Seems like a lot of people have heard about this stuff and are looking for something to do with it.

Nickel is not friendly.

A grab is the best bet, feeding onto a chain. Make sure the stuff doesn't touch the receiving chute sides or you'll be digging it out for ever and a day.

By the way, where did you find your units? 2.4 decitonne per hour is 240kgh^-1. ■

Originally Posted by BabakHoudeh

Thanks John. The ore is sticky iron laterite with 24% moisture content. I should be more clear on the units, dtph refers to dry tonnes per hour.

Some does not recommend the chain conveyor. They believe because of the ore stickiness and tendency to agglomerate it will not drop out from the flights. Same problem using a screw conveyor. Looks like flat belt conveyor equipped with good scraping devices at the head end of the belt would be an appropriate option.

-------------------------------------

Should considered optimized withdrawal slot, shear opening, and skirtboards as well. CDI just completed an upgrade of an existing belt feeder that eliminated skirtboard damage and reduced power draw with success. All was validated with ROCKY or DEM code. ■

Larry,

With all due respects I have seen a large number of DEM simulations and as you are aware I am somewhat familiar with Rocky. Your claim that you can simulate the flow of "sticky" aka cohesive/adhesive materials needs to be scrutinized as all the literature that is published questions/denies that the flow of such materials can be computationally described, one recent publication describes the challenge as "beyond current science". DEM programmes do not have the facility to input the unique flow characteristics of micro fines and ultra fines with variable water contents that is characteristic of "sticky" materials. What I have seen is DEM proponents video or observe such flow and then manipulate the software to duplicate what they have seen but this is not a scientific approach nor would we expect that such an approach can be projected into other applications so I have no faith in your claims that what you designed was "verified" using DEM unless by chance. Our approach which we have documented and which we have subject to open scrutiny has however proved very accurate and definitely does have a creditable and logical base.

Cheers

Colin Benjamin

Gulf Conveyor systems Pty Ltd

www:conveyorsystemstechnology.com ■

Originally Posted by Colin Benjamin

Larry,

With all due respects I have seen a large number of DEM simulations and as you are aware I am somewhat familiar with Rocky. Your claim that you can simulate the flow of "sticky" aka cohesive/adhesive materials needs to be scrutinized as all the literature that is published questions/denies that the flow of such materials can be computationally described, one recent publication describes the challenge as "beyond current science". DEM programmes do not have the facility to input the unique flow characteristics of micro fines and ultra fines with variable water contents that is characteristic of "sticky" materials. What I have seen is DEM proponents video or observe such flow and then manipulate the software to duplicate what they have seen but this is not a scientific approach nor would we expect that such an approach can be projected into other applications so I have no faith in your claims that what you designed was "verified" using DEM unless by chance. Our approach which we have documented and which we have subject to open scrutiny has however proved very accurate and definitely does have a creditable and logical base.

Cheers

Colin Benjamin

Gulf Conveyor systems Pty Ltd

www:conveyorsystemstechnology.com

--------------------------------------------------------------------------------------------------

With Due Respect Colin:

CDI has repeatedly and successfully corrected sticky ore problems that consistently plugged chutes on many continents. I do not believe you are up to date. The key word is "somewhat". I do not recall ever sharing our successes with you.

Peter's ROCKY code is over one year out of date, as I recall. I believe that version did have adhesion and cohesion included.

I stated in an earlier email, I see no advantage in debating you or showing you where you are wrong, since you do not offer a quid pro quo assessment.

I see where you claim we have not published on this subject, which is true. I also note you have not published any details of your claims that others can verify and acknowledge your scientific approach. Why not?

I would have taken your interest seriously, if you were serious in knowing what can be done.

We have sufficient body of working sites that attest to our claims. We are always refining our insights to the many rheology models that need to be developed.

Micro fines can be calibrated into ROCKY. You just need to know how to do it. Maybe you should try to update ROCKY and then make a plan. I did tell Peter to be polite, if we were to continue in a positive relationship.

You diss the idea of calibration. Why? This is the essence of what is needed to model granular flow. I claim moving material through a laboratory apparatus is just that, calibrating. It plugs, you change the features, it plugs, you change the features till it works. Only difference is we can do the same with a computer.

I would be careful of your claims of science. I doubt you have any functional rheology based material mechanics in Ab initio. Prove me wrong. I and many others read your writings and wish to be enlightened by your advances and thereby offer due homage - on Zeta Potential, Einstonian Forces, Van der Waal Forces, various Liquid Bridge models, from nano to micro to macro conditions. ■

Originally Posted by nordell

--------------------------------------------------------------------------------------------------

With Due Respect Colin:

CDI has repeatedly and successfully corrected sticky ore problems that consistently plugged chutes on many continents. I do not believe you are up to date. The key word is "somewhat". I do not recall ever sharing our successes with you.

Peter's ROCKY code is over one year out of date, as I recall. I believe that version did have adhesion and cohesion included.

I stated in an earlier email, I see no advantage in debating you or showing you where you are wrong, since you do not offer a quid pro quo assessment.

I see where you claim we have not published on this subject, which is true. I also note you have not published any details of your claims that others can verify and acknowledge your scientific approach. Why not?

I would have taken your interest seriously, if you were serious in knowing what can be done.

We have sufficient body of working sites that attest to our claims. We are always refining our insights to the many rheology models that need to be developed.

Micro fines can be calibrated into ROCKY. You just need to know how to do it. Maybe you should try to update ROCKY and then make a plan. I did tell Peter to be polite, if we were to continue in a positive relationship.

