Hello Shane,

4.5 meters would be to far apart for even a return pulley flat idler.

when we ran salt ROM belts, they were no more than 5 foot apart-the troughing idler pulleys with the return idlers at 10 feet.

if you run you belt like that it will tear up the belt faster and put more stress on the take up unit- I have not forgotten you in regard to the conveyor belt retriever- pleae send me your address again so I can send you a simple sketch.

lzaharis@lightlink.com By the way I am sorry I was not home when you called me :^(

leon z ■

Hello Shane,

In general, snub pulley and drive pulley (or head pulley or tail pulley) are kept as close as possible, keeping about 250 mm to 500 mm free length of belt.

The belt will get subjected to reverse bending when reaching the other pulley. This free length between the pulleys gives flexibility to belt in quick reversal of bending. More free length would be beneficial on belt consideration but would demand more space and cost for installation. Somewhat more distance would be preferred for exceptionally strong and thick belt. But this is a very finer aspect of positioning the pulley and one can take concurrence from belt designer / manufacturer in such situation, for possible implication on belt fatigue.

Regards,

Ishwar G Mulani.

Author of Book : Engineering Science and Application Design for Belt Conveyors.

Advisor / Consultant for Bulk Material Handling System & Issues.

Email : parimul@pn2.vsnl.net.in

Tel.: 0091 (0)20 25882916 ■

Hi Shane,

For drive/snub combinations the distance between tangential pulley faces is variable depending on the amount of snubbing. The values given by Mr. Mulani are reasonable for high angle snubbing and for dual drive units. It would be advisable to use the higher end of that range for high speed belt applications.

If the deflection is less, i.e. snubbing from a 180 degree wrap to 200 degrees, or if the belt speed is relatively low, the snub pulley can be positioned closer.

If you are in a situation where you are changing belt types on an existing conveyor (different ply count, different ply type (e.g. multiple reduced ply to straight warp), different cord diameter, et cetera), take care to evaluate the flexibility of the new belt. Stiff belts can fatigue quickly if asked to bend too much - and the splice can give you problems.

As far as the "rule of thumb" you quoted is concerned - the values given appear to be extremely high. I cannot recall a snub pulley that is 4.5 meters away from the drive pulley (bend pulleys yes, but not snubs).

As always, check with your belt manufacturer technical department. The belt is high cost component of most systems and it makes sense to protect your investment by using that resourse.

Regards, ■

Dear Shane:

I offer a professional difference of opinion wrt others on your request. Fabric and steel cord construction must be treated individually. Without getting to pedantic I offer the following.

Today, we identify a number of important facts that influence the pulley center distances.

1. Old rule of thumb was 1 second between changing wrap to minimize overlapping and additive internal belt splice stresses due to the rubber's relaxation rate between pulley wraps. Rubber can recover to near full strain relaxation in less than 1 second. Many core gum rubbers recover to 95 % of pulley induced strain change in about 0.5-0.75 seconds. However, its not just the rubber. Fabric and steel cords also need a relaxation period. We have measured their influence in dynamic fatigue testing of splices.

By example, a difference can be seen between the Hannover splice fatigue tester and the Goodyear tester. Initially, the Hannover machine had 6 m centers between pulleys. The Goodyear splice testing machine,, commisioned in 1996, has 12 meter pulley centers. Hannover increased their centers to 8 m to reduce premature cable and rubbber failures when running the belt at 6 m/s and 6m centers. Furthermore, the Hannover machine does not test the true splice relaxation time across a drive pulley of a working conveyor. The Gooodyear machine does. It relaxes the splice in about 0.5 seconds, similating drive tension drop from T1 to T2. Hannover takes 2 seconds. This fact shows up in the shorter cycles to failure on the Goodyear machine when operated at 0.5 second drive relaxation time. We have tested 0.5, 1.0, and 2.0 seconds relaxation time on the Goodyear machine and our Instron in-line splice tester. All results agree that the higher rate of strain or shorter relaxation times prematurely fail splices. Thus, placing drive and bend pulleys with too short a rubber and tensile member recovery distance is not good.

