Contact Applied DEM at http://applieddem.com/. Their software will do everything you are asking. Contact Clint Hudson for pricing. ■

DEM business is not plug and play. I respectfully suggest that you ask a professional like Gary to run your simulation rather than risk crashes and errors.

There is also an AVESTA manual which gives some examples of the very subject which you are interested in. It only deals in the usual hard Swedish steel stuff but you might find it a useful background. There are at least 3 springs in series to be considered: 4 if you choose Trelleborg rubber body panels. ■

In addition to DEM Solutions response. The more detail the information on the product the more accurate the analysis will be, ie; moisture content, particle size distribution, insitu density (required for impact absorption of particles), etc. It is highly recommended to perform material testing to characterize the material in order to insure accurate simulations. Properties of materials can vary so drastically from location to location that it can only be estimated without having the material testing done.

From your information during registration it appears that you are a student. If this is so there are student versions of the software available at a much lower price. We also offer a free version of Bulk Flow Analyst with a very limited particle count that would allow yourself to get familiar with the software. Cloud simulation processing service is also available in a "PAY AS YOU GO" format. ■

Hi Danish, to summarize, it sounds like Discrete Element Method (DEM) simulation is what you are looking for. At DEM Solutions we create the DEM software package EDEM. EDEM is used by heavy equipment manufacturers around the world to design their truck body, excavator and other bulk material handling equipment. It provides engineers with a great deal of insight into loads on equipment, and material flow behaviours during equipment operation.

We have recently further enhanced the capability our software provides to engineers designing such equipment by launching our 'EDEM Add-In for ANSYS Workbench' . This takes the realistic material load forces and lets them be used directly in an FEA simulation to determine structural strengths etc. for equipment.

If you want to talk more about this – then please get in touch with us on info@dem-solutions.com

Thanks

The DEM Solutions Team ■

Originally Posted by johngateley

DEM business is not plug and play. I respectfully suggest that you ask a professional like Gary to run your simulation rather than risk crashes and errors.

There is also an AVESTA manual which gives some examples of the very subject which you are interested in. It only deals in the usual hard Swedish steel stuff but you might find it a useful background. There are at least 3 springs in series to be considered: 4 if you choose Trelleborg rubber body panels.

Good point about all the springs.

My mind boggles at what material properties might be measured in the laboratory to assist in simulating ROM ore dumping into a truck. Angle of repose? Really? Of course any such measurement would assume a constant material coming from the mine. Unlikely, I think. I can only imagine that the largest lumps will tend to dominate the effects in this situation.

If this is an existing operation, the possibility of taking measurements with an accelerometer attached to the truck tray and perhaps at other points comes to mind. ■

Originally Posted by Danish

Thanks Gary for this useful information. I want to ask that are there any learning manuals available for Bulk Flow Analyst?

Thanks,

You can view our video tutorials here. http://applieddem.com/training/training-videos.aspx ■

Guys,

I think there is time you all started looking carefully at the mathematics and algorithms behind the development of DEM software. To say the least the developers make some heroic assumptions that based on the latest research and published data is fundamentally flawed. Once you get to complex materials i.e. ones where there is significant variation in particle size and worse, water present DEM software is useless. It will give you a reasonable approximation for free flowing, sized material but that is it. The promotion of DEM as a design tool and its use by so many designers is why there is a maintenance disaster in most bulk materials handling operations. Simply we have persuaded a generation of engineers to stop thinking, stop observing and follow the little round balls (sorry one has some odd shapes) and pretend this is good engineering. We have two recent posts on ongoing education in materials handling and one on engineers standing up and being counted as professionals. If we want to take this seriously then start questioning the mathematical logic of DEM instead of accepting it and you might find your designs may work a lot better. I speak from one who first looked at DEM software as a design tool in 1994 and while I accept the mathematics and algorithms have improved the complexity of what this software is trying to mimic as far as material flow requires enormous simplification and many heroic assumptions as there are a very large amount of variables.

