Belt mistracking theory

David Pelc
(not verified)
Posted in: , on 19. Sep. 2002 - 23:04

Hello everybody!

To compile a part of my thesis, I´m looking for some materials concerning conveyor belt mistracking and training theory. Simply said, whats going on when the belt is sliding to one side of conveyor and why is it doing it.

If you have some idea or suggestion where to look for it, I'll be grateful for your notice.

Have a good day,


helenea - FLEXCO, USA
(not verified)

Re: Belt Mistracking Theory

Posted on 20. Sep. 2002 - 06:15

Hello David,

There are some excellent, very experienced conveyor design specialists contributing to this site and I hope some of them respond to you also. However this is a start,

There are many reasons that conveyor belts mistrack and there are training idlers that can be designed into the system initially to take care of this or they can be added in to an existing system. For long complex conveyors the alignment during installation may not be accurate enough. Also every aspect of the design affects the mistracking, eg. is the counterweight sized correctly? does material load centrally on to the belt? is the splice good? If it's mechanically fastened, then are the correct rated fasteners being used? Sometimes the belt itself is the problem, it may be bowed or damaged in some way. Or the lagging on the drive pulley may have worn inconsistently and causes the belt to run off.

To improve belt tracking you can use "V" return idlers instead of the flat returns (for some reason they don't do this much in the States) or add in training idlers, either on the carrying or return side. These are pivoting devices which have sensors which steer the belt back into the center when it drifts to one side.

The best ones in the market can "pivot" and "tilt" and should preferably have self lubricating bearings so that there is not a problem with the pivot point seizing up. Flexco manufactures the Persuader line of trainers and positioners which includes the PT Max and PT Smart. I can send you information on these products if you'd like.

Good luck with the thesis!

Helene Acock

Applications Engineering Manager

Parramatta Group

Belt Tracking - Causes

Posted on 20. Sep. 2002 - 11:10


Good question. Wonder if it has been asked before. From my experience, I will identify some of the likely expectations of belt mal-tracking.

1. material load centering & ore irregularities- (gravity forces)

2. idler alignment (friction and gravity forces)

a) plan view - off normal (90 deg) to belt center - install. error

b) elev. xsection - roll - earth settling (gravity)

c) elev. side view - twist or yaw - (friction) - idler frame warp

d) installation errors/corrections - rain chgs friction=misalign.

3. broken cables, cable alignment (mfg. construction) press etc.

4. belt (fabric & steel cord)construction errors in uniform tension

5. storage - poor support, telescoping, causing permanent set

6. pulley alignment

7. belt wear - crossection wedge shape wear from ceramic pulleys

8. splice alignment and pattern errors

9. training idler negative influence - kinematics can cause mistrkg

10. thermal expansion - sun orientation on steel cord belts no cvr

11. harmonic vibration that causes load to shift - frame compliance

12. residual effects from horizontal and vertical concave curve rad.

13. wind

Hope this provides fodder for the mill. The above points drive design criteria for long overlands and become guidelines for limiting the errors from such.

One hint - friction alignment error and mal-steering force, to belt axis, reaches a natural limit at about 2 degree tilt in horizontal plane. You should find some of these effects are also solved by understanding horizontal curve engineering.

There are formulations for the above. I think it good to do the sums to appreciate the complexities.

I am sure others will have more to contribute. It is a big field.

Lawrence Nordell


Conveyor Dynamics, Inc.

1111 West Holly St.

Bellingham, Wa 98225


Lawrence Nordell Conveyor Dynamics, Inc. website, email & phone contacts: phone: USA 360-671-2200 fax: USA 360-671-8450
David Pelc
(not verified)


Posted on 21. Sep. 2002 - 12:43

Thank you for your answers. Basicaly you answered me what I expected it may be. That's fine, but I´m trying to find out ,let's say, some kind of mathematical-physical model of belt behavior based on theoretical basis. Some kind of this was tried by Sachmajsters book "Teoria i razcet lentocnych konveyerov" but I'm not sure if this the right basis for further work. Of course I know this problem is mostly object of experiences and there are many primary and aditional influences, but I'm realy interested in the way of mathematical description of this problem.