2. Newer Rule is split into the following:

a) for rubber relaxation time between pulleys - no less than 0.5 seconds of belt distance between bend changes on pulleys. More is better but with diminshing returns beyond 1.0 second.

b) pulley geometry anomolies due to lagging dimensional errors from material build-up, lagging wear, belt surface bulges, and mfgs. errors must be considered between high and low tension zones. A 1mm radial error in surface shape on 1000mm drive pulley diameters, at 1% motor slip, will produce approximately 20% variation in motor nameplate torque/power oscillation. The rate of change in geometry is also of importance. To reduce torque/power cross-talk between drives and between dirty side pulley surfaces and drive pulleys an elastic stretch zone is required. This is the straight section between pulleys that aids in longitudinal stretch forgiveness. Too close together, they can produce broken pulleys, drive power oscillations, and failed splices due to pulley diameter anomolies and the other points above. This has to be engineered and is of course more sensitive with higher strength belts.

Lawrence Nordell

Conveyor Dynamics, Inc.

www.conveyor-dynamics.com ■

Hello again Shane,

The test equipment that Mr. Nordell refers to for testing splices at the University of Hannover and the Goodyear Test faciulity both are large machines with a drive pulley and a tail pulley. Snub pulleys, or other bend pulleys, are not used. Also, these machines are primarily used to test high tensile belts that tend to be stiff.

Although I agree that the rubber, fabric carcass, and steel cords require time to relax, the question is how much time and how much excercise (bending) is practical. Unfortunately I do not know of a "standard" for this.

If we examine Mr. Nordell's center-to-center pulley approach of 1 second relaxation time before any path change it would mean that the in your example of a belt running at 4.5m/s the closest that your snub pulley could be is 4.5 meters. This would restrict your snubbing ability to approximately 10 degrees. Although a 190 degree wrap is better than 180, many design engineers want more. This is why there are a number of successful high speed systems that use a tight snub arrangement, as indicated by Mr. Mulani.

The problem with using a rule of thumb is that the inherent belt flexibility has so to do with the system requirements. A highly flexible belt, such as an ST1000, running at 5m/s will require less relaxation space than an ST7200 belt running at the same speed.

Regards, ■

Mr. Miller,

I follow your insightful discussion on this forum often. We seldom disagree. I offer the following comments to keep a technical perspective and moderation of differences with the main focus on steel cord. The comments are the product of extensive laboratory testing, failure analysis, and field forensic studies.

1. My earlier comments wrt 1 sec relaxation time was not my inventions but of belt suppliers and in circulation for many years. You and Mr. Mulani agree this may be a bit excessive. However, under some circumstances may be not.

2. Comments on Hannover and Goodyear test equipment pulley geometries was to inform, by example, the importance of distances in the general. Since many tests have been conducted on both machines where there is a tracable history of damage mechanisms. Such is normally not the case with field examples making scientific conclusions fague. I hope this corrects your misconception on the purpose of my point.

3. I support no less than 0.5 sec. relaxation time for rubber. This is a product of core rubber dynamic properties, not belt strength. This time may need to be increased for very high strength belts. I will not delve into high strength belt origins but to say that shock loading and strain rate play a part as the splice approaches the pulley. The shock tends to twist cables and break the cable's interior beyond the splice trailing end.

4. Pulley spacing should include the arrangement of belt wrap angles. Closer pulley proximity is possible with small belt wraps of less than 70 degrees. The angle of steel cord torsional twist locks solid after about 3 full helical turns subjected to a reasonable tensile force (ie. the internal radial contact stresses between wires can not increase with further bending on the same pulley - added wrap carries not cord fatigue penalty beyond 70 degrees such as a snub pulley).

5. Mr. Miller has not given any comment to the errors of pulley surfaces from lagging irregularity, material build up, ets that affect load sharing between drives, telegraphing of tensile strain from one pulley to the next - greater length to stretch a given strain (pulley geometry anomoies) lowers the stress. This tends to be a significant influence on steel cord belt and may require greater pulley spacing that the 0.5 seconds provides for major installations. A 0.5 second delay, at 5 m/s belt speed, requires a minimum of 2.5 meter straight

6. There are no standards to guide us, on the question but our education, experience, perceptions, and desire to minimize large and costly risks.