By way of comparison, look at the design of an overland conveyor of some complexity. We have a number of accepted design standards. We also have about 5 key variables that require assumptions to be made mainly around various friction factors. Using the 5 most popular standard software you can get power demand variations of up to 20%. Now we have got better and refining these variables by manually inputting data based on our experience but just looking at the paper published a couple of years ago on the Worsley Alumina overland conveyor, the variations were still in the order of 15% and most accurate was based on a part empirical approach to the design calculation. This after 30 odd years of pretty intensive work. Now take a step into granular flow which is the realm of DEM and the number of variables is many fold that a simple overland conveyor system. If we cannot get the mathematics accurate where the variables we have to estimate are but 5, how accurate do you think we can be when the key variables are but 40. Guys stop putting your head in the sand and pretending DEM is some sort of savior. It is an evolving mathematical model that has some uses with facile materials but as of now all it is doing is leading young engineers up the garden path so to speak

Cheers and Merry Christmas

Colin Benjamin

Gulf Conveyor Systems Pty Ltd

www.conveyorsystemstechnology.com ■

While examining the behaviour of laterite ores in North Maluku recently I was presented with a design where the 90tonne dump truck stood an extremely excellent chance of dropping a 7tonne boulder onto anyone walking in the pathway below the grizzly. The design had quite reasonably started in 2D sectional view and taken it from there. I pointed out that smaller material would not slide away as quickly as large rock would roll down once they were exposed. When, nearly said if, the downstream sticky finer ore presented sufficient obstruction the boulders would roll sideways, over the truck body and at least severely damage the concrete floor some 18metres below.

So, a large part this thread is rather irrelevant in this context. Body strength is very well catered for by...bodybuilders...and we should concentrate on the possibility of load fluctuations in the first instance. Health and safety procedures forbid anyone walking across, or standing by, a vehicle loading operation. Operators accordingly walk around the blind side of the operation in the assumption that it is the safer side. Of course it is more dangerous because the guy, or doll, cannot see the rockfall coming. By all means examine the spread of ore in the truck body and in years to come, as the software develops, let those of us who are still around know the results of your study. Body redesign is a useless quest whatever software is presented. You have to design for the harshest condition with the material thicknesses available and then admit that the panels and stringers will get beaten to hell and back by awkward rock shapes and shovel impingement. Remember the robust canopy over the driver's head, although it shouldn't take any load and is therefore redundantly outside the scope of this exercise, isn't just stuck up there for fun!

Merry Christmas to all! ■

I am amazed that certain individuals feel threatened by the use of DEM in such a way that the cost to use DEM is so much cheaper than their method of scale modeling. One would expect that the customer would want a much cheaper alternative to visualizing whether or not their design will work instead of the major expense of building scale models. Yes DEM has some imperfections and may not be 100% accurate in some conditions but for most products it is a very useful and sought after tool for designers around the world.

The telegraph felt threatened by the invention of the telephone and it was not perfect at first. The telephone took time to become a household necessity. Do you see a telegraph in use today???

Have a Very Merry Christmas one and all and a Happy and Prosperous New Year. ■

Perhaps this thread could have been started in 'Engineering Software Aspects' so other experts might be encouraged to reply.

There is an ongoing conflict between protagonists of DEM and hard copy texts/scale modelling. I am not open minded on this issue for 2 reasons.

Firstly I cannot presently justify the expense of such software. By the time it becomes generally affordable I and many of my contemporaries might not be around to buy, or too old to care. In my book this equates to a fair chance of running blind for future generations. Secondly, as most replies mention, scale modelling and testing is out of the question in large scale operations. This is what keeps most practitioners in business. In the land of the blind the one eyed man is king.

