INDO GULF COPPER SMELTER PIPE CONVEYOR

In a continuing effort to remove the fears surrounding the use of pipe conveyor technology as an alternative conveying option, the writers have revisited the Indo Gulf pipe conveyor operation, (in India), to assess what can be learned from the present operation and the improvements that can be made for future development of the concept.

The 3,2 km pipe conveyor at Indo Gulf has by and large been considered a triumph for the pipe conveyor industry as a whole.

This development has proved conclusively that there is a place for the pipe conveyor in an environmentally conscious society and that we are far from reaching the limits of its length and capacity.

There was also a lot of skepticism that it was possible to develop the 90-degree horizontal curve, which was proved to be possible.

However, where attentions to alignment, level and component installation accuracies are not significant on conventional conveyors, pipe conveyors must be installed to a far higher degree of alignment accuracy to counter the tendency the carry has to twist especially for the empty belt.

The Indo Gulf pipe conveyor has now been operational for 1½ years and there is much to be learned from its present performance, with the aim of identifying performance shortfalls, possible operating improvements, and measure the actual power used and arrive at an accurate operating cost model.

Also as the pipe conveyor has experienced some operational problems it is interesting to identify the cause and effect of such problems.

Prior to entering into the problem areas of the conveyor we will detail the conveyors basic parameters.

1.	Conveyor length, pulley centers	3 120 m.
2.	Overall lift	22,0 m
3.	Belt speed	0-4,5 m/s. variable speed
4.	Conveyor capacity	1600tph (initial); 2200tph (future).
5.	Belt Factor of safety (design)	~8,0 (initial); 5,2 (future)
6.	Pipe diameter	450 mm
7.	Belt width x thickness	1650 mm x 17,8 mm.
8.	Belt specification and covers	ST 1000, 8 x 6 mm covers.
9.	Horizontal radius	300m with a 90 deg horizontal curve.
10.	Vertical inclination (max.)	5,74 degrees.
11.	Vertical radius	300m
12.	Location of take-up	Head-end.
13.	Type of take-up	Horizontal trolley, gravity mass.
14.	Take-up travel distance	14,0 m live; 19,0 m available.
15.	Take-up mass	19,4 tons.
16.	Belt sag	1% to 3%
17.	Installed power	Head=2x500kW; Tail=1x500kW
18.	High tension and drive pulley dia.	1 000 mm.
19.	High tension pulley face width	1 900 mm.
20.	Gearbox ratio (Normal)	17,73 : 1
21.	Type of reducer	Elecon – Helical gear.
22.	Motor type	GEC(Alsthom) Frame DC355F900R
23.	Motor shift speed	0 – 1490 rpm
24.	Type of high and low speed couplings	Flexible geared couplings
25.	Low tension / take-up pulley diameter	800 mm.
26.	Low tension pulley face width	1 900 mm
27.	Idler diameter	152 mm.
28.	Idler gauge length	325 mm.
29.	Number of idlers per panel	12 off.
30.	Pitch of idlers – straight sections	2 000 mm.
31.	Pitch of idlers – curved sections	1 000 mm.

Table 1 & 2 indicates the actual power consumption for the various offloading capacities and the amount of material conveyed in its initial operation.

The justification for the pipe conveyor was based on both an environmental sensitive and cost analysis when compared against alternative options of 2 belt conveyors with a transfer point.

Figure 1 shows an environmentally sensitive area that had to be circumvented requiring either a curved conveyor or a conventional conveyor with a transfer point.

The transfer point was considered unacceptable because of the high cost of the materials being conveyed and the damage spillage could cause to the environment.

An open troughed belt could not negotiate the tight curve and would have been susceptible to wind blown dust.

During our latest inspection, we noticed a number of operational problems, which we considered needed addressing to achieve an acceptable system.

None of the items in themselves adversely affected the operating performance of the conveyor, however it was felt that prior to bringing the conveyor up to full design capacity and for information for future projects, one should at least be able to identify the reasons for the problems, and where practical, effect modifications which would eliminate them.

This observation was initially made at the commissioning stage of the conveyor. It was seen that three belt sections, (of the 30 installed), caused the belt to rotate when empty by almost 180 degrees in the carry strand primarily along the straight section.

