GearheadGrrrl
Active Member
Just wrapping up my 2nd valve check on the S10, and after 8+ hours wrenching I have again found all valves in spec. Same story with my other bike with a similar engine design, a BMW F800S where after a day's work each time I found all valves in spec at 12k, 24k, 48k, and 72k. So we're spending a lot of time searching for signs of a mostly non existent problem, valve lash loss, while creating a whole bunch of potential new problems like valve cover oil leaks and worse. There has to be a better way...
BMW "airhead" owners are taught that if they can't hear valve clatter, adjust the valves until they do. Railroads are using sensors to monitor locomotive's health so they get sent to the shop when actually needed rather than by a fixed schedule. There are distinct changes in a valves temperature, conductivity, vibration, sound, etc. signatures as it approaches failure. Of those measures, probably the easiest to use is sound, and even our own experienced ears can often note the telltale signs of valve train components going amok. So why not take a database of valve train sounds, then by computer find the distinct sound features of valve trains that are out of spec and headed for failure? Then use that data to build a simple app that can diagnose by sound and predict when valve adjustment is really needed.
Whacha think?
BMW "airhead" owners are taught that if they can't hear valve clatter, adjust the valves until they do. Railroads are using sensors to monitor locomotive's health so they get sent to the shop when actually needed rather than by a fixed schedule. There are distinct changes in a valves temperature, conductivity, vibration, sound, etc. signatures as it approaches failure. Of those measures, probably the easiest to use is sound, and even our own experienced ears can often note the telltale signs of valve train components going amok. So why not take a database of valve train sounds, then by computer find the distinct sound features of valve trains that are out of spec and headed for failure? Then use that data to build a simple app that can diagnose by sound and predict when valve adjustment is really needed.
Whacha think?