We, as a body, risk ridicule if we admit that we are telling finance that a job will run into billions and then later on tell finance, and the Client, that we don't know what we are doing. It is not the same as the bloke who told his record company that the Beatles demo disk indicated they had no future. He made a very very expensive mistake but he could not be blamed for lost revenue because they never spent the money.

Many colleagues do a very good job at predicting plant behaviour with the existing tools. It is only when we try to improve the status quo that things can sometimes go horribly wrong. The most learned in this business are running into difficulties with overlands and tubes and although I am confident that they will overcome the problems and emerge with even more credit and respect the fact remains that the business is too big. It's easy to say this...so I'll say it...progress should make life easier for all. Whether or not DEM can address the acknowledged material misbehavior-ism within a reasonable time and cost remains to be seen. However scale modelling is agreed, within this thread, to be inadequate and so we are going to have to rely more on DEM for problem solving in the future. Or we can keep it simple: and get back on topic.

The largest lump to fall into the truck body is the largest that the digger can lift over the side. The smallest lump is the one which the shovel teeth can lift without rolling under. Smaller particulates will adhere to this bottom size but if the base lump rolls under then the debris doesn't count. That sorts out the spectrum and the definition of the largest lump is borne out by the shovel operator who won't be congratulated by other face workers if they have to consistently break down rocks which he can't get into the trucks. So if the thread starter manages to develop a truck body which will sustain an absolutely stable operation then the digger driver will be obliged to try and load a still bigger boulder aboard and so the cycle persists. That's progress. ■

Hello,

My quick reply with some 15 - 20 minute thought is as below. Therefore do not consider it as solution. Readers are welcome to improve this.

In this type of situation, when bulk material is hitting any surface, the basic feature about solution can be as below:

As per Newton’s law of motion,

Force F = mass x acceleration.

Therefore, force F = mass x (Change in velocity) / (Time).

Therefore, force F = (Mass per second) x (Change in velocity)

Therefore, force F = (Mass per second) x (Initial velocity perpendicular to surface - Final velocity which is zero)

Therefore, force F = (Mass per second) x (Velocity perpendicular to surface), as a first solution.

There will be emptying time for bucket. The bucket material mass divided by this actual emptying time will give mass per second. The velocity you can compute as per drop.

When this impact force acts on the truck body, the spring supported body itself will deflect. So, you have to consider certain deduction in momentum which is transferred directly to truck, and thereby it will reduce the value of F as above. It will be a closed circuit solution.

Regards,

Ishwar G. Mulani

Author of Book: Engineering Science And Application Design For Belt Conveyors (new print November, 2012)

Author of Book: Belt Feeder Design And Hopper Bin Silo

Advisor / Consultant for Bulk Material Handling System & Issues.

Pune, India.

Tel.: 0091 (0)20 25871916

Email: conveyor.ishwar.mulani@gmail.com

Website: www.conveyor.ishwarmulani.com ■

Hello,

The maximum impact force exists when the front of a material stream touches the impact surface.

This is called the dynamic factor of an impact.

(also existing in earth quakes)

Impact-force formula

The impact force is given by : F = m * dv/dt

furthermore : m = vol * rho

vol/dt = v1 * A ---> vol = v1 * A * dt

dv = v1 - v2

resulting in :

m = v1 * A * rho * dt

Substitution gives :

F = v1 * A * rho * dt * dv/dt

or :

F = v1 * A * rho * dv

In case v2 = 0 then dv = v1 - v2 = v1



F = rho * v1^2 * A

Dynamic factor on an impact force

Energy delivered by F(impact) = Energy absorbed by structure

x=x(max)

F(impact) * x(max) = ∫ c * x *dx

X=0

F(impact) * x(max) = * c * x(max)^2

F(struct) = c * x(max)

F(impact) = * F(max)

F(structure) = 2 * F(impact)

Using

F(impact) = rho * v1^2 * A

Results in :

F(structure) = 2 * rho * v1^2 * A

The dynamic factor of an impact is :2 ■

Hello,

Referring to my earlier reply, some additional information is as below:

1) The arriving material front first slice of depth say equal to lump size can be considered to have direct impact on truck floor with rebound velocity like solid object (as per coefficient of restitution and instantaneous impact force accordingly). But at this moment, full quantity of material from bucket is yet to land on truck body. Such situation will occur when the material is falling on truck empty body. It will not occur if material is landing on earlier emptied material.