You diss the idea of calibration. Why? This is the essence of what is needed to model granular flow. I claim moving material through a laboratory apparatus is just that, calibrating. It plugs, you change the features, it plugs, you change the features till it works. Only difference is we can do the same with a computer.

I would be careful of your claims of science. I doubt you have any functional rheology based material mechanics in Ab initio. Prove me wrong. I and many others read your writings and wish to be enlightened by your advances and thereby offer due homage - on Zeta Potential, Einstonian Forces, Van der Waal Forces, various Liquid Bridge models, from nano to micro to macro conditions.

------------------------------------------------------------------------------

"Our approach which we have documented and which we have subject to open scrutiny". I will stand corrected if you can offer the subject details by your claim of published details on rheology modeling that are "documented ". I have read a fair bit, but, maybe I missed something. Please offer the information on the subject of "Open to Scrutiny". Is is true you have publically offered your scientific principles with working proof? Let us read your foundations. I will truly apologize for any doubts. ■

Hi Larry,

A few issues to cover some questions. We have published and there is in "Technical Articles" a paper I wrote and access to one Peter wrote. We have also written a book albeit it only covers the principles of transfer design and is designed as a practical starting text. You further state that you have updated "Rocky in the last year yet so I presume that as Peter has a purchased copy of "Rocky" that his version contains the latest version. Having said all this I am pretty long in the tooth now and so questioning DEM and the design logic of others is not about any intent to gain extra mileage for my consulting work ( I am too busy as it is), it is about getting others to start realising that we have for too long been going down the wrong path on transfers and possibly bins and hoppers.

Over the last 40 odd years I have been asked to look at many thousands of transfers and like you started with the premise that what had been published was the correct starting point. The processes were well established and were basically seen as industry standards. The problem is that as I advanced my work on transfer designs things did not add up always. I was an early user of DEM (1994 I think), I used testing facilities at the Universities here, I sponsored a PhD student and other R&D. The book I wrote with Peter and others was in some ways about putting a peg in the ground from which we could go forward. Our focus over the last 10 or so years has been transfers handling cohesive or adhesive materials. The very large number of transfers we have looked at over this recent period do include transfers you have designed and in all cases the reason we were asked to look was because of various issues they were a problem for the site. If I am to summarise what where we are at;

1) We, like others focused on the angle of repose when designing transfers. With cohesive/adhesive ores we did simple tests to adjust the angle to reflect the reality that these ores hang up at steeper angles. Simply working around such a concept did not translate to what we observed. Reverting to the latest literature and research of others was enlightening only to the point that they acknowledged that such material flow is extremely difficult to predict. What we now know is that you can use the angle of repose to design a reasonably good chute for free flowing materials by way of creating an interference flow (e.g. the WEBA chute) but not if the materials are adhesive or cohesive. Such materials can build-up at vertical or over vertical angles in some instances depending on the chute design. We now understand the build-up mechanism and this was an important part of the jigsaw.

2) We, like others, started with the premise we must know the physical properties of the ores we are processing. The techniques are still in use today and they involve taking an ore sample, sizing it and putting it through various tests and from this giving a very complete analysis of the ore. We know and have known for a while that sizing ore and then doing tests that then are meant to represent the ores properties is a nonsense. It is even worse if the the ore has cohesive or adhesive properties as this invariably means that there is a reasonable proportion of minus 200 micron material in the ores. Let's look at just one aspect and that is moisture content. We get told the ore is 6-8% moisture content yet if we examine such ores and separate the lump from the fines you will find that the lump ore has maybe a moisture content of less than 1% and the minus 200 micron material maybe 15-20%. Simply water covers the surface area of each particle and the fines with a higher surface area for a given mass naturally contain more water.

3) In transfers and any dynamic flow environment lump and fines separate. Moisture laden fines separate even more. Why? Well it all gets down to the effective friction which is not really a friction but an aggregate of all the forces that impact on the way the granular material flows. Lump material will be more affected by gravity and material angulation, minus 200 micron material by liquid bridge forces and Van Der Waal forces. That is very different mechanisms are in play that control the way the material flows. Given this separation and the different forces in play dynamic flow in a transfer is not about some "coherent" flow mass but a series of different flow masses that interact but ultimately will create a very broad dynamic from what the lump does to what the fines do. So treating the ore we process as some sort of coherent, logically described entity is also a nonsense.

4) The literature describes the flow in transfers as "dense granular flow" or continuous flow whereas in bins or most hoppers as "quasi static flow" or stop, start flow. We dispute this as our observations and work shows that any interference to flow from intercepting the ore at the wrong angle or a ledge of some sort will create a "quasi static flow". What is the relevance, well when handling cohesive or adhesive materials the angles at which the material will build varies enormously in these two flow regimes.

5) We have designed chutes for materials with ledges and with smooth liners yet the build-up patterns and flow patterns after a short period of time are remarkably similar. Conclusion, coulomb and wall friction are not issues in transfer chute design.

6) What happens in transfers happens over time. What can be seen as a very minor interaction in the initial flow pattern can create a nucleation point for cohesive fines to build-up and eventually significantly change the flow patterns in the transfer and maybe cause the transfer to block. You cannot observe or test this in a 20-30 second grab on a computor. You need to do such evaluation over time.