7. Arab (sic) proverb captures my thoughts:

" he who knows not and knows not that he knows -

question his wisdom"

" he who knows not and knows he knows not -

teach him"

" he who knows and knows not he knows is asleep -

awaken him"

" he who knows and knows that he knows -

holds the truth"

8. I do enjoy your provocative response within this forum and further encourage you to continue - some are too timid to query further and thus leave and apply their false conceptions.

Lawrence Nordell

www.conveyor-dynamics.com ■

Dr. Wolbier:

What would it take to get you to provide a spell checker on this forum ?????

Poor, humble, vasal without dictionary and synaptic failures between fingers and brain. ■

Shane,

Although it may appear that Mr. Nordell and I are fighting over the issue you started with this thread, we are trying to get to something that will help you and others. As you probably have already determined, a standard for pulley placement does not exist. The reason for this is that pulley placement is determined by a number of factors – of which the belt construction is one.

Now, back to the battle – Mr. Nordell:

I realize and appreciate that the 1 second relaxation time is based on testing done by the laboratories mentioned and that this, and other, testing are scientifically accurate. However, I believe that the tests are geared towards the high end of the belt spectrum. I have no argument with the relaxation time requirements for high tension steel cord belts and relatively high tension or stiff fabric carcass belts.

As the original question was focused on snub pulley placement, which normally involves wraps of 60 to 90 degrees on the snub, the 1 second “rule” can be relaxed – as you state in your point #4.

With regard to you fifth point, you are correct in that I have not addressed irregular pulley surfaces, irregular lagging (either initial or through uneven wear), or material build-up. I also have not addressed short transition zones, bad pulley bearings, or other factors that can have an adverse affect on both pulleys and belts. If any of these factors are involved, increasing the distance between belt bends will help to minimize the transmission of the negative forces from one pulley to the next, and will help to minimize damage to the belt. However, fixing those problems would be a more effective response.

The problem that I am having with this question (rule of thumb) is that I have been exposed to a number of conveyor systems running in excess of 5 m/s where there is a snub pulley placed within 10 inches (face to face) of the drive or tail pulley. Also, I’ve seen a number of tandem drive pulleys that are placed closely together. Both practices seem to contradict the “rule”. So I ask you, what am I missing? (This question should place on the second rung of your proverb!)

A Canadian addition to the proverb would be: “he who knows not and does not care is in management - he has engineers to take care of things”

Regards to both Shane & Larry ■

Dear Shane and David,

I would encourage not to use the words "fight; battle". Are we jesting or josting? People like to misquote - we are trying to inform. We can also have differences of opinion that cannot be resolved here. I welcome all learning experiences. Others can read and take the pleasure of their choice.

I will make a final attempt to correct some details.

1. One second rule is not laboratory based. It has been a quoted guideline by belt suppliers which incorporates rubber properties, belt construction, and material buildup to have safe harbor for their belts.

2. Low tension belts and high tension belts can suffer by too close a proximity between drive and adjacent snub pulleys, not by the one second rule, but mainly mineral buildup on the snub in contact with the dirty side. Note, belt cleaners often don't do the job. Do the math on elongation stresses wrt belt splice. Calculate the strain by a local buildup, on the snub pulley, such as you would do in defining when to use a crowned pulley and when the belt strain is too great. Assume 5m/s, 15mm local lagging buildup on 36 in. drive (not uncommon) and 18 in. snub, a 600 PIW polyester belt and a 10 inch face to face (FTF) distance (0.05 second). Then try a 70 inch FTF (~2m or 0 .4 secs) distance and tell us your result.

3. What about trapped rock and pulley proximities?

3. I added the other notes on steel cord belt and cable bending to talk on the other design issues.

4. CDI has extensive experience, with identifying material buildup problems and identifying splice and pulley failures associated with buildup and pulley arrangements. Fabric distortions with edge flap and center losenessis are issues which are exacerbated by undesirable pulley placements. This is true with drives, non-drive bends, takeup mass oscillation and all their associated stresses which may lead to premature splice and pulley failures.