2) The governing situation seems to be by last material falling from bucket, for which rebound velocity can be practically considered as zero. But now the force will be sum of material weight already emptied by bucket + impact force (as per rebound velocity zero), as sum total maximum force (mostly) occurring during unloading process of bucket as one specific event.

Regards,

Ishwar G. Mulani

Author of Book: Engineering Science And Application Design For Belt Conveyors (new print November, 2012)

Author of Book: Belt Feeder Design And Hopper Bin Silo

Advisor / Consultant for Bulk Material Handling System & Issues.

Pune, India.

Tel.: 0091 (0)20 25871916

Email: conveyor.ishwar.mulani@gmail.com

Website: www.conveyor.ishwarmulani.com ■

Originally Posted by I G Mulani

Hello,

My quick reply with some 15 - 20 minute thought is as below. Therefore do not consider it as solution. Readers are welcome to improve this.

In this type of situation, when bulk material is hitting any surface, the basic feature about solution can be as below:

As per Newton’s law of motion,

Force F = mass x acceleration.

Therefore, force F = mass x (Change in velocity) / (Time).

Therefore, force F = (Mass per second) x (Change in velocity)

Therefore, force F = (Mass per second) x (Initial velocity perpendicular to surface - Final velocity which is zero)

Therefore, force F = (Mass per second) x (Velocity perpendicular to surface), as a first solution.

There will be emptying time for bucket. The bucket material mass divided by this actual emptying time will give mass per second. The velocity you can compute as per drop.

When this impact force acts on the truck body, the spring supported body itself will deflect. So, you have to consider certain deduction in momentum which is transferred directly to truck, and thereby it will reduce the value of F as above. It will be a closed circuit solution.

Regards,

Ishwar G. Mulani

Author of Book: Engineering Science And Application Design For Belt Conveyors (new print November, 2012)

Author of Book: Belt Feeder Design And Hopper Bin Silo

Advisor / Consultant for Bulk Material Handling System & Issues.

Pune, India.

Tel.: 0091 (0)20 25871916

Email: conveyor.ishwar.mulani@gmail.com

Website: www.conveyor.ishwarmulani.com

Dear experts,

If bulk material passes from one conveyor to another conveyor at 1800 TPH through a chute, which is having one meter of vertical height, how much would be the force produced on its receiving conveyor? Assuming that both conveyors are running at 2 meter/ second,

Force = ( 1800 Tonnes / 3600 seconds ) x 2 Meter

= 500 kg/second x 2 meter

= 1000 kgm/sec

= 10,000 Nm/sec ( approx.)



Is there any limiting factor of above force in numbers ? Would not this depend on exact material behavior also which is being handled ? The iron ore fines may have less force than un-calibrated lump ore, which can be very harsh on the receiving conveyor, when it vertically falls.

Kindly enlighten further.

Thanks & regards to all, ■

Originally Posted by Danish

Hello Everyone,

I want to know the impact force that the material puts on the dump truck body when that material is dumped by the shovel. And latter i need to simulate the material flow as well from shovel bucket into the dump truck body. Can anyone please guide me that which is the best software to do these stuff.

Thanks,

Danish

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

A Late Comment:

ROCKY DEM (http://www.rocky-dem.com/index.php?pg=release)can also resolve the impact force acting on all components. You will need to identify the suspension compliance properties to obtain an accurate measure. More shock absorption = less impact. However, you also need to know truck bed mass and other suspended masses over the springs or shock absorbers.