When entering the curve the belt would correct itself and run correctly through to the discharge pulley. The return strand ran, perfectly throughout its length, with the overlap on the bottom roll.

The conveyor is equipped with training idlers, but because the movement was confined to only 10% of the belt, correction to the bad sections resulted in movement to the good sections.

As it was not apparent why the belt sections caused the belt to twist, it was decided to compromise the belt twist and set the empty belt to move between –30 degree and +45 degree about the vertical center line through its length.

After 1 years operation we were informed that the belt was behaving worse than initially and that we should investigate the matter further.

To identify the reasons for the erratic movement, and to locate the problem sections of belt. A series of inspections and tests were conducted at site.

The first test performed was intended to verify that the belt sections were constructed correctly and that there was no bow in the belt. A set of alignment arms (tilt switches) were positioned either side of the belt in a flat straight portion, then a simple pen recorder indicated the movement and misalignment of the belt.

It can be seen from the graph figure 2 that generally there is no clearly identified belt section causing alignment problems, the belt ran between +/- 20mm throughout its length with round 3 producing some bigger variations at sections 5,6,7 and 20,21,22.

A second test conducted in the splices set out to prove the correctness of alignment of each splice.

Here only one splice was out of tolerance and was remade. However, this had no effect on the belt performance.

Having eliminated the possible problems with belt manufacture and splice construction we considered the possible problems associated with the way the pulling of the belt into the structure was carried out.

In past years little has been written about the effects of installing the belt in the correct sequence, in fact little or no literature exists about the problems that could occur with incorrectly installed belt.

When one considers the pipe conveyor and the problems of carry strand empty belt stability then the procedure of belt installation was revisited.

The procedure followed for belt pulling and splicing the belt sections was shown in figure 3 above, which indicates that the belt was pulled into the top and bottom strands with alternating splicing between the strands.

This allows us to pull the belt into one strand while splicing in the other strand, which was done to expedite the construction which had been delayed by the extended jetty construction phase with the final splice being carried out at the tail.

Today convention dictates that to counter cable tensing problems during belt construction, the belt should be pulled onto the conveyor in the same way as it is constructed.

Therefore it is necessary to ensure that the procedure of rolling the belt during manufacture is repeated during installation. Again referring to figure 3, we see that splice 14 and 30 were made end-to-end and start-to-start respectively. Therefore if there were any variations in cable tension, these would be amplified in the way the belt was installed.

To confirm this possible problem a further belt inspection was carried out by the writers, which made it possible to confirm that as indicated in figure 3, two splices were in fact end-to-end and start-to-start (splices 30 & 14) and a third splice (19) was also incorrect.

The reason for splice 19 being turned around cannot be fully explained, however it was noted that during belt inspection at the factory one belt was re-rolled because it was not correctly packed. The re-rolling procedure would also have the effect of reversing the belt.

Highlighting this type of problem, which was not anticipated at the time of installation, has given us an answer to the cause of the problem. Therefore it has been concluded that it could be necessary to reverse those belt sections to eliminate the erratic belt movements in the carry side.

At the time of writing this modification has not been carried out, in fact as the problem only occurs on the empty belt it may be possible to allow the belt to stay as it is for the foreseeable future. Possibly carrying out the change when a replacement belt has to be installed.

However what this problem has highlighted is the degree of sensitivity experienced by the pipe belt and the lengths one must go to, to ensure an accurate installation because carry belt stability is very difficult to guarantee.

The pipe belt construction cannot be considered as very forgiving for structural alignment, or variations in belt construction, which makes us very concerned about specifying a replacement section of belt that will perform correctly in an existing installation, especially if that section is supplied by an alternative belting manufacturer.

The conveyor is presently operating with 2 of the 3 drives because the carrying capacity is dictated by the ship off loader, which is set at 910tph.

When the second ship off loader is commissioned the third drive will be added to load up to 1850 tph.

Investigation shows that the return belt is absorbing 20% less power than initially calculated.

Inspection shows throughout the straight portion of the return length, the belt only contacted the lower three rolls of the return set, asking the question of why the other three rolls were installed in the first place.

This section is almost horizontal and the tensions are reasonable low in this area, the question is asked why the smaller diameter.