Thank you for your replies once more.


Belt Mis-Tracking Analysis

Posted on 21. Sep. 2002 - 03:02


If I understand the second resonse, you wish to analyze the belt's response to a condition that is causing mis-tracking by either gravity forces or friction forces.

To resolve this you must solve the local belt transverse bending behavior as an indeterminate beam with local transverse gravity and/or friction forces. This is a far bigger problem. Alas, it has been done in two ways.

First Approach of transverse Bending:

The first derivation of the belt (beam) in bending, with a flux variation between cables, was solved for steel cord belt construction by Profs. Oehem (father and son) at the Univ. of Hanover, Germany. He wrote a treatment of the strain field in the cords, due to actions of curving the belt, approximating an indeterminate beam solution.

Combine this with the horizontal curve analysis (gravity banking forces) of Hans Grimmer in 1963 (I believe also from Hanover) and you begin to approach the physics. If you derive Oehmen and Grimmer's works, you should find some flaws in their assumptions and results. Oehmen's work would take some explaining. Grimmer is easier.

To calculate the belts transverse equilibrium, Grimmer first placed the ore as a dense field mapped to the belt width. Both act as a singular membrane in cross section when the belt is symmetrically loaded. This produces a variable density along the belt width. As the belt moves transverse, the equilibrium forces are calculated from the shifting belt. This does not take into account the true ore motion above the belt or ore resettling from idler agitation. It will give a so-so approximation. Lots of designers use it.

The steering force equations of idler-to-belt during mis-alignment, have been solved, but I don't recall who or where. The function resembles a graph where the steering force (SF) is plotted against the angular misalignment (AM). Zero AM equals zero SF. Two degrees AM equals 100%SF. Pick your friction coef. and wa la.

Second Method for Tranverse Beam (Belt) Deflection:

The indeterminate beam is best solved using FEA. Ryan Lemmon (Conveyor Dynamics, Inc.) published a work on belt turnovers, for the Society of Mining Engineers (SME) that may be of some help in this area. He has one of Oehmen's references. I believe the proper belt reference from Oehmen (father or son) is in German published many years ago (1960-1970's).


Lawrence Nordell


Lawrence Nordell Conveyor Dynamics, Inc. website, email & phone contacts: phone: USA 360-671-2200 fax: USA 360-671-8450
(not verified)

Belt Mistracking

Posted on 28. Sep. 2002 - 12:56

Hello David,

You may want to check the articles which were published in the journal "bulk solids handling" during the past 20 years by searching for "belt alignment", or "mistrakcing" or similar.

Please visit the eLibrary on this website and see if this might be of help.

I wish you a high degree of efficiency with your thesis.

Reinhard Wöhlbier

Belt Mistracking Theory

Posted on 5. Oct. 2002 - 08:19

Dear Mr.David Pelc

I have written one book on belt conveyor. Its information is on bulk-online. This book also has one big topic on your subject. Please contact me if you intend to know about the topic etc.

IG Mulani

Re: Belt Mistracking Theory

Posted on 13. Dec. 2002 - 04:51

Hello David,

As mentioned in the replies, there are a number of factors that can cause a belt to mistrack -- both external forces (idler misalignment, material buid-up on idlers and pulleys, frozen or sluggish bearings, uneven loading, load impact direction, side obstructions causing drag on the belt edges, over or under tensioning, et al) and internal forces (unevenly tensioned carcass, wavy carcass, crooked splice, belt bowing, et cetera).

