E
Ed Cory
Guest
Fools rush in, so I thought about it for a few days before I tried it.
I disassembled a 6 spd NOS Shimano freewheel, degreased and reassembled using JB Weld in place of
lube. I did NOT reinstall the ball bearings. I left the pawls and spring in, attempted to epoxy the
pawls into an interference position, I doubt that this succeeded.
I did not pay enough attention to aligning the two parts of the freewheel, there is misalignment
between the centers although the axes seems close to parallel.
Installed, there is no noticeable chain droop. Perhaps the eye is sensitive to errors which are too
small to matter.
I used red (permanent) threadlocker to attach the "used to freewheel" to the hub. Some rbt posters
say that interference friction is sufficient, I expect that the threadlock can only help.
The first test used 52:24 (cogs are 13-15-17-20-24-28). On a very hilly 7 mi loop considered
challenging by local runners, I ascended hills much faster than I would have thought possible.
This on a heavy 80's steel Bridgestone, compared to an all AL road bike where I would have
dropped to 30:23.
I did this ride unclipped due to my lack of fixed gear experience. I did pedal-brake to the best of
my ability. A stronger rider and/or more time will be a better durability test of this approach.
Switching to 52:20 (excellent chainline since my BB is a bit longer than the orig.) and spd pedals,
I moved to the nearby flatlands. After 12 miles of spinning for all I'm worth (my riding skills
rival my hack wrenching so I doubt that I averaged 100rpm) I turned around and discovered the wind
had been aiding.
After slogging along for 1/2mi, I stopped, gave thanks that Sheldon Brown's page had convinced me to
switch the solid axle to a steel QR and switched back to 52:24 for a slow but comfortable return
ride. This change was entirely accomodated by the horiz chainstays.
I plan to install 42t inside and 39t outside rings and hope that 42:20, 42:17, 39:17 and 39:15 will
be reachable without chain games. These would all be zero or one gear chainline "errors". The 52:24
(one gear error) is quiet and smooth- credit the SRAM PC-48? The 52 is a new "Rocket" ebay-ring with
no ramps nor pins.
Chain tensioning: lightly clamp the QR, crank a bit to allow the chain to slide the axle in the stay
a bit if needed, re-center the wheel if needed and clamp the QR tight to ride.
Thanks to Sheldon Brown in particular, and a few more
of you who were responsible for the inspiration. Ed
(remove spam to reply)
I disassembled a 6 spd NOS Shimano freewheel, degreased and reassembled using JB Weld in place of
lube. I did NOT reinstall the ball bearings. I left the pawls and spring in, attempted to epoxy the
pawls into an interference position, I doubt that this succeeded.
I did not pay enough attention to aligning the two parts of the freewheel, there is misalignment
between the centers although the axes seems close to parallel.
Installed, there is no noticeable chain droop. Perhaps the eye is sensitive to errors which are too
small to matter.
I used red (permanent) threadlocker to attach the "used to freewheel" to the hub. Some rbt posters
say that interference friction is sufficient, I expect that the threadlock can only help.
The first test used 52:24 (cogs are 13-15-17-20-24-28). On a very hilly 7 mi loop considered
challenging by local runners, I ascended hills much faster than I would have thought possible.
This on a heavy 80's steel Bridgestone, compared to an all AL road bike where I would have
dropped to 30:23.
I did this ride unclipped due to my lack of fixed gear experience. I did pedal-brake to the best of
my ability. A stronger rider and/or more time will be a better durability test of this approach.
Switching to 52:20 (excellent chainline since my BB is a bit longer than the orig.) and spd pedals,
I moved to the nearby flatlands. After 12 miles of spinning for all I'm worth (my riding skills
rival my hack wrenching so I doubt that I averaged 100rpm) I turned around and discovered the wind
had been aiding.
After slogging along for 1/2mi, I stopped, gave thanks that Sheldon Brown's page had convinced me to
switch the solid axle to a steel QR and switched back to 52:24 for a slow but comfortable return
ride. This change was entirely accomodated by the horiz chainstays.
I plan to install 42t inside and 39t outside rings and hope that 42:20, 42:17, 39:17 and 39:15 will
be reachable without chain games. These would all be zero or one gear chainline "errors". The 52:24
(one gear error) is quiet and smooth- credit the SRAM PC-48? The 52 is a new "Rocket" ebay-ring with
no ramps nor pins.
Chain tensioning: lightly clamp the QR, crank a bit to allow the chain to slide the axle in the stay
a bit if needed, re-center the wheel if needed and clamp the QR tight to ride.
Thanks to Sheldon Brown in particular, and a few more
of you who were responsible for the inspiration. Ed
(remove spam to reply)