In summary we believe the models used to develop much of the logic used today in transfer chute design are fundamentally flawed. Sure there are some transfers that work well using this logic but generally this is for free flowing material where there is far more scope to manipulate the design to create a good outcome. The problem is even worse if the material being handled is abrasive as the general approach to cohesive/adhesive ores is to basically create a steep sided box and then try to intercept the flow at the bottom of the box in some way so as to steer the ore a bit. Consequence is high abrasion, skirt damage and skirting issues, dust, spillage and belt damage and even then sometimes these chutes block. All great for the belt manufacturers, suppliers of belt trackers, suppliers of wear liners, impact beds. air cannons and the like and skirting systems. Our work has allowed us to avoid such pitfalls and as we design and put in more chutes using the principles we have developed, the more compelling the logic we have developed becomes and the more we can confidently question the approach that has been the norm and which has delivered transfers as the number 1 fixed plant maintenance problem in the Pilbara iron ore province. As I said in an earlier reply to you, this is not about DEM, it is about the whole approach to transfer design of which DEM is just an extension

Cheers

Colin Benjamin

Gulf Conveyor Systems Pty Ltd

www.conveyorsystemstechnology.com ■

"Micro fines can be calibrated into ROCKY. You just need to know how to do it. Maybe you should try to update ROCKY and then make a plan. I did tell Peter to be polite, if we were to continue in a positive relationship.

You diss the idea of calibration. Why? This is the essence of what is needed to model granular flow. I claim moving material through a laboratory apparatus is just that, calibrating. It plugs, you change the features, it plugs, you change the features till it works. Only difference is we can do the same with a computer. "

Larry, I don't believe that I have been impolite.

You can account for some aspects of micro-fines in DEM by various means. Paul Cleary did this in a paper on samplers, recently referred to here. In essence, he coated the ultra-fines onto the surface of larger particles and modified his particle interaction algorithm to account for this. That's fine. I think Paul does some excellent work. He is also very clear to describe the assumptions he has made and the limitations of what he is doing. This approach might be great with something like washed pisolitic bauxite, where the ultra-fines do indeed exist in the surface coating and the particles are close to spherical. Beyond that, it becomes increasingly more tenuous. In particular, one has to acknowledge that the ores flowing through many process plants are highly variable, so the concept of characterising them approaches absurdity.

I would like to focus on this particular point. "I claim moving material through a laboratory apparatus is just that, calibrating. It plugs, you change the features, it plugs, you change the features till it works. Only difference is we can do the same with a computer. I think this encapsulates the prevailing myth very well. From that, one would assume that if you take a scaled down version of the ore in question, and run it through a laboratory apparatus, you would have the ideal situation. Its just that you can do it better with a computer. I contest that view. Cohesive ores do not scale down and tests based on scaled down ores are not a guide to anything. I don't do tests on scaled down ores because I know that this is nonsense. It is quite a simple point to prove and Paul talks of it in his publications. Quite simply put, you cannot scale down a cohesive ore. End of story. As to some of the other stuff that you refer to, I suggest that you are grabbing words from Wikipedia without any understanding of their meaning. Things don't 'plug' as you suggest, in a couple of minutes. If that were the case, then the task would be simple. In reality is it far, far more complex than that and has almost nothing to do with bulk ore properties under quasi-static conditions.

Once again, I would point out to you that I am the person who has invested in Rocky, not Colin. If it is out of date and now useless, then I am disappointed. At the time it wasn't described that way. But the practical results that I am getting from physical modelling with the designs that Colin is doing are so far beyond the capabilities of any DEM software, that it isn't really isn't an issue.

I shun the whole concept of 'ore characterisation' and 'calibration' for the type of work that we do. Instead, we run the chute through the whole range of possibilities, from highly cohesive to completely free flowing. Interestingly, I am getting feedback that the proponents of DEM are offering the same thing. In fact, that is just what Paul does in his paper; he varies the granular Bond number, just as we do (though we have a few more tricks than that). Its just that we can do it all in real time, once the model is built. The starting point for building a model is the same sort of file used in DEM and the time for model construction grows shorter with the use of modern laser cutters and CNC machining.

In the long run, since DEM is not capable of including the fine particles themselves in the calculation, because of their sheer numbers, it relies on a set of assumptions and approximations. It is not, as you tend to imply, some form of well defined physics. It is simply an approximation of reality that relies on feedback from the full scale, real world not from feed forward data from a set of bench scale calibration tests.

My techniques too are approximations that rely of field feedback, except that I have recognised the nonsense of the bench scale ore sample tests for a long time.

At the end of the day, I provide a service to designers such as Colin. If I could provide something better, I would. That fact that he finds the results so useful and is having success from it is all good news to me. Where I can find a niche for DEM, I will use it. ■

Originally Posted by Colin Benjamin

Hi Larry,

A few issues to cover some questions. We have published and there is in "Technical Articles" a paper I wrote and access to one Peter wrote. We have also written a book albeit it only covers the principles of transfer design and is designed as a practical starting text. You further state that you have updated "Rocky in the last year yet so I presume that as Peter has a purchased copy of "Rocky" that his version contains the latest version. Having said all this I am pretty long in the tooth now and so questioning DEM and the design logic of others is not about any intent to gain extra mileage for my consulting work ( I am too busy as it is), it is about getting others to start realising that we have for too long been going down the wrong path on transfers and possibly bins and hoppers.