5. Too short or not too short a transition distance telegraphs its stress field onto the pulley arrangement such as the drive and snub. I contend it should be a part of the proper choice of pulley placements. The fix to problems, such as these, is to select the best pulley placements, which is the question at hand.

6. Are you Missing something?? Yes! I agree that you can have a system with very close pulley proximities and they may appear to work. Does this make it acceptable when you look at the belt rating, utilization, buildup? Can it be that the system is over-designed? Engineers are responsible for proper design and fixing errors in designs. We can have very close proximities and weigh the cost and benefit of this choice. Sometimes it is the best choices. Most often it is not. I do await for your elongation stress result and am willing to place my bet on who eats the crow.

7. Most large hardrock owners are willing to pay large sums to remove pulleys where possible to avoid the prior stated risks.

Anticipating Daves reply.

Lawrence Nordell

www.conveyor-dynamics.com ■

Mr. Nordell:

I am glad that we agree – there are a many systems that successfully violate the pulley placement rule of thumb for stress relaxation. Although they may be “over” or “under” engineered depending on your viewpoint, they do work without compromising belt life, splice life, or pulley assembly life. I know of one such system that has been running the same belt for close to thirty years without any splice repairs or pulley changeouts required – although the lagging on the dual drives does get changed every five to ten years, depending on wear.

Although I have not run the calculations (and will not until I get back to my office at the end of next week), I am confident that they will show that the “1 second rule” is required to relieve the strain on the belt from bending around pulleys. Therefore, I will gladly “eat crow” on the calculations.

I realize that I have acknowledged the validity of the “rule” and have also claimed that it can be ignored, based on the evidence of working systems - even though the fact that something works does not mean it's right. This suggests the following:

1) The rule must be followed if the intent is to minimize stress on the belt and other conveyor components, minimize power requirements, and use the weakest belt possible for the application.

2) If the rule is ignored, there will be a price. The belt, pulleys, and idlers will be more sensitive to potential damaging influences. Also, the capital and operating costs will be higher than is necessary.

Finally, I am in complete agreement with respect to your points on rock entrapment problems and on the benefits of removing extraneous pulleys, although I consider those issues as other factors, albeit related, in proper belt conveyor design and use.

Regards, ■

Snub pulleys are a pain with sticky materials as well.......get build up causing the belt to flap and this creates issues with the secondary scraper tryin got clean the belt.

Then U have the issue of dealing with the "thrown spillage" from the bend pulley - chute design issues.

We are getting rid of our snubbers.

David........it is my understanding that snubbers are typ. used on travelling tripper type conveyors to reduce belt tensions/curves as well as shiploaders - boom conveyors (for the same reason).

Interestingly, belt splice life on shiploader boom conveyors is an issue eg Dally Bay coal loader - operator prefers steel cord belts to fabric belts. On the shuttle type of designs, there are quiote a few bends/pulleys and with the high loading rate > belt speed, cycle time is quite low.

Cheers

James Morrish ■

Whew ..... debates like that are simply breathtaking !

Just one small question though : this 1 second time-frame ... would it remain the same during the lifetime of the belt ? Would it not undergo a change as the composite belt modulus changes with time with ageing, flex fatigue etc. ?

If so, would the thumb-rule not depend on the loading cycle : a function of the conveyor length and the belt speed ?

Regards ■

Good point.

The life-time fatigue factor is considered in the recommendation. Belts do not fail immediately but accumulate damage over time. The whole issue of rubber rheology changing over time and its properties with fatigue damage would not be practiced by the lay engineer.

With large investments clients tend to be more chosey on the engineer that can quantify the differences and will have peace of mind on the belts risk free performance.

Most often the engineering community say.... "why make such a science of it?", They have not made the investment in research and cannnot quantify the benefit so they down play its importance. This is true on many technical issues wrt to design criteria, performance guarantees and life expectancies. You don't know how much your lack of understanding costs your client.

