Hello Brad,

We’re not really replying to your ‘best practice’ strategies but giving you suggestions on how to improve idler life.

There are more and more facilities available for tracking events of idler failure. In South Africa Sasol have implemented a procedure which is now running for about 15 years where all roll failures on all conveyors are tracked (manually) and the roll supplier is expected to replace the roll if it fails within the warranty period. Whilst not identifying or giving prior warning of roll failure the system has identified specific areas along the conveyors where failures are prevalent (usually conveyor related problems such as vertical misalignment (bearing failures) or idlers being used as trackers (shell wear)).

Richards Bay Coal Terminal used a thermal imaging procedure (infra-red camera) to determine increased bearing temperatures and then replaced the affected rolls during the normal maintenance period. In this instance the procedure was combined with a refurbishment policy whereby (with roll being replaced prior to collapse) the bearing seals (and possibly shaft) were replaced into a slightly worn shell.

Regarding identification of imminent bearing failure Huwood developed the ‘whistling’ roller where an audible signal was emitted once the bearing temperature achieved critical levels )I believe Larry Nordell discussed something similar in a previous forum).

I believe that new procedures with advanced electronic chips are being developed by Delft University, C-Kit and Lorbrand (see posting).

Regards,

Adi Frittella ■

Hello Brad,

Thought I’d add my thoughts on idler failures. Perhaps you could advise the belt parameters and the idler configuration being used. Give the forum an opportunity to bench mark.

By the way has anybody heard whether the problem of premature failure on wing rolls on three roll garland sets operating on stacker-reclaimer has been solved?

1.In what environment are the rolls used? The 3 year life-time is relatively short for a well maintained, correctly aligned and cleaned surface application, however in an underground environment where alignment, belt cleanliness and general housekeeping around the conveyor often take second place behind production needs , the 3 year life may be optimistic.

2.Previous responses have identified the need to recognise all possible influencing factors in selecting a suitable shaft diameter and bearing rating for the conditions. The two are interlinked in that the bearing (typically deep groove ball) will not perform if the inner and outer races run in a misaligned state (shaft deflection under load). It appears that a general rule of 100,000h is stated as design L10 life for bearings. There are significant differences in opinion on this with European standards generally opting for 30,000h, USA (CEMA) having recently changed from an overall 30,000h to a difference (dependant on rating) of either 30,000h or 60,000h. Locally (South Africa), the move has been to reduce from the previously standard 100,000h to 50,000h. A factor influencing this decision is grease life which is typically guaranteed to a maximum of 40,000h.

3.Having selected a suitable shaft/bearing combination the life of the roll is significantly influenced by the effectiveness of the sealing arrangement in preventing ingress of dust/moisture into the bearing. Note that a highly effective seal may result in excessive ‘rim-drag’ with resultant increased wear of the shell.

4.Your information states that failure is ‘bearing failure’ – is this confirmed? Very often the roller is replaced once some form of collapse has occurred, at which stage typically the shaft, shell and bearing housing are too severely damaged to establish the root cause of failure. Have the bearings of failed rollers been examined – the effects of bearing misalignment often leave a very clear on the races.

5.Excessive failure of return rolls is often the result of excessive carry-over due to incorrectly set belt scrapers and also of the return roll being used to excessively deflect the belt, hence acting as a bend pulley.

6.Premature failure of correctly designed, effectively sealed rolls operating on well installed systems is still possible if correct tolerances between shaft/bearing and bearing/housing are not adhered to. (Quality of manufacture).

Regards,

Adi Frittella ■

Brad,

As stated above, there are a large number of variables that need to be taken into consideration, and rechecked, when determining why the life expectancy of your rollers is so low. (Design, etc.)

Most Idler manufacturers you talk to, will be happy to run a quick design check to confirm what you are using is the correctly sized component.

From what you have said, the bearings appear to be the area to fail first. Nine times out of ten (provided the roller is correctly designed) this means the sealing system is not adequate for your conditions. (dusty, wet, salty etc.)

May I would suggest trialing other roller manufacturers that may be able to offer a better sealing system or other alternatives, with proven results.

The salt industry for example, have done numerous trials on different idler sealing systems and found that life expectancy can be from two weeks to four years plus, given the extreme conditions.

Other influencing factors could also be the components used in the roller itself. Type of grease, make of bearing, end cap alignment etc.:- all contribute to saving costs and decreasing the life expectancy of the roller. - Check how your roller warranty (18 months is standard for Australia - depending on the conditions) stacks up against the B10.