Over the last 40 odd years I have been asked to look at many thousands of transfers and like you started with the premise that what had been published was the correct starting point. The processes were well established and were basically seen as industry standards. The problem is that as I advanced my work on transfer designs things did not add up always. I was an early user of DEM (1994 I think), I used testing facilities at the Universities here, I sponsored a PhD student and other R&D. The book I wrote with Peter and others was in some ways about putting a peg in the ground from which we could go forward. Our focus over the last 10 or so years has been transfers handling cohesive or adhesive materials. The very large number of transfers we have looked at over this recent period do include transfers you have designed and in all cases the reason we were asked to look was because of various issues they were a problem for the site. If I am to summarise what where we are at;

1) We, like others focused on the angle of repose when designing transfers. With cohesive/adhesive ores we did simple tests to adjust the angle to reflect the reality that these ores hang up at steeper angles. Simply working around such a concept did not translate to what we observed. Reverting to the latest literature and research of others was enlightening only to the point that they acknowledged that such material flow is extremely difficult to predict. What we now know is that you can use the angle of repose to design a reasonably good chute for free flowing materials by way of creating an interference flow (e.g. the WEBA chute) but not if the materials are adhesive or cohesive. Such materials can build-up at vertical or over vertical angles in some instances depending on the chute design. We now understand the build-up mechanism and this was an important part of the jigsaw.

2) We, like others, started with the premise we must know the physical properties of the ores we are processing. The techniques are still in use today and they involve taking an ore sample, sizing it and putting it through various tests and from this giving a very complete analysis of the ore. We know and have known for a while that sizing ore and then doing tests that then are meant to represent the ores properties is a nonsense. It is even worse if the the ore has cohesive or adhesive properties as this invariably means that there is a reasonable proportion of minus 200 micron material in the ores. Let's look at just one aspect and that is moisture content. We get told the ore is 6-8% moisture content yet if we examine such ores and separate the lump from the fines you will find that the lump ore has maybe a moisture content of less than 1% and the minus 200 micron material maybe 15-20%. Simply water covers the surface area of each particle and the fines with a higher surface area for a given mass naturally contain more water.

3) In transfers and any dynamic flow environment lump and fines separate. Moisture laden fines separate even more. Why? Well it all gets down to the effective friction which is not really a friction but an aggregate of all the forces that impact on the way the granular material flows. Lump material will be more affected by gravity and material angulation, minus 200 micron material by liquid bridge forces and Van Der Waal forces. That is very different mechanisms are in play that control the way the material flows. Given this separation and the different forces in play dynamic flow in a transfer is not about some "coherent" flow mass but a series of different flow masses that interact but ultimately will create a very broad dynamic from what the lump does to what the fines do. So treating the ore we process as some sort of coherent, logically described entity is also a nonsense.

4) The literature describes the flow in transfers as "dense granular flow" or continuous flow whereas in bins or most hoppers as "quasi static flow" or stop, start flow. We dispute this as our observations and work shows that any interference to flow from intercepting the ore at the wrong angle or a ledge of some sort will create a "quasi static flow". What is the relevance, well when handling cohesive or adhesive materials the angles at which the material will build varies enormously in these two flow regimes.

5) We have designed chutes for materials with ledges and with smooth liners yet the build-up patterns and flow patterns after a short period of time are remarkably similar. Conclusion, coulomb and wall friction are not issues in transfer chute design.

6) What happens in transfers happens over time. What can be seen as a very minor interaction in the initial flow pattern can create a nucleation point for cohesive fines to build-up and eventually significantly change the flow patterns in the transfer and maybe cause the transfer to block. You cannot observe or test this in a 20-30 second grab on a computor. You need to do such evaluation over time.

In summary we believe the models used to develop much of the logic used today in transfer chute design are fundamentally flawed. Sure there are some transfers that work well using this logic but generally this is for free flowing material where there is far more scope to manipulate the design to create a good outcome. The problem is even worse if the material being handled is abrasive as the general approach to cohesive/adhesive ores is to basically create a steep sided box and then try to intercept the flow at the bottom of the box in some way so as to steer the ore a bit. Consequence is high abrasion, skirt damage and skirting issues, dust, spillage and belt damage and even then sometimes these chutes block. All great for the belt manufacturers, suppliers of belt trackers, suppliers of wear liners, impact beds. air cannons and the like and skirting systems. Our work has allowed us to avoid such pitfalls and as we design and put in more chutes using the principles we have developed, the more compelling the logic we have developed becomes and the more we can confidently question the approach that has been the norm and which has delivered transfers as the number 1 fixed plant maintenance problem in the Pilbara iron ore province. As I said in an earlier reply to you, this is not about DEM, it is about the whole approach to transfer design of which DEM is just an extension

Cheers

Colin Benjamin

Gulf Conveyor Systems Pty Ltd

www.conveyorsystemstechnology.com

---------------------------------------------------------------------------------------------------------

Since you have little understanding on what ROCKY can do, but, you claim to know we cannot do it, what is the point of the discussion?

I have entered into the dialog with a hope that it would enlighten the readers on various solutions to bulk flow in chutes - wet or dry. I believe we are deviating from this course. ■

Larry,

It has nothing to do with the capabilities of Rocky, it is informing the readers that we believe the fundamental approach used in the design of transfers and the evolution of this of various computational evaluation methods have serious and fundamental flaws and therefore cannot be as accurate as their proponents claim. This is reflected in the extremely large number of poorly designed transfers especially those handling cohesive or adhesive materials. Conversely we have developed a completely different approach that is producing extremely good results consistently and very predictably

Cheers

Colin Benjamin

Gulf Conveyor Systems Pty Ltd

www.conveyorsystemstechnology.com ■

Originally Posted by BabakHoudeh

What are the viable options for conveying sticky wet lateritic ore with 22% moisture and a top size of 125 mm at low throughput up to 2.4 dtph continuously from ground level to the third level of a pilot plant? Belt conveyor? Chain conveyor? any other option?

