Sensitize more your model...

[iacob alex]

     ...  or prototype,if you have in mind an experiment,not reducing the friction,only(    www.geocities.com/iacob_alex/Some_Drafts/text014.jpg    ),but increasing the output/input ratio
,also.
        In what manner?!  Simply: enlarging the size (   www.geocities.com/iacob_alex/Some_Drafts/text045.jpg   ).

        You can think this as an "input-black box-output" configuration;

           -black box= your model

           -input = a constant  value (gravity potential difference).

           -output= a variable  value(kinetic energy).

          With the same input("arm difference..."),we can get an ever increasing output( "stored energy"...),if we have time,that is free fall velocity,that is diameter/size of the model.

                      All the Bests! / Alex

[iacob alex]

.....or imagine it bigger and bigger,as you can see at :   www.geocities.com/iacob_alex/Some_Drafts/text045.jpg

     The title "A small confusion?!" is reffering to something that,usually  "disturbs" the thought process...

     Here,the torque( due to a constant  long arm-short arm difference) is becoming greater and greater,for  increasing dimensions.

     In other words:for the same input(we "pay"),we gain an ever expanded output.

           All the Bests! / Alex

[iacob alex]

   .....if your objective is to discard easy a non-working prototype.

        The observation is simple : rolling friction << sliding friction.

        The viewpoint is uncomplicated : test the static unbalance of your model,with the pivot supported on two horizontal planes.

         You can face two occurences:

           -to and fro/swinging motion

           -rolling motion

          Only the second circumstance (a well-advanced rotation ) can be an evidence ,that your model can be a solution.

                    All the Bests! / Alex