Possible solutions to your other queries;are;-

Belt Damage - Steel end caps produce a "pizza cutter" effect once the shell is warn away and the roller is seized - Polyurethane End caps will ware away rather than cut the belt.

Monitoring - We have used transponders in our end caps, this gives each roller a unique number and in turn makes monitoring idler life as simple as scanning the the failed roller to check when it's installation date was. - This also helps to monitor rollers for extended warranties.

Hope this helps - for more information please drop me a line. ■

Hi Brad,

If belt damage is a result of separation of the bearing housing (end-cap) from the roller shell then an alternative method to the use of a polymer end cap would be the use of a collar type (or turn-back) bearing housing as used by some idler manufacturers. See for example Melco in South Africa (www.melco.co.za)

Regards,

Adi Frittella ■

Dear Brad,

Quanitfying the roll failure rate has many components. The full discourse is more than can be described in this forum. I will try to give a generic answer as follows:

1. What did the engineer design as a bearing L-10 life?

2. I calculate you have a 1.5 m carry and 4.5 m return spacing

3, I assume you have 3 equal roll carry and 2 Vee return.

4. At 4000m, carry roll count is 2666 x 3/assy = 8000 rollers

5. reurn roll count is 889 x 2/assy = 1778 rollers

6. Points 4 and 5 equal 9778 or about your 10,000 total rolls

7. Assume there are 160 hours of operation in one week with 8 hours for maintenance.

10. Assume all rolls have the same life and fail 40 per week out of 10000 rollers.

11. Then 160 hours/ 40 failures per week = 4 hrs mean time between failures (MTBF)

12. Then 10,000 rolls x 4 MTBF hours/roll = 40,000 MTBF hours/roll

13. This equates to a bearing MTBF = 40,000 x 2brgs = 80,000 hrs.

14. This equates to a L10-hrs bearing life = 80,000/5 = 16,000 hrs. a very poor designa by any engineering standard

14. The L-10 hour number should be about 6 times this number or have a failure rate 1/6 or 6-7 rolls per week.

Maybe the engineer did not include all things that contribute to the idler loading to have such a high failure rate and low MTBF.

We would be happy to provide a more formal evaluation and bring the failures under control. There are many things to consider such as"

1. tonnage histogram varations for each wing and center roller



2. off center loading allowance ( poor transfer station design)

3. material buildup on idlers ( poor belt cleaning - may be need turnovers)

4. structural alignment - settlement or bad installaition

5. roller design tolerances

6. vertical curve geometries, spacings and installation tolerances

7. belt flap vibration forces and their influence on life

8. all correction factors on bearing rating

Wishing you luck,

Lawrence Nordell

www.conveyor-dynamics.com ■

Nice job Nordell !

The L10 (or B10 Life) of idlers are calculated for 100,000 hours usage. Obviously as Nordell said, you should get 6 times the life you are getting.

Are the failures consistent? Are they on both troughs and returns? Shaft deflection, substandard sealing arrangement for the application, thin shells? check the grease temperature ranges in the rolls. Are they regreasable (tapered roller bearings?) or non-regreaseable ball bearing (most likely in your area).

Any idler manufacturer should be able to tell you why they are failing and/or do proper L10 or B10 calculations for you if you provide all the specs. they require. ■

Dear Mr. Brad Frost,

The earlier respondent has given a detailed answer. I would like to add following regarding the failures of your idlers:

1) The bearing can fail if the bearing dynamic capacity is not adequate in context of expected life.

2) The bearing can fail if the spindle is not adequately stiff and deflection at bearing is crossing the allowable limit or it is very close to allowable limit.

3) Incidentally, what is the material you are conveying by your conveyor? If this material happens to be very abrasive type, ingress of abrasive particles in bearing assembly can also reduce the bearing life.

4) Imperfect sealing of idlers can also result into bearing failures.

5) If the climate has very sharp variations in temperature during night time and daytime, it can promote ingress of dirt particles into the bearing assembly.

6) The mis-alignment of idlers in installations can also promote the idler / bearing failures.

Therefore, while analysing the idler failure, just do not focus the attention on bearing only, but analyze the other parameters also, which can indirectly contribute to bearing failures.

While quoting for the important project, the experienced customer normally asks the supplier about the required spares (with quote) for one-year operation, 3-year operation and 5-year operation. So, while quoting for such spares, the concerned companies make the failure analysis of 3-rollers of one idler set, the load pattern is different for side rollers and for middle rollers. This failure projection is done using SKF / equal make formula and life factor a1.

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 ■

Brad,

One of the main reason why idlers do not reach L10 = 100,000+ hours is because of bearing lube issues incl contamination.