Review of the original profile in relation to the preset profile leads the writers to postulate that the reasons for this phenomena is the relationship between belt mass and sliding friction between the overlapping belt sections.

This phenomenon is currently being studied and is the subject of a future paper to be written on the subject and will not be touched upon in this paper.

However this dose answers the question of reduced tension in the return strand and therefore the inability of the tail drive to share power correctly.

The solution to this problem is either to increase the return belt tension by increasing the take up tension or replacing the tail drive with the proposed future secondary drive at the head station.

Prior to leaving this subject of pipe redirection as identified in the previous section, the writers would like to take this opportunity to address the problem of cause of pipe collapse, or diameter reduction on many other installations.

During the past 3 years a number of installations have been visited to investigate similar problems often resulting in premature belt failure. Without exception the blame has been leveled at the belt construction and also without exception after inspection the reasons for these failures can be attributed directly to the conveyor layout, and always the problem of inadequate vertical or horizontal curve radii.

Reviewing the design procedure of many proprietary suppliers of such systems clearly indicates the reasons for such problems. Little or no attention is actually given to the forces present in the design of the curve and subsequently belt failure at these junctures.

When one considers the conveyor, which is the subject of this paper and the degree of attention paid to the curve design, one understands that almost all belt failures are a result of inadequate curve design.

To expand on this critical issue of curve design, we would like to explain the procedures followed in selecting the correct cross belt rigidity factor which would ensure that the belt at Indo Gulf would not buckle in the curves.

As there was no published literature available in the degree of rigidity necessary to ensure the belt would not buckle, and we were operating in the area of uncertainty when one considers that we were constructing a 90deg high-tension curve we decided to test a section of the belt in the laboratory.

A 10m section of the proposed structure was built, refer to figure 5, and an actual section of the proposed belt, manufactured to our anticipated requirements, was installed in the structure.

The structure was designed to form the required radius and the belt was then tensioned to the operating limits.

As the tension was applied, the indentation measurements were taken at one of the idler rolls. These tests were repeated at various smaller radii to ensure there would be no chance of buckling, and that the construction was in fact safe. Obviously it would have been more accurate to conduct the test on a moving pipe, but this was not practical.

As a result of the experiment, it was possible to be confident that the belt as tested would not buckle at the radius selected and it was a simple exercise from this point to measure the actual belt rigidity factor which could be used to ensure the correct construction of the total belt (Refer graph figures 6-9) and which was used to test the belt during the actual manufacture.

During the operation of the conveyor there has been no observed pipe collapsing, except the reduction in diameter noted above.

As this reduction in diameter in a straight section of the conveyor has not been explained at this time and as this does not affect the conveyor performance and results in a lower power usage then one is tempted to ignore it.

However as it is our hope for the future that we will be able to push pipe conveyors to the lengths traveled by conventional conveyors we must be able to give an explanation for this effect.

To this end we are embarking onto a research project to assist in identifying the cause and procedure of eliminating or predicting the effect thus being able to offer a more efficient design.

This problem was identified through the vertical curve and was caused by the belt edge pressing hard into the roll.

The problem was more prevalent at the top of the curve and diminished as we advanced down the curve. The cause of the problem is obviously a function of the belt edge rubbing into the roll with high pressure at the top of the curve and thus diminishes as the tension reduces.

Figure 10 shows the problem and to overcome this problem it is necessary to change the overlap of the return belt so that the overlapped edge is on the inside of the belt on the inside of the curve.

This problem which is typical of pipe conveyors and therefore critical that it is resolved results in the belt moving horizontally in the loading point and thus reducing the effectiveness of the pipe forming section.

The design of the chute was such that the loading was based on 1850 tph and when operating at the lower capacities the load shape impinged on the side impact idler and pushed the belt off line.

There it can be seen that when one has variable loading conditions, the belt tends to be pushed down the inclined troughing idler and thus the belt is misaligned as it starts the pipe forming.

To overcome this problem we have introduced a completely redesigned feed chute, which incorporates a number of improved features.

As the cost of power in India is very high, an optimization procedure has been undertaken to lower the total system operating cost.

By running the conveyor at an optimum speed for the various materials and capacities, we consider that we will be able to lower the overall operating cost.

Send mail to Webmaster with questions or comments about this web site.