tumblr stats

Peter's Nostalgia Site

Interests    Photos  Links  Home Page Feedback

 

 

 

Gearbox Overhaul  25th  March '96
 

"The gearbox deserves high praise because the synchromesh action is particularly effortless and rapid."
"The change from top to third demands simply declutching and pushing the lever straight through."

How often have you read pre-war  SS road tests and thought , how uncritical the testers were?

In the three years that I have had my SS Jaguar on the road I have always suspected that the synchromesh action was never very effective even when new. Consequently I persuaded myself that the effort of restoring it would be less than worthwhile. How wrong I was.

I decided to have a go, not because I objected to double declutching on downward changes but because of the slowness of making upward changes without 'snicking' the gears. In our gearboxes the main and layshaft gear sets, for second and third, run in mesh regardless of the gear selected and the mainshaft cogs not selected are allowed to rotate freely at their own speed. When a particular gear is required the selector locks the chosen cog  to the mainshaft with a dog clutch and this clutch can only be engaged when the mating dog teeth are rotating at the same speed.

 In synchromesh boxes a cone clutch is used to force the chosen cog to rotate at the mainshaft's speed and allow the dog clutch to lock it to the shaft.  The effectiveness of the cone clutches and the consequent effectiveness of the synchromesh depends on two ingredients: One, the quality of the friction surfaces and two, the pressure applied to the cones.

 With my gearbox I found that weak synchromesh resulted from deterioration in both departments. The cones were nicely polished and slippery and the springs in the synchro hubs were not exerting enough pressure.  My rather thin workshop manual gives no test for the springs in the synchro hubs but I found that the Mark V manual is much more informative and quotes 42-45 lbs pressure for the third and top gear synchro hub and 62-65 lbs for the second gear hub. All very well, but I found on dismantling my old pre-war double helical box that the synchro balls and springs were 1/4" diameter as opposed to the 5/16" of the later cars. Prior to dismantling I had obtained a set of 5/16" balls and springs, misled by identical numbers in the parts list for balls from early and late boxes.

I then found some notes on repairing Standard gearboxes (with 1/4" springs) quoting 42 lbs for third and top and 50 lbs for second. The question now was, should my early box with its smaller springs offer the same cone clutch pressure as the later boxes? My quandary was knowing that stronger pressure would improve the synchromesh but would increase the gear lever pressure required for engagement. I figured that the mechancial advantage of the gearlever in my box was very similar to the later boxes and decided to match the Mark V pressure figures. (In retrospect my assumption about the difference between SS and later gear changes was wrong. The later cars have much longer fore and aft gear lever travel than the SS and the increased leverage permits the use of greater spring pressure.)

        Comparison of fulcrum points


                       SS                           Mark IV/V

(You can see from the above diagram that the Mark IV & V cleverly separates the leverage for fore and aft travel from that of travel across the gate. The red line represents the single fulcrum point of the SS gear lever and you can see that the centre line of the across gate leverage in the later cars gives a slightly wider gate but the fulcrum for fore and aft gives a very much longer lever travel.)



 If guessing the original strength of individual springs proved difficult our trusty bathroom scales came to the rescue for pressure testing of the complete synchro hub assemblies. When testing the hubs you need to limit their travel or the sliding sleeve comes completely off the hub and tests your powers of observation as you search for the balls and springs which have flown in all directions. If you are doubtful about limiting the sleeve travel then wrapping the hub in an old rag saves much wasted time.

 Having restored the clutch pressure, refacing the cone surfaces is an easy if somewhat tedious process involving grinding-in the cones with fine grinding paste. This is exactly the same as grinding-in valves, but on a bigger scale, and cleaning away all traces of grinding paste after the event is just as important.



 All in all the differences between my ineffective synchro-mechanism before the job and that after seemed very small and my first test drive after reassembly was one of profound distrust. However, I am delighted to report that my distrust has now turned into astonishment at the effectiveness of the mechanism combined with slowly decreasing scepticism that the wonderous qualities are here to stay.

 Accelerating away is now a pleasure with no more feelings of guilt stretching the compromise between a rapid upward change and a nasty  'snick'. Even more amazing than my hoped for silent upward changes is the silent downward change, which is probably a more arduous test. The only limiting factor seems to be self-confidence which is now building by the mile.

 If you have weak synchromesh like I had then I thoroughly recommend the above overhaul. It really is well worth the effort. Those road testers were absolutely right.

A few other tips that I found useful:

 In assembling a synchrohub a piston ring clamp is highly effective for compressing and retaining the springs and balls.

 After removing the front seats, floor boards, starter motor and N/S exhaust pipe the gearbox can be unbolted and slid back off the engine. A static jack supports the rear of the engine and a small trolley jack supports the gearbox.

 If you spread the load with a hefty wooden block you can support the engine by jacking under the rear of the sump and the alloy casing does not collapse under the weight. (One of my fears.)

 If you leave the clutch pedal in place it acts as a useful handle when manoeuvring the gearbox off the engine and onto the rear floor.