It is mainly depend on the application that you are simulating or the purpose of the experiments that you are running. In addition, the rest conditions can be specified based on the conditions of the application that you are simulating or the purpose of you experiments.
Its depend upon the mechanical and thermal properties of your pin material and counter face material, environment also. Generally increasing applied load proportional to wear rate but in some case it may reverse. Secondly in what aspect you want to investigate and severity level.
There is no standard load for the pin-on-disc wear test. As pointed out by Mr. Vashishtha the load should be decided depending on your experimental condition and your requirements only.
In most of the cases, the wear of the pin is proportional to the load applied (based on Archard Eqn). Therefore, you can review any prior available literature on what load they have used for your particular material and you can optimise your load requirement based on that.
Dr. Simhan is correct in that ASTM G99 does not specify a load, and Dr. al-sallami has also provided good advice - that the load (and speeds and sliding distance) should be chosen with a specific application in mind. So I have little to add to these previous answers.
If you have no such application from which you can try to reproduce the service conditions, then you will be well served to do some "scoping tests" to see what combinations of load, speed, "pin-end geometry" and sliding distance will give you a good means of comparing materials.
I can tell you that the original Round Robin tests for G99 were run using a 10 mm diameter ball, at 1 cm/sec sliding speed with a 10 N load and a 1000 meter sliding distance. Wear scar diameters were compared and COF was recorded. There is nothing special about these choices for test conditions, and we are starting to discuss a new round robin, with different conditions (higher speeds and different materials).
Sure it should be based on ASTM standard and it depends on the load and surface characteristic. I do not remember but in your area it should be ASTM2266. Or try to review some literatures
As Dr. Steve Shaffer Said, the wear round robin tests within the framework of the Versailles Agreements on Materials and Standards (VAMAS) have been already performed on the wear behavior of steel and ceramics under given sliding conditions of the ball-on-disk test system. Considering these works as well as previous tribological studies done on your material and also with respect to the interlaboratory wear test conditions cited in the ASTM standard, the wear load will be well chosen.
In lubrication case, the load will be chosen based on the determination of a particular lubrication mode which corresponds to contact pressures in the tribological system application. For example, the boundary lubrication can happen if the oil film becomes thin enough at higher loads and low sliding speeds.
All depend on the properties of the materials used and the lubrication regime. Whether it EHL or boundary or extreme pressure. ASTM set certain standard for testing that you can compare your results to it.
Just to add to the previous comments, take into account that all depends on the tribopairs that you want to study and also what you want to achieve. For instance, if you want to cause wear on a hard coating (like DLC); you will probably need to use a 1 mm diameter ruby ball (http://www.glassballlens.com/ruby-ball.html) against the coating to cause any measurable wear. Remember that on typical pin-on-disk tribometers load is limited (tens of N), then you rely on contact mechanics to cause yielding on the surface (point contact in this case). You can use the basic contact mechanics equations for point and line contacts (cylindrical), then based on the geometry find the amount of load that will generate yielding (plastic deformation and wear) on your tribological interface depending on the lubrication regime.