WHat is the particle size of this material? ■

[Colin:

Please enlighten us on what you are measuring and how you use them in Van der Waal mechanics. After you have repeatedly extold the virtues of using Van der Waal forces, I am sure the readers are very curious on your special insite. ■

Please go to "Technical Articles" on the Forum under "New Developments In Transfer Chute Design" for an insight on how we do this

Cheers

Colin Benjamin

Gulf Conveyor systems Pty Ltd

www:conveyorsystemstechnology.com ■

Hi there Babak..

We have handled numerous sticky materials on conventional belt conveyors.

Probably the worst is the power station fly ash mixed with water (about 17 to 20% H2O) which is mixed with coarse bottom ash to form a rough sticky grey goo, that sticks to the ceiling if you throw it upwards.

You can't even get it off your hands without washing it off.

However, we have successfully used conveyors over the last 30 years to handle it.

Got to know what you are doing though.

Cheers

LSL Tekpro ■

Hello,

I have not tackled with the particular mentioned material. However, general guidelines to deal with wet sticky materials are as below:

1) In case of belt conveyor, the material is conveyed on smooth plain surface (without corners and three dimensional cavities). Therefore, material sticking on belt would be least, as compared to any other conveyor (chain conveyor with apron pan, screw conveyor, etc.). So belt conveyor is having maximum self cleaning features.

2) The belt speed should be kept relatively more and the discharge pulley diameter should be least so that centrifugal force at discharge pulley will tend to throw off material forcefully. Centrifugal G-effect can be checked by vsquare / r.

3) The carrying idlers should be rubber lagged, and of least diameter (belt higher speed and and lesser diameter for more centrifugal force). Both these feature will result into continuous self cleaning of rollers.

4) Conveyor should be without snub pulley.

5) Return rollers should be of spaced rubber discs type, and thereby they are self cleaning.

6) Use UHMW polymer liners for skirt and chute which are resistant to sticking. Alternatively, use SS liners if material is containing heavy lumps. Design the chute for higher speed of material within the chute. The skirtboard and chute internal surface should be plain without protruding bolt-nuts.

7) Water cleaning pipe line and air cleaning pipe line should be placed along conveyor, for periodic cleaning (daily, weekly, etc. as per need).

8) Use appropriate type of scrapers.

The handling of wet, sticky and hygroscopic materials by belt conveyor is not unusual. Mined lignite is sometime wet and sticky and also has presence of sandy particles making it more abrasive compared to coal. Such lignite handling by conveyor is routine affair.

As for the feeder, high frequency vibrating feeder is likely to be resistant to sticking and would tend to be self cleaning. The next option could be belt feeder.

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 Colin:

I have read your article and Peter's article. I find no scientific detail in either that would lead a educated engineer on how to use your "Van der Waal forces" and how to calibrate the same. I do believe you are mixing physical concepts to wit:

First, Van der Waal force is an molecular electron dipolar imbalance that mainly occurs at very close "normal planar" particle to particle relationships, unless you are talking about moisture. When scientist talk about particle interactions and Van der Waal force, they refer to a dry contact and the molecular electron magnetic attraction with two molecules in a dipolar interaction. The concept is a physical state with probabilities and instabilities due to the flux of electron positions. It is normally a qualitative condition that cannot be quantified, or so I am told by the physicist on staff at Granular Dynamics. Maybe, if you are skilled in Quantum Stochastic solutions this can be found.

Second, Van der Waal force in water is basically the "liquid bridge" event you reference, but, you do it in vague terms. This force is the surface tension phenomenon that can be quantified as water and the relationship of particle sizes. This is very different than the solid-to-solid particle relationship.

Neither of these to relationships are well defined in your article. If you wish to add clarity to these comments, I sincerely welcome your input. ■

Larry,

You will note that in our article we stated quite clearly that the work was done empirically. It was the result of many years work and observation. Now simply we have found flow properties that we can use to design transfers with a great deal of reliability and accuracy. We then went to the literature and research of others to relate this to the known and projected science. We have our theories to explain what we believe is happening but other than what we have said in this article we have not extended this further in print. We do not need to for that matter as we have no interest at this stage in developing a computational model of what we are doing and we do not need one for our design approach. We are getting significant recognition now for our work and through this maybe we will find ways where through new partnerships we can extend what we have done further in the manner you describe. At the moment our focus is fine tuning and extending our knowledge in more practical ways so that the design process can be easily documented. I believe once we have a documented approach then others will look at the mathematical implications and then maybe over time a computational approach will be developed based on what we have done.

Cheers

Colin Benjamin

Gulf Conveyor systems Pty Ltd

www:conveyorsystemstechnology.com ■

.....Go

Gentlemen,

This thread concerns laterites, not software.

But, if DEM is developed to reliably study laterites etc. then why am I not able to buy a pirate copy here in Jakarta? I'll be looking again, in Mangga Dua Mall but I have no great expectations.Similarly there is much discussion about the merits of DEM and textbooks.Why not stop bickering and put it all together in the interests of progress and sales?

All I can say to the thread starter is that laterites do not convey well.

Recently I became involved in laterite nickel handling for a large French outfit who have been working with the crud for over a century and lovingly demonstrated their inabilities. In New Caledonia there was constant digging out of blocked feeders and chutes, incessant spillage, choked scrapers etc. etc.. Even worse problems were forecast for Weda Bay despite a considerable investment in very impressive specialsed test equipment. When shovelling and trucking was proposed the Australian Scania sales engineer remarked that the crud hangs up in tipper bodies. I later witnessed this on site visits where the truck payload was seriously compromised by the incremental accretions.

During my time with this fiasco I observed that the problem only occured when the material was allowed to touch the sides. As soon as the crud saw skirtplates; chute walls; reclaimer buckets and bins it was there to stay.

