Testing for Belt Longevity - 20ft Pipe

[CDAndrews]

Hi All,

I have a vendor wanting to run belt sanding tests to prove his belts can significantly decrease cost (over time) by running faster and increase longevity. He has requested to run 5 pieces (20ft in length and about 2" in diameter) repeatedly. These belts (4 in total and 5" wide each) are designed to give a grained appearance to the pipe. The belts will not differ between individual sanders. Each sander will have one belt. Here is my question:

Will running these same 5 pieces repeatedly give me a reliable result as to belt fatigue, load, wear out, etc.?

I would assume, dangerously enough, that once the initial grain has been achieved - on the first pass - that subsequent passes will be easier on the belts than running an unfinished piece each time (as we would in reality). Am I correct?

Thank you all very much!

 

[Miner]

You need to be able to compare the new to the current baseline. There are two ways to do this: 1) Use the new in the same manner as the current, or 2) Use the current in a parallel study the same as the new.

 

[normzone]

Yeah, you'd be best served by testing both at the same time, same conditions. Most reliable data achieved that way.

 

[CDAndrews]

Yeah, you'd be best served by testing both at the same time, same conditions. Most reliable data achieved that way.

Thank you both very much; I really appreciate your feedback. I was thinking the same. I believe that to be the only way to get a good comparison, without getting into a wrestling match regarding supposed specifics and variables. Unfortunately, while I am the foreman and project manager, I'm not the final decision-maker on the project. I have to ride the line between what I know to be true (and can prove) and how the "more experienced" guys have handled vendors for years. Wish me luck; I'm walking in with my head held high...

 

[CDAndrews]

You need to be able to compare the new to the current baseline. There are two ways to do this: 1) Use the new in the same manner as the current, or 2) Use the current in a parallel study the same as the new.

I realized the way I replied only showed up in response to normzone's post. I wanted to make sure I thank you directly as well. So, thank you very much for your early and accurate response!