By Howard Grivna

There are several factors that affect a widebelt sanders ability to hold a close thickness tolerance. When a sander is new, with proper operating procedures, any rigid orifice type machine should reasonably hold plus or minus .005” tolerance. If a machine has been specifically designed and has the right characteristics to hold a close tolerance, thickness tolerances of plus or minus .0025” are achievable. However, within a short period of time [less then one year], certain wear factors require machine maintenance procedures along with proper operating procedures in order to continue to obtain tight thickness tolerances.

Side To Side Head Adjustments

Individual sanding heads must be adjusted parallel to the conveyor belt and for best sanding result, to each other. The degree of this parallelism should be within a total

of .001”. This can be achieved and maintained with a “Set-Up Device.” 

Care must also be exercised to make sure that the outboard spacer and its contact surfaces are dust free. The hold down bolt must also be tightened with a uniform tightening force. A torque wrench may be required to accomplish this. 

Conveyor Belt 

Many conveyor belt factors can have a dramatic effect on a sanders ability to hold tolerance. First of all, all conveyor belts must be callipered [dressed to a uniform thickness]. New belts must be callipered [dressed] and then redressed approximately every six months to eliminate wear variances. This process is described in section 19 of my book. 

The conveyor belt hardness can have a major impact on a sanders capability to hold tolerance. For maximum tolerance holding capability, belt hardness should be 85-90 durometer or more. The top surface pattern of the conveyor belt also impacts tolerance capability. A smooth face belt results in optimum tolerance holding capability. Be advised that belt hardness and top surface type can impact part feeding capabilities. Smooth face hard belts have less feeding grip then softer, rough textured belts [see section 19 of my book]. 

Contact Drums

Contact drums are one of the most critical components in a widebelt sander. For optimum dimensional control, contact drums should be steel or hard rubber. Because they wear unevenly, they must be periodically ground true. Also, for optimum dimensioning capability, they should be running at belt speeds of 5,000 to 7,000 SFPM.

·         Run out should be less than .0005” T.I.R. [measured when drum is running in its bearings]

·         Balance should be as fine as possible [within 2 grams]. Displacement should not exceed .0002” - .0004” at operating R.P.M.       

·         Wall thickness should be sufficiently heavy so drum will not deflect when heavy cut is made. 

Thickness Adjustin  

All widebelt sanders have inherent backlash [slop] in their thickness setting devices. Thickness setting changes are accomplished through the use of jack screws rotated either by means of a worm and worm gear device or a chain and sprocket arrangement. Therefore it is essential that thickness changes are always made with the final adjusting movement in the same direction. If this is not done, side to side thicknesses can change by as much as .015” - .020”.

Top Machines: On most top sanding machines, the conveyor bed is moved up and down relative to the sanding heads to adjust thickness settings. In those cases, the final setting should always be made by moving the bed upwards to the exact setting. If the bed is moved upward beyond the final set point, it should be lowered by at least one full hand wheel turn and then slowly raised taking care not to go past the required setting. 

Bottom Machines: Almost all bottom sanding machines are built so that the conveyor feed system hangs from the adjusting jack screw arrangements. This means that the final movement of the adjustment should be in a downward direction. If the set point is missed, the conveyor feed system should be raised up by at least one full handwheel turn and then slowly lowered, taking care not to go past the required setting. 

Machines with Entire Top Frame Adjusting: On some sanding machines, the entire upper frame assembly is moved up or down instead of just the feed bed. In those cases, the final movement of the adjustment should be in a downward direction. If the set point is missed, the entire top frame should be raised up by at least one full handwheel turn and then slowly lowered, taking care not to go past the required setting. 

Inherent Machine Design Deficiency

Not all widebelt sanders are designed for optimum dimensional thickness control. In fact, some machines are purposely designed and built with yielding orifices for optimum results when sanding sealer or thin face veneers. If close thickness control is one of your requirements, make sure that you purchase a machine with the proper characteristics to optimize those requirements. If you find that your machine has thickness control design deficiencies, it may be possible to modify in the field and gain some improvement. 

Abrasive Belt Wear Differential

Be aware of the fact that abrasive belts change in thickness across their width due to uneven wear. This uneven wear can be the result of uneven part feeding or a differential in the cutting characteristics of multi-density products. The coarser the minerals that an abrasive belt has, the greater is the potential for uneven wear. The following belt thickness data reflects information that I have personally gained by measuring various used belts across their entire width. New belts will generally measure within .001” or .002” across their entire width. 

Belt Thickness Variation Example [Cloth Backing] and Belt Wear Differential [all values vary and are therefore approximate]

GRIT                BACKING                     Total Thickness            Est.Wear Diff.

20                     Cloth                             .092                              .030”

                        24                     Cloth                             .078”                             .025”

                        36                     Cloth                             .070”                             .020”    

                        40                     Cloth                             .066”                             .018”

                        50                     Cloth                             .058”                             .014”

                        60                     Cloth                             .053”                             .012”

                        80                     Cloth                             .046”                             .010”                

                        100                   Cloth                             .037”                             .004”

                        P120                 Cloth                             .035”                             .003”

                        P150                 Cloth                             .033”                             .002

                        P180                 Cloth                             .032”                            .0015”              

The following is actual data that I obtained at a plant that was trying to hold plus or minus .002” thickness on truck flooring material. When the abrasive belts were new this was achievable, but due to uneven feeding, the belts wore as indicated making this tolerance goal impossible. 

Worn Belt Thickness Analysis  

                        ‘A’ Belt ‘B’ Belt ‘C’ Belt ‘D’ Belt ‘E’ Belt 

New Belt         

Thickness         .096”                .096”                .087”                .087”                .087” 

Run Time         8 hrs                 8 hrs                 “Long run”      8 hrs                 8 hrs                  

Measurement

Location [Lane] 

Edge                .090”                .093”                .080”                .082”                .085”

1                      .092”                .092”                .078”                .081”                .079”

2                      .088”                .083”                .075”                .076”                .072” 

3                      .083”                .081”                .074”                .072”                .071”

4                      .082”                .083”                .070”                .076”                .075”

5                      .084”                .084”                .068”                .077”                .075”

6                      .084”                .083”                .069”                .075”                .076”

7                      .082”                .082”                .068”                .075”                .071”

8                      .082”                .079”                .069”                .073”                .072”

9                      .081”                .083”                .065”                .071”                .073”

10                     .085”                .083”                .065”                .069”                .071”

11                     .082”                .082”                .068”                .073”                .070”

12                     .081”                .074”                .068”                .072”                .070”

13                     .083”                .080”                .070”                .073”                .072” 

14                     .085”                .084”                .069”                .073”                .078”

15                     .090”                .083”                .073”                .079”                .081”

Edge                .096”                .090”                .073”                .081”                .084”

                                                            Test Methodology

a.      On belts A, B, C, & D, a section was removed out of test belts across the entire width.

b.       Belts A & B were from opposed heads and had run eight hours.

c.       Belts D & E were from opposed heads and had run eight hours.

d.      Belt C was from one head only and had run “long hours?”

e.      Measurement locations 1 & 15 were taken 1” in from each edge.

f.        Measurement locations 2 & 14 were taken 3” in from each edge.

g.      All other measurements were taken every 2” across the belt widths. 

This graph takes the above dimensional data and coverts it into a “profile picture”  

It becomes obvious that a final product thickness tolerance of plus or minus .002” is impossible to achieve once the abrasive belts are unevenly worn. In this example, the belt profile varied by approximately .015”. This effect could have been minimized by employing my “SANTRAC” device, which prompts an operator to achieve even feeding across the machine width.