Overall processing improvement, eg. Ravensthorpe, might help with sustained effort. All that Babek wants is to raise 2.4 tonnes of dry material about 25m. If dry material is talked about then probably it could be dried at ground level and then the pilot plant might get a reliable feed. A pilot plant which cannot receive proper material throughput won't go far. ■

John,

"Recently I became involved in laterite nickel handling for a large French outfit who have been working with the crud for over a century and lovingly demonstrated their inabilities. In New Caledonia there was constant digging out of blocked feeders and chutes, incessant spillage, choked scrapers etc. etc.."

This is what the discussion is all about, we are successfully designing transfers for this type of material that do not block and to get there we had to examine and question the conventional logic and found it wanting. What we have found does have broader implications.

Cheers

Colin Benjamin

Gulf Conveyor Systems Pty Ltd

www.conveyorsystemstechnology.com ■

I have summarily explained the difficulties with wet laterite ore. In this thread's diatribe I have not seen the word laterite used once outside of myself and the thread starter. Unfortunately the preceding reply goes no further than "this type of material". If you meant to say laterites please do so and then try to give the thread starter some advice.

What has foregone is a bunch of claims and counter claims which contradict the experiences of a major international metals refiner who has been handling this stuff for well over a century and is hardly satisfied. ■

Laterite ores are by their nature either cohesive or adhesive if accompanied by water so rather than describe them as laterites we have lumped them into a more generic classification therefore the comments are relevant as far as transfers are concerned and this was the basis of our original response to a claim re using DEM for this type of material. As far as gaining some knowledge there are papers we referenced.

Cheers

Col Benjamin ■

Originally Posted by Colin Benjamin

Laterite ores are by their nature either cohesive or adhesive if accompanied by water so rather than describe them as laterites we have lumped them into a more generic classification therefore the comments are relevant as far as transfers are concerned and this was the basis of our original response to a claim re using DEM for this type of material. As far as gaining some knowledge there are papers we referenced.

Cheers

Col Benjamin

--------------------------------------------------

We also see you offer no scientific details. Recently, you did admit to doing it emperically. This is the same method used by most operators for over the many years plants have been in operation noted by John.

You go on to state that you have found a new method (now emperically resolved) that results in ?? First attempt at sizing a chute geometry = getting it right? Second attempt?.... No guesswork? Since your method is emperical. How do you resolve moisture: a) which is surface moisture; b) which is internal to the ore granulation? Only (a) matters until you apply significant pressure. "Significant Pressure" must be quantified such as in changes in cohesive force by testing as must "surface moisture".

You use of the term Van der Waal forces has not been explained. Just curious about the strong objection to our understanding of the term and your not explaining how you use it?

You get more respect if you offer some examples than denegrate others' works and then claim you have apriori knowledge. What part of the discussion do we observe that gives the reader insight into your improved technique/examples without you having to divulge your blackbox? Sort of a cause and effect analysis. ■

Originally Posted by johngateley

I have summarily explained the difficulties with wet laterite ore. In this thread's diatribe I have not seen the word laterite used once outside of myself and the thread starter. Unfortunately the preceding reply goes no further than "this type of material". If you meant to say laterites please do so and then try to give the thread starter some advice.

What has foregone is a bunch of claims and counter claims which contradict the experiences of a major international metals refiner who has been handling this stuff for well over a century and is hardly satisfied.

-----------------------------------

John,

You are right in we have not addressed the question.

Am I wrong in the units used? If he is referring to 240,000 kg/hr then we need to define the container size or other lift device.

At 2000 kg/cm/container, you would need 120 lifts/hr (2 /min) to lift 3 floors and dump. Probably you need twice this size to do one cycle per minute. This might get a little tiring. The point is a time motion-study is in order. A key question is what container did the material arrive in at the point to be hoisted? If you can connect it to a powered hoist and does it have a bottom opening device? If so, then you only need a receiving device on the third floor and have a strategy on how to get from this container to process.

Our experience with highly cohesive materials, in such containers, is to use a vibrator to cause the release of offensive residue and that the container has steep side walls. The question then will be: a) how long to vibrate; b) what frequency, and c) what amplitude. More data on container is then in order and so on, if we are to consider batch movement. ■

You never cease to surprise me with what you are prepared to say when someone disagrees with you but I won't come down to your level. Let me reiterate, we

" Developed our design theory empirically over many years"

" The process we now use to design transfers chutes is not based on guess work but follows a rigorous procedure based on this work and it has produced a large number of very different transfers that all work extremely well including we must stress, with materials that are highly adhesive/cohesive by nature"

" Based on the research of others there is a very sound theoretical (read scientific) base to our work"

" We use dynamic scale modelling as developed by Peter Donecker to evaluate some of our designs as it has proven far more accurate than DEM and also allows us to extend the evaluation over extended time periods"

" Based on our work over many years we do seriously question much of the logic used to develop the computational models used by others including you"

Finally in respect to the original question posed by the post, other than transfers the key to designing conveyor systems for sticky material is cleaning and this can be facilitated by selecting the correct cover compound for the belt, appropriate scrapers usually with water sprays (or a belt wash station in extreme cases).

Cheers

Colin Benjamin

Gulf Conveyor Systems Pty Ltd

www.conveyorsystemstechnology.com ■

Originally Posted by Colin Benjamin

You never cease to surprise me with what you are prepared to say when someone disagrees with you but I won't come down to your level. Let me reiterate, we

" Developed our design theory empirically over many years"

" The process we now use to design transfers chutes is not based on guess work but follows a rigorous procedure based on this work and it has produced a large number of very different transfers that all work extremely well including we must stress, with materials that are highly adhesive/cohesive by nature"

" Based on the research of others there is a very sound theoretical (read scientific) base to our work"

" We use dynamic scale modelling as developed by Peter Donecker to evaluate some of our designs as it has proven far more accurate than DEM and also allows us to extend the evaluation over extended time periods"

" Based on our work over many years we do seriously question much of the logic used to develop the computational models used by others including you"

Finally in respect to the original question posed by the post, other than transfers the key to designing conveyor systems for sticky material is cleaning and this can be facilitated by selecting the correct cover compound for the belt, appropriate scrapers usually with water sprays (or a belt wash station in extreme cases).

Cheers

Colin Benjamin

Gulf Conveyor Systems Pty Ltd

www.conveyorsystemstechnology.com

------------------------------------------------------------------------------

Colin,

I have no qualm with you not accepting DEM as a granular mechanics tool. I do object to how you do so in public, without any demonstration of why, other than "I know". I therefore ask you to show why/how our DEM code fails, publically. Peter has a year old version that, I claim can still illustrate cohesive actions. Not perfect, but by most accounts workable. ROCKY now has rheology that includes pressure sensitive cohesion. From your refusal to show why DEM is wrong, by, scientific example, I claim you are only advertising your skill by experience like any other operator. I also believe, there are many flow regimes you cannot scale accurately and model without trial and error. This implies you will not achieve success on the initial configurations. This is only my belief, since you do not provide evidence.

Yes, the dialog has not favored either party. I will cease after this last comment.

Peter’s method of scaling gravity is flawed according to major geomechanic studies. In order to accurately include gravity in scaled down models, the model must apply the necessary gravity force. This is done in a centrifuge. Would you like scientific evidence of this fact? ■

Originally Posted by BabakHoudeh

Thanks John. The ore is sticky iron laterite with 24% moisture content. I should be more clear on the units, dtph refers to dry tonnes per hour.

Some does not recommend the chain conveyor. They believe because of the ore stickiness and tendency to agglomerate it will not drop out from the flights. Same problem using a screw conveyor. Looks like flat belt conveyor equipped with good scraping devices at the head end of the belt would be an appropriate option.

I have had experience with mineral sands and mangrove mud transfers and belt transportation of both and it is not an easy issue to overcome. In both cases that are particularly relevant was the movement of mineral sands from around an extinct volcanic site and mangrove where the dug material had to be scraped and ejected by use of water & air blast from behind the steel faces of the ejection plate in corrugated form to release the block of sticky product.

The adhesive sandy volcanic burden contained the micro-fine particulates of mineral sands we were harvesting.The products were then transferred by means of a belt conveyor to the preparation plant. Again the cleaning apparatus was not a success in just a scraping manner so a specialised cleaning solution was employed which worked well.

In a second experience was the transfer of mangrove mud from the excavation area to the reclaimed ground 1,000 metres away was not an option for trucks and again a belt conveyor was utilized.

My solution at the time was the use of washing up detergent that was used to breakdown the stickiness and adherence capability and at that time was not an environmental issue. Initially, the foaming was an issue but research and the help of BP Australia 'Comprox' in those days was extremely helpful in producing a very low frothing agent and the concentration allowed a small volume application to the in-feed that reduced the 'Cling' attributes to be reduced to a point that belt conveyors could discharge the products with minimal carry-back past the cleaner 'Scrapers and washers' apparatus. The detergent was utilized at the loading area and also the excavator bucket was kept moist with the detergent in the water & air blast system employed. This was in the 1970's and research on environmental issues would now have to be undertaken for application under EIS[Environmental Impact Studies]. A word of warning, the inclination angle of the conveyor is much reduced and my experience was that the maximum fully operational angle was reduced to between 7 - 9 degrees due to intermittent feed and back slip of the product on a flat belt. I have not tried using a fluted or ribbed chevron or dimpled belt surface.

I look forward to hearing how you approach the task and your solutions to the issues. If a 'Cling Breaker' can be of use, I would also be chuffed to hear if it helps your situation.

Cheers

Les Dunn ■

Originally Posted by nordell

------------------------------------------------------------------------------

Peter’s method of scaling gravity is flawed according to major geomechanic studies. In order to accurately include gravity in scaled down models, the model must apply the necessary gravity force. This is done in a centrifuge. Would you like scientific evidence of this fact?

The method is certainly not flawed.

What is flawed is your perception that we are dealing with geomechanics.

For the last and final time, transfer chutes do not operate in the quasi-static flow regime that applies to geomechanics. They operate in the dense granular flow regime at overpressures that don't even approach those in geological systems. I am astonished that you do not appear to be able to grasp this simple fact and its implications, but then neither do those who market quasi-static tests for chute design or model calibration.

Froude Number scaling does not require gravity to be altered. It simply requires Froude Number to be kept constant. This isn't rocket science. In fact it is marine science. Ship designers have been using it for 150 years. No centrifuge required.

http://www.iwm.org.uk/collections/item/object/205191748

Apart from the fact that I have been successfully applying the technique for 20 years, the application of Froude Number scaling to dense granular flows is well documented in the literature.

Froude number scaling is also implicit in such things as the Beverloo equation for hopper discharge. That is controlled largely by the dense granular flow in the discharge arch. I don't have to go into a detailed explanation of this as I can (hoist you with your own petard and) conveniently point you to a paper by Hilton, Mason and Cleary, where they explain the history and physics of this and then use DEM to model hopper flow to demonstrate the validity of Froude Number scaling over a huge range of scales. There are dozens of other papers that come to similar conclusions by experimental means.

http://www.cfd.com.au/cfdconf09/PDFs/121HIL.pdf

This is just one simple example. In my paper I illustrate Froude Number scaling for all the different points in a transfer chute.

The move from non-cohesive to cohesive granular flows requires some additional considerations, but I will not cloud the water here by bringing them into the discussion again at this point.

Cleary et al are, to their great credit, quite open about the limitations of DEM. Some of these can be overcome by physical scale modelling. One such limitation is that DEM simulations have very restricted time ranges, so miss some complex processes that occur over long time frames. In the work we have done for Colin, this has been one critical factor.

Lastly, I will point out to you that your calibration tests for Rocky are conducted with scaled down ores in laboratory scale apparatus at normal gravity. This rather undermines your argument, I think. ■

Originally Posted by nordell

Dear Colin:

I have read your article and Peter's article. I find no scientific detail in either that would lead a educated engineer on how to use your "Van der Waal forces" and how to calibrate the same. I do believe you are mixing physical concepts to wit:

First, Van der Waal force is an molecular electron dipolar imbalance that mainly occurs at very close "normal planar" particle to particle relationships, unless you are talking about moisture. When scientist talk about particle interactions and Van der Waal force, they refer to a dry contact and the molecular electron magnetic attraction with two molecules in a dipolar interaction. The concept is a physical state with probabilities and instabilities due to the flux of electron positions. It is normally a qualitative condition that cannot be quantified, or so I am told by the physicist on staff at Granular Dynamics. Maybe, if you are skilled in Quantum Stochastic solutions this can be found.

Second, Van der Waal force in water is basically the "liquid bridge" event you reference, but, you do it in vague terms. This force is the surface tension phenomenon that can be quantified as water and the relationship of particle sizes. This is very different than the solid-to-solid particle relationship.

Neither of these to relationships are well defined in your article. If you wish to add clarity to these comments, I sincerely welcome your input.

"Second, Van der Waal force in water is basically the "liquid bridge" event you reference, but, you do it in vague terms. This force is the surface tension phenomenon that can be quantified as water and the relationship of particle sizes. "

Such confusion. I barely know where to start.

Van der Waals forces are not the same as liquid bridge forces. These are two completely different phenomena. I will leave you to further explore these concepts.

I will make it simple

It is you, Larry, who need to be able to calculate and quantify all the forces in the system.

Since you cannot include the particles concerned (those that are influenced by Van der Waals and liquid bridge forces) in your simulations, then you have a basic problem. You have to adopt an approximation. That is why I pointed these out.

These are physical forces. In actual physical systems, they occur all on their own. All we have to do is vary the total cohesive force, which we can do simply and continuously in either direction in real time. You, on the other hand, have to approximate them and vary them in discrete, tedious, individual, short range simulations. ■

Originally Posted by johngateley

I have summarily explained the difficulties with wet laterite ore. In this thread's diatribe I have not seen the word laterite used once outside of myself and the thread starter. Unfortunately the preceding reply goes no further than "this type of material". If you meant to say laterites please do so and then try to give the thread starter some advice.

What has foregone is a bunch of claims and counter claims which contradict the experiences of a major international metals refiner who has been handling this stuff for well over a century and is hardly satisfied.

I have been a bit dismayed that this thread has evolved as it has, which is why I have abstained for so long from posting.

I cut my teeth on laterite ores.

Moisture content is critical with laterites. In the plant where I first worked, there was a belt that carried it up to the top of a screening plant. At the bottom of the belt there was a man permanently stationed to watch the belt to observe any slide back. This often happened because of rogue water ingress. It is critical to keep water addition under control. You are generally working one side or the other of the Atterberg limits for the material, usually on the lower side. Any rogue water introduced to the system can cause you to suddenly operate on the other side of the limit and that can result in disaster. I have a lovely set of slides of these effects propagating right through a system from the mine to the port. Rogue water can come from many sources. One key thing is to interlock water sprays with belt loading, so that you don't have the belt carrying water when it isn't loaded. Another is controlling chute wash-down procedures. Operators sticking hoses into chutes and leaving them running is another thing that can cause big problems. Mine to process plant communications also contribute.

If you can keep the water under control, you can operate either side of the limit. It is the transition from one side to the other that causes the worst problems. For the case of elevating the ore, I would design for the upper moisture excursion case.

As far as modelling chutes for laterites goes, I first did this successfully in 1995 and 1996 and still have the videos. ■

Originally Posted by BabakHoudeh

What are the viable options for conveying sticky wet lateritic ore with 22% moisture and a top size of 125 mm at low throughput up to 2.4 dtph continuously from ground level to the third level of a pilot plant? Belt conveyor? Chain conveyor? any other option?

After this prolonged discussion it is quite likely that your pilot plant is already well under development.

Have you considered the belt washing possibilities as mentioned variously in some replies and can you confirm, or otherwise, the acceptability of extra water within the downstream plant? In my experiences in DRY bulk handling, water has always been the troublesome constituent: loathed in all but screening and milling applications,and even there the control of water was recommended.

Would you expect the additional washdown to encourage avalanches? If not, then you could wash with fresh water, or brine, according to your process requirements. As Roland H mentions, feedback would be helpful and appreciated. This is one of the more worthwhile threads, very process critical and deserves more background. ■

Dear All

We are manufacturers of conveyor belts . Please contact us in any needs

regards

Vivesea

vivsea@gmail.com ■

This late reply is not the kind of input or feedback that we were hoping for.

Please examine the foregoing discussion before offering trite sales babble. ■