There is a wealth of information available to you on belt mistraining causes and remedial actions (Mr. Mulani's book, CEMA, NIBA, most belt manufacturers, et cetera). Unfortunately there is little written on mathematical models concerning the various causes. Thus, if your thesis requires this, you will be conducting basic research and, thereby, doing the industry a favour. I would recommend that you contact belt manufacturers, conveyor manufacturers, idler manufacturers, and universities (such as the University of Hanover) to determine if they can offer assistance -- either in providing past study information or in conducting tests.

Basically, a belt reacts to the "braking" forces placed on it by the system and load, assuming good belt construction and splicing. The degree of belt reaction will depend on belt speed, load, tension, and braking (id est friction) force being applied. You should be focusing on models for each of the various mistracking causes, instead of attempting to develop an all-inclusive or unified model. For example, develop a model that shows the effect of a misaligned idler on a belt of a given weight (construction), travelling at a given speed, under a set load condition, and with a specific take-up tension. Show the effect of various misalignment angles for those conditions -- then change each of the

set conditions and repeat in turn. You can then follow this by repeating the exercise with a frozen idler and then with a sluggish idler. Further testing would be with material build-up of various amounts on one side of your test idler, through the same test range. Then do the same exercise with pulleys.

Further testing on uneven loading and impact direction would also be of use. As would side drag mistraining actions of skirting, side guide rollers, and structure contact. Finally, a study on the environmental effects of temperature, wind, moisture, dust storms, and icing would be of use.

Note that this list does not include internal causes.

Given the size of this field, you may want to limit your thesis to developing a mathematical model for just a few of the common belt mistraining causes -- my personal favorite is with the deleterious effect of over tensioning the belt, since this tends to increase the mistraining effects of all the other causes and a majority of users tend to run their belts too tight.

Good luck with your thesis, I hope to see the results in an article in Bulk Material Handling.


Dave Miller ADM Consulting 10668 Newbury Ave., N.W., Uniontown, Ohio 44685 USA Tel: 001 330 265 5881 FAX: 001 330 494 1704 E-mail:

Re: Belt Mistracking Theory

Posted on 13. Dec. 2002 - 10:44

Mr. Miller:

You note emphatically that mal-tracking may be caused by over tensioning. I beg to differ until proven otherwise. High tension in and of it self does not induce mal-tracking, in fact, it has just the opposite effect. If it did, then all attempts at high ST rated belts and belts under higher tension in horizontal curves would be at risk of controlling the tracking behavior. I have not witnessed such in high tension belts. El Abra, in Chile, has a rating of ST-6800 N/mm. The horizontal curves and the belt at-large tracks well at last inspection. The belt is designed to run with a safety factor SF= 5.5:1. This is 22% higher than the Safety Factor norm SF =6.7:1. Not many belts exceed this rating and none to my knowledge have high tension tracking sensitivity:

1. Selby (UK) ST-7000 SF= 5:1

2. Prosper Haniel ST-7500 SF= 5.5:1

3. Los Pelambres ST-7800 SF=5.51

4. Muskege River ST-7000 ??

However, too low a tension can cause mal-tracking. The irregularity of x-sectional loading and errors in belt construction can then dominate tracking response. Do a test, as we have done, measure the tracking of the belt edge in the high and low tesnion regions. You will find the high tension head station has a lower wander than the low tension tail station as the same section of belt passes.

Wish you well,

Lawrence Nordell

Conveyor Dynamics, Inc.

Lawrence Nordell Conveyor Dynamics, Inc. website, email & phone contacts: phone: USA 360-671-2200 fax: USA 360-671-8450
Gabriel Lodewijks
(not verified)

Belt Tracking Models

Posted on 13. Dec. 2002 - 11:44

Dear David,

We have, over the years, developed a couple of mathematical models that describe belt tracking phenomena. To validate these models experiments were performed. If you are interested then please let me know and we can send you some of the report available.

