Honda S90 - Engine and Gearbox Rebuild


Following a couple of test drives, the gearbox was found to be slipping in second gear and the engine was smoking.

This page details the investigation and rework that was done to correct these problems.

The Honda on its first drive - June 2008

 

 

 

 

 

 

 

 

The slipping second gear needed closer inspection, so the engine and gearbox assembly was removed from the bike.

These images show the inside of the gearbox and the parts on the bench

(Click images)

Gearbox

Parts on the bench

 

 

 

 

 

 

 

 

The follow defects were found.

The gear selector forks were worn at the contact areas.

The dog clutch for second gear had quite deep grooves worn between the engagement holes.

The prongs on gear 2 were also worm and rounded off.

Other parts showed mild wear but these were the areas of concern.

Selector Forks

The prongs on gear 2 were worn where they should engage with the dog clutch.

The dog clutch had grooves worn between the holes.

 

 

 

 

 

 

 

 

The result of these defects was that the gearbox wouldn't cleanly select second gear. This photo shows the gearbox after trying to select 2nd gear. The 'roller' hasn't full engaged in the 'gear cam' and so inside the gearbox the second gear is not fully engaged either. By turning the input shaft a notchiness could be felt which was the dog clutch partly engaging. It was also difficult to then get the gearbox out of this state into any other gear.

These were the 2 problems encountered when the bike was road tested.

Gearbox stuck between 2nd gear and neutral (click)

 

 

 

 

 

 

 

 

The sequence in the tables below shows the theory behind the gear selection failure and was based on observations on the gearbox.


Good gear change sequence

This first image shows the 2nd gear dog clutch before meshing.

Note that the gears will be turning at different speeds as they approach each other.

As the gears touch even though the dog clutch is not lined up there is little friction because the touching surfaces are both flat and perpendicular to the rotation of the gears.

Because the gears are turning at different speeds, when the holes and clutch teeth line up the gear is engaged correctly.


Gear jam problem with worn dog clutch

This image shows the same gears but with rounded clutch teeth on the upper one and wear around the holes in the lower gear.

This time when the gears touch they are not touching with surfaces which are perpendicular to the direction of rotation, and so are not able to slip into place easily.

Because the gears are turning at different speeds and because the touching surfaces are not perpendicular the dog clutch cannot merge and mesh fully. In this situation 2nd gear cannot be engaged.

The purple arrow shows the contact point.


 

 

 

 

 

 

 

 

The selector drum was inspected for wear. The grooves were supposed to be a maximum width of 6.4mm. They were approximately 6.1mm throughout.

The clearance between the drum and the fork sleeves was acceptable ie: less than 0.025mm.

 

 

 

 

 

 

 

 

Fixing the gearbox

To improve the fork condition some brass shim was brazed onto the worn face of one of the fork tips to improve its fit in the dog clutch.

This had the risk of changing the ductility of the fork so only a long-term test would show if this would be a problem.

 

 

 

 

 

 

 

 

The bolts used to guide the forks (within the slots) both had mild wear, but by swapping them between the forks, new faces were exposed to the slot sides to giving a tighter fit in the grooves. This was possible because the start position of the thread was different in each of the forks.

 

 

 

 

 

 

 

 

These images show the old and new gears and give a feel for the wear that had occurred on the old parts.

Both gears were fitted to the input shaft without issue.

 

 

 

 

 

 

 

 


Fixing the Engine

To try and fix the smoking engine 2 new valves were fitted, because they had been measured as slightly undersized, however the new ones were only 0.01mm greater in diameter.

 

 

 

 

 

 

 

 

After this work was completed the engine was put back in the bike for another test run.

 HOME

 

EMAIL

 

BACK