydraulic pressure is applied from above, I don't see how the bolts can apply a preload to the load cell. With the clearance shown at the spacer ODs and the gap between the washers and the flange, it looks like the lower half of the tool is free to drop down a bit.

 

[MechEngineerNT]

Tbuelna, yes i have recently changed the design because of this. So know to get an accurate load cell measurement i must subtract the weight of the piston.

 

[BrianE22]

I don't get the spacers under the screws. Are they screwed into the flange? I also don't know what you mean by "free floating piston". The term "hydraulic press" implies to me that you are applying a load through a piston rod and applying hydraulic pressure against the piston.

At any rate, it sure looks like there are multiple paths for the load to take to get from your upper rod to your lower rod. Your load cell will see only a fraction of those loads.

 

[MechEngineerNT]

BrianE22,

This is a piston the is operated by a hydraulic system. The force is coming from the top down, which is 1 load path thru the pressure trandsucer.

 

[rb1957]

the way i see the pic, load from the top will only go through the 2 squares (section of transducer) ?

the preload on the bolts is applied to the sleeves, and doesn't clamp the flange (what you call "floating" ?)

with no hydraulic load applied, the transducer should be reading the weight of the piston. why not just "tare out" the transducer, so it's reading zero with no hydraulic load applied ?

another day in paradise, or is paradise one day closer ?

 

[BrianE22]

You've got 2 rods on top of the load cell (inner and outer). If the outer rod reacts all of the load from the lower rod then you are o.k. If the inner rod reacts any of the load then you have a problem.

If you change the design so the bolts along the flange preload the load cell then you have a problem. If the bolts/sleeves bottom out in tension (not floating anymore) then you have a problem.

 

[tbuelna]

Like rb1957 notes, as the sketch shows without any hydraulic pressure applied and the lower piston resting on the part face, the load cell will only be subject to the mass of the upper piston and the hydraulic fluid within the cylinder minus any friction from seals. If the lower piston assembly is resting on the part face, its mass will produce a force on the part that is not measured by the load cell.

If the intent is to tighten the flange bolts just enough to create a preload force on the load cell to offset the mass of the lower piston assy, this might be a bit difficult in practice. If the mass of the lower piston assy is say 25lbs, and there are six flange bolts, you would need to torque each bolt to approx. 4lbf preload. Not easy to do with the size of bolts your sketch shows.