Kind regards, Gabriel Lodewijks

Delft University of Technology

Re: Belt Mistracking Theory

Posted on 13. Dec. 2002 - 04:52

Dear David,

Misalignment of belt can be for varied reasons such as faulty structure, faulty belt, faulty material load, etc. The chapter in book discusses and have mathematical calculations for aligning force due to 1) Pivoted self aligning idlers 2) Trough shape 3) Forward tilt of side rollers 4) Bow string effect. All these calculation procedure is based on material plus belt pressure (grip) on idlers, in accordance with alternating plus / minus strains in bulk material, as it is being conveyed. The material force calculations are based on Rankine and Coulumb theories for bulk materials as applied to belt conveyors. The four methods described in book can be further extended to calculate aligning force due to idler banking in vertical plane or idler positioning in (skew) in horizontal plane. 20 pages in my book (page 96 - 115) have mathematical calculations to work out aligning force with numerical examples. These have many figures, formulae etc and cannot be reproduced here. However, if you inform me your email, I can email you calculation of one case. However, still it will not be adequate for undestanding, as material forces are based on one full chapter-8 (about 60 pages).


I G Mulani

Author - Book on Engineering Science and Application Design for Belt Conveyors.

Re: Belt Mistracking Theory

Posted on 13. Dec. 2002 - 07:23

Mr. Nordell:

With respect to belt overtensioning and its effect on mistraining. Highn tension systems and horizontal curves such as you described are designed, both system and belt, to run with the belt well placed when running within the designed tension parameters. However, if the take-up tension is increased so that the belt no longer makes contact with the center cans of the troughing idler sets, the return run is taut, and/or the belt lifts off idlers in concave vertical curves, the belt will be more sensitive to other mistraining sources.

An overly tight belt will run harder on the troughing cans than on the center can and will, thereby, be more sensitive to slight idler misalignments. Likewise, if pulleys are not square, the mistracking effect will be more pronouced if the belt is overly tight around them. Finally, belts that are overly tensioned cannot absorb mistraining sources such as material build-up on idlers and pulleys, off center loading, side loading, or uneven skirt rubber pressure.

Conversely, low tensioning allows the belt to react better to the system - with the caveat that the tension cannot be low enough to permit slippage on the drive pulley(s) or to "float" in horizontal turns or turnovers.

From this, I can only conclude that it is best to tension belts to the belt manufacturer's and conveyor designer's specifications (if given), or to follow the general guideline of setting the take-up tension to a point slightly greater than that required to prevent slippage at the drive. Although many engineers prefer to use the T2 calculation to set take-up tension, I've had better success by using the belt sag between idler measurement method (percentage changes depending on the system).

Having said this, I also realize that specialized conveyor systems may not follow this rule of thumb -- that's why it is important to gather information from the conveyor designer. Unfortunately, getting information from conveyor fabricators and designers can be difficult. It would be ideal if every conveyor system came with an Operation and Maintenance Manual that provided the operator (owner) with system information, preventive maintenance tables, maintenance procedures, and troubleshooting guides. However, most conveyor operators are not given this information in a useful format, if at all.

With regard to your comment on the relative belt tracking at high and low tension areas of the system -- this does not apply. I've studied the relative mistracking between drive pulleys and tail pulleys; and I've made the same observation. However, when you overtension the belt, this mistracking worsens on both pulleys. The point is, operating a conveyor in either an overtensioned or undertensioned mode is not good for belt training. I've highlighted overtensioning simply because most operators who run at improper belt tension levels do so on the high side, not on the low.

I suspect that we are not in disagreement, but that I did not adequately stress the over part of overtensioning. The tensioning forces that I am discussing are the same that lead to splice failures, puncturing at load zones, increased bearing wear, etc. That is, running the belt with take-up tensions at two to ten times that indicated by T2 calculations -- or running the belt through piles of spilled and hardened material.


Dave Miller ADM Consulting 10668 Newbury Ave., N.W., Uniontown, Ohio 44685 USA Tel: 001 330 265 5881 FAX: 001 330 494 1704 E-mail: