Fluid-like Impact

[pbc825]

This is my first time, so please be gentle.

I need to estimate the impact force acting on a conveyor system at the loading point. The system continuously receives a granular product dropped from 3' with an initial velocity of 0 ft/second (the product comes from another conveyor system). The product is typically composed of gravel sized particles. There may be some boulder sized particles weighing as much as 300 lbs in the flow sporadically. The maximum mass flow rate is 42 tons per minute.

Two questions:

1) I understand the steady-state "fluid" flow force to be approximately 1936N or 435 lbs (Force = mass flow rate [635 kg/sec]* velocity [4.24m/s given the drop]) if we assume the product is a fluid; however, is this still accurate during start-up, or is there an initial transient state with higher force?

2) How might we estimate the boulder force? Do you recommend assuming a set impulse time and average the force over this time? Should I assume a deceleration distance and kinetic energy? Should I superimpose the boulder impact force with the stead-state fluid flow force?

Any advice would be appreciated.

Thank you in advance.

 

[Archie264]

Well, since no one else has chimed in I'll throw in my two cents worth. First off, welcome to the forum, from a relative newbie myself.

In your first question you mentioned the grain mimicking fluid flow so you might want to search for thrust block calculations to get some guidance on that issue. The initial startup in particular imparts a "water hammer" and whatever force they tell you to use be aware that something hitting the system like that can shake things loose in a way that might otherwise seem hard to quantify. It's certainly nothing to be trifled with.

As for your second question, well, the mechanical engineers and the dynamics gurus may have other methods, but I would calculate the kinetic energy of the bolder (i.e., .5mv^2) and set it equal to the work done on the system, force x distance, and solving for the force. This would require assuming an initial distance and refining the calculation through an interritive (sp?) process. In other words, assume a distance and calculate the force associated with it separate from the system. Then check what deflected distance that force would impart upon the system and repeat this cycle to the degree of accuracy you need. And watch your units.

 

[PersonalProfile]

1. Have seen this used for operational state, however there are various other methodologies (Bruff / Bridgestone etc) prior to flow commencing (assuming this is relevant to your application).

2. Have seen the use of the "strain energy" approach.

Regards,
Lyle

 

[hetgen]

Check this thread Link. With so many unknowns you should be conservative on your design unless you do on site measurement.

 

[Compositepro]

Kinetic energy is not conserved in this type of process. Much of it is converted to heat. Momentum is always conserved. Force is change in momentum with time.

 

[Archie264]

I'm sure that's correct, but when putting one's name behind a design to resist impact it might be worth neglecting losses due to such things as heat, friction, sound wave propagation, etc. Structures can do some wonky things when hit. Without full knowledge of the system or a proven nonlinear dynamic analysis model at my disposal I'd tend to err on the side of being conservative.

 

[pbc825]

Thank you everyone. If we all can't think of the "right" answer from an analytical approach, then I'm sure the client won't mind some conservatism.

 

[IRstuff]

What are the typical particle sizes, relative to the boulders?

One thing you might consider, since you're treating it like a fluid flow is what in EE parlance is called "shot noise", which is the standard deviation of the nominal flow. If you know the particle size, you can compute the nominal number of particles per sec in the flow. The square root of that number would be representative of the standard deviation of the flow. This would at least give you some feel for the worst case of the "normal" flow.

TTFN
faq731-376

Need help writing a question or understanding a reply? forum1529