Short belts tend to be run slower. This will not make up for the long center distance of an overland. Note with short belts, the investment is much smaller and life expectancy is substantially lower. This becomes less critical to the owner. If it does fail prematurely, he upgrades its rating. Not a good answer but fits to the knowledge of engineering normally applied.

The attempt is to keep excessive stress levels in check on long belts. Eliminate unacceptable buildup. If not possible, extend pulley to pulley elasticity with the 1 second rule or longer. Apply good enginnering judgement when faced with evaluating the stress level. Recall, the main reason for the one second rule is buildup, or other geometry anomolies that cause local belt splice stresses from excessive or amplified strain associated with geometry.

Breathless.

LKN ■

Hello Kayem and Shane,

Kayem,

As Lawrence Nordell stated, the one-second rule of thumb has a safety factor built into it for the changing properties of the cover rubber and carcass (fabric or steel cord). It is meant as a guideline to ensure that the belt is not flexed too rapidly by changes in the belt path. The problem with the rule is that it is a guideline only – it does not distinguish between the various belt constructions. Therefore, engineers faced with a new system design or a system retrofit should use the rule in conjunction with discussions with the belt manufacture to determine what is economic, practical, and safe for the long-term life of the belt.

The use of snub pulleys, dual drive pulleys, short take-up bend arrangements, and hold-down pulleys put extra stress/strain forces on the belt, which are exacerbated the introduction of trapped material (lumps or simple build-up). By using the one-second rule, engineers are providing a relaxation time for the belt and space for auxiliary belt cleaners, if needed. If pulleys are placed too closely together, the belt and pulley components may experience premature wear and have to be replaced before their design life expectancy. This may, or may not, be acceptable – depending on the design criteria. The counter argument to the rule is the historical success of many systems that violate the rule (conveniently ignoring the historical failure of an equal or greater number of systems that violate the rule).

As far as your specific question on various aging and use phenomena are concerned, although information on the aging properties of both the rubber and carcass should be available from the belt manufacturers or via direct studies of existing systems, the actual deterioration of belt properties is very system specific – taking into account weather conditions, stress conditions, and running conditions. As such, your question highlights the most compelling reason for following the one-second rule – it provides a safety margin for long-term belt life.

The examples provided by David Beckley illustrate some of the problems associated with snub pulleys and drive systems that include two pulleys located close to each other. His point regarding splice failure rates is particularly noteworthy.

Shane,

To get back to your original question – there is not a standard on this issue, but there are conventions (rules) used by engineers for system design. Your enquiry for the purposes of proper snub pulley placement has generated a remarkable amount of dialog, which is good for the industry.

It appears that the use of a snub pulley to increase the efficiency of the drive pulley, to help counteract slippage, or whatever problem the engineer on site thought that a snub would correct, is undesirable for various reasons. Although I do not know how the one-second rule evolved, I find it curious that the primary issue is one of belt cleanliness and material entrapment that can cause localized stresses, which can be transmitted from pulley to pulley and cause harm to the belt and pulleys. To my mind, this sounds like a belt cleaning issue as much as a pulley placement/belt relaxation issue and suggests that you can shorten the placement distance if you run a clean belt.

Regardless of my thoughts on pulley placement and general conveyor system design, all can agree upon the following:

1)Getting rocks trapped between the belt and pulley is BAD.

2)Dirt buildup on belts and pulleys is BAD.

3)Overstressing belts and pulleys is BAD.

4)Studying a problem and doing research before acting is GOOD.

5)Using all resources (belt manufactures, forums, industry organizations, etc.) is GOOD.

I do not know if your question was for a specific conveyor application or was general in nature, but it did spark interest. If you are concerned about a problem with one of your conveyors or plan to make a change in the conveyor’s configuration, it is advisable to discuss it with the conveyor designer/fabricator and the belt manufacturer.

Regards, ■

Hey, I don't know how I ended up in this thread and really don't know why I wasted the last 15 minutes reading the comments but please tell me why, with today's quality of belting and our current understanding of conveyor design, you would even think to install a snub pulley in the first place. Please take that CEMA handbook and toss it out the closest open window as it has no place in a good engineering office. By the way, Larry's right...... ■