Idlers are made as cheap as possible and this means cheaper seal systems, even the well known and regarded Prok/Sandvik idlers.

If U send them away to the OEM, U will be able to get an answer of the Root Cause of failure.

On our idlers, we went to the same supplier but the better seal system....with marked changes in results. Also, if U have FRAS rubber in the rubbing external seals, this is sensitive to UV light and will crack/age...

Continental make a very elaborate seal saystem for their idlers but the roll resistance is higher.

On the wing idlers, the seal is at an angle and dirt/dust can accumulate near the seal and work its way into the bearing assy.

Next, U need to have a strategy on maintaining.

We have regular a walkaround checks and tagging/ID and then planned replacement - idlers do not suddenly fail but give warning signals.

There have been products on the Aussi market wrt idler checking eg thermography and sound.. Check the internet or your ilder supplier for advise.

If the belt is critical, U may need to look at "on the run" idler changeouts - Laurence Nordell has a design for such a device.

Cheers

James ■

Hello,

The major quantitative reply (Nordell) assumes 8 hours for maintenance (point 7). If you are spending 2-3 hours replacing idlers then you are within the reasonably accepted service timeframe; whatever the quality of your proprietary equipment and the tenacity of your maintenance crews.

Investment in an infra red camera ( see the reply from James Morrish) will accurately identify the deterioration of bearing performance throughout the system. The camera will also show a lot more, if properly used. Send me your email address and I'll send you a gif file of an on-line idler replacement gadget. ■

Dear Brad

One way of planning conveyor maintenance is to use a

PDA based system to record defects, generate

maintenance schedules and enforce warranties.

Lorbrand has developed a customised conveyor

maintenance software package

which can be used on most popular PDA's .

Our website illustrates the operation of the software

(www.lorbrand.com)

Regards

Hans Granig ■

Hi Brad

Have U made a decision/come to a conclusion yet wioth yr maintenance strategy??

Do U wish to share it with us.

Thanks

James ■

Wow! What a great topic. Great answers also. I will add my two cents worth. Make certain the belt is as clean as possible. A belt like this will require several belt cleaners (I have installed as many as five on an application such as this.) The cleaner the belt is at the head pulley, the less strain on the return rolls. ■

Great answers on conveyor system requirements for optimum life of the system components.

I understand that you are looking for an approach to deal with the maintenance aspect of the system.

-The system is operational and currently you have an idler related downtime of 3 hours per week at some cost that includes cost of repair, lost production, manpower etc. That accounts for near 2% downtime if the system operates 24-7.

-The major cause of failure is due to bearing failure (bearings giving up).

-I'm not sure if you identify 30-50 idlers per week or replace 30-50 idlers per week but in any case, you have a method to qualify what is a good idler and what is a bad idler.

-You state that the life of the idler is three years... That gives a MTBF. A horrible one but at least you have one.

-Replacement of 50 idlers in three hours is very good effort only depending on the work force. 2 workers-->7minutes per idler...

3 workers-->11 minutes per idler... My opinion is that, with some planing a worker should be able to do 7 minutes per idler with 70% wrench time. With this you could replace close to 100 idlers in three hours with a crew of five.

-You need 100 spares.

-You need to create an "Idler Crew".

I'm sure that there must be an alignment reference in the conveyor system. If there is not one, you need to create one.

May be that the support system moved... You need to verify alignment.

You need to create a procedure for repairing the idlers and a procedure to intall the same with an effort to return the system to the aligned state. It may be that the fast replacement of idlers is adding to the low reliability of the system.

At best, you should be capable to replace enough idlers to the point were you can schedule the time for the effort.

You need to keep the cleaning system well maintained so that there aren't periods of "rough ride" between adjustments of the belt cleaners.

One last point is lubrication/sealing. It usually goes both ways. Too much grease and poped seals is as bad as not enough grease.

Lab seals in general are only as good as the environment.

Little by little you will improve the availability of the conveying system. By the way, 2% downtime is not very bad.

Good luck,

Antonio Reis

Vitrom Mfg Consultants

Your Process and Manufacturing Solutions

Phone: 209.834.1900

Fax: 209.834.1039

www.vitrom.com ■

Hello:

The main reasons for bearing failing are seal and bearing block problem, if seal system was not good, the bearing will open to bad working environment that will cause bearing the rapid decreasing of working life, another reason is that if the bearing block was not precise by axial direction, that may cause big run-out phenomenon during delivery, that will make the bearing in a extremely vibrating.

Jienasi roller use SKF bearing which can make sure the working life of conveyor idler, for more information, please log on http://www.jienasi.com for checking.

With best regards for all..

Steven ■