Moreover, it couples with ANSYS multi-body physics and stress analysis of related components, together with ANSYS computational fluid dynamics (CFD) (double coupled between fluid and rocks = resolves dynamic fluid force on rock and rock for on fluid). Very fast code using both CPU and GPU.

ROCKY is:

1. much faster ( up to 20x in some cases on solid particles) than other commercial DEM packages, Solid particles can use inexpensive GPU ( 2000 cores)

2. configures non-round rocks not spheres - polyhedrons

3. able to resolve rock breakage - using energy spectra, JKMRC T-10 protocol w/ A x b constants

4. able to handles millions of particles (rocks)

5. able to handle extreme size range with much less time computation penalty of other commercial codes

6. able to resolve dust turbulence, dust emission and dust control guidance - either using Lattice Boltzmann or ANSYS Fluent.

7. able to resolve sticky cohesive and adhesive properties between rocks and boundaries

DEMOS are possible. ■

Originally Posted by sganesh

Dear experts,

If bulk material passes from one conveyor to another conveyor at 1800 TPH through a chute, which is having one meter of vertical height, how much would be the force produced on its receiving conveyor? Assuming that both conveyors are running at 2 meter/ second,

Force = ( 1800 Tonnes / 3600 seconds ) x 2 Meter

= 500 kg/second x 2 meter

= 1000 kgm/sec

= 10,000 Nm/sec ( approx.)



Is there any limiting factor of above force in numbers ? Would not this depend on exact material behavior also which is being handled ? The iron ore fines may have less force than un-calibrated lump ore, which can be very harsh on the receiving conveyor, when it vertically falls.

Kindly enlighten further.

Thanks & regards to all,

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

As with truck body, ROCKY DEM can resolve all forces acting on belt in 3-planes (down, forward, lateral). Impact on belt axially can damage belt and either apply a speed up or retarding axial force as well as apply a miss-tracking force that are quantifiable and thereby predict degree of miss-tracking. ■

TO further clarify the granular flow stream force behavior onto belt you need to know:

1. 3-D belt modulus between support idlers

2. Type and properties of impact idlers

3. Suspension compliance properties of idler supports

4. chute geometry, liner system, and position wrt belt

This is not trivial but can be solved using multi-body dynamics code together with a coupled Finite Element code such as ANSYS, and a good DEM code.

Yes, particle size has a major influence on force. However, proper chute geometry can mitigate most, if not all, of the damaging forces. Minimize the vertical velocity as it hits the belt and match the belt forward velocity are today common practices used in conjunction with a good DEM code. ROCKY does this daily. ■

Hello,

Flow rate is kg/sec which can be easily calculated by mtph x 1000 ÷ 3600. In your case the conveyor is discharging material at 2 mps. This velocity will mostly increase during material travel in transfer chute (as an overall effect of gravity and interim impacts in chute as vector sums). Find the material velocity at the exist end of feed chute. This velocity component perpendicular to receiving belt will be causing impact. If material size is small compared to flow cross section, then the calculation as per formula will be governing. For excessive lumpy material, calculation as above, and in addition analysis is done for lumps falling on belt.

Refer literature for calculating material velocity in chute.

In properly designed chute, the chute shape / arrangement is designed to get the required output velocity. Unfortunately, often chute is simply drawn as a convenient enclosure connecting material inlet to outlet. If the plant buyers start asking material velocity in chute, then its related practice will become widespread, to distinguish quality of engineering, and to promote better performance.

Regards,

Ishwar G. Mulani

Author of Book: Engineering Science And Application Design For Belt Conveyors (new print November, 2012)

Author of Book: Belt Feeder Design And Hopper Bin Silo

Advisor / Consultant for Bulk Material Handling System & Issues.

Pune, India.

Tel.: 0091 (0)20 25871916

Email: conveyor.ishwar.mulani@gmail.com

Website: www.conveyor.ishwarmulani.com ■