Kick Starter Shaft Repair.

Jim

Beyond the edge is the unknown. Here be Dragons
Messages
20,510
Reaction score
87,680
Points
813
Location
Kansas City Mo.
The splines are buggered. As y'all know, a pretty common problem with old Japanese bikes.
This shaft is off an RD200, but everything done here applies to about all Japanese kick starters.
Here's the video that made me think this was doable.




Ordered a single wheel knurling tool that showed up today. Straight knurl, with wheels at .5, 1.0, 1.5 and 2.0 mm pitch. It's too large for my AXA holders, so a little moddin' is in order. Two reasons I didn't get one to fit my holders... first, I wanted something kinda beefy lookin. The ones to fit mine looked like matchstick toys compared to this one. And second, I couldn't find one with 2mm pitch unless I handed over a kings ransom.

1770344185408.png


I quickly discovered the axle circlip was spring loaded to the flight ready position. :doh:
My local Ace hardware has 'em, so....

1770344903752.png



To make it fit my holders, I cut out a piece of 1/4" by 1" steel and welded it to the bottom of the knurler.


1770345041898.png


1770345114776.png


1770345135310.png


Even used a little gun blue to feed the OCD for corrosion protection. :rolleyes:

1770345268758.png


1770345322246.png


Put it on the tool holder and found out that A, I didn't take everything into account when I dreamed this up, and B... I got pretty friggin' lucky. :cautious:

The knurling wheel was way too high with the adjustment knob down pretty low. Started raising it (dropping the tool) and quickly figured out I was prolly short on adjustment. Finally just raised the adjustment all the way up so it's not touching anything and let the AXA block bottom out.
Sumbitch if I ain't but just 5 or 10 thou high.
Here's the adjustment wheel all the way up...

1770345837131.png


And here's a dead center stuck in the axle hole...

1770345897199.png


And here's a view through the hole. Looks to be maybe 5 or 10 thou high. I figure I'll get a good 5 to 10 thou flex out if the Atlas when the wheel's under load, so I could actually be dead on. I think I'll shave a good 10 thou off the bottom of the block just in case... On the other hand... might just give it a go as it sits. But.... goin' on the assumption I've used up my share of luck for the day, I'm callin' it.

1770347412868.png


1770346317363.png
 
Last edited:
Ordered a single wheel knurling tool that showed up today. Straight knurl, with wheels at .5, 1.0, 1.5 and 2.0 mm pitch. It's too large for my AXA holders, so a little moddin' is in order. Two reasons I didn't get one to fit my holders... first, I wanted something kinda beefy lookin. The ones to fit mine looked like matchstick toys compared to this one. And second, I couldn't find one with 2mm pitch unless I handed over a kings ransom.

View attachment 365650

I quickly discovered the axle circlip was spring loaded to the flight ready position. :doh:
My local Ace hardware has 'em, so....

View attachment 365651


To make it fit my holders, I cut out a piece of 1/4" by 1" steel and welded it to the bottom of the knurler.


View attachment 365652

View attachment 365653

View attachment 365654

Even used a little gun blue to feed the OCD for corrosion protection. :rolleyes:

View attachment 365655

View attachment 365656

Put it on the tool holder and found out that A, I didn't take everything into account when I dreamed this up, and B... I got pretty friggin' lucky. :cautious:

The knurling wheel was way too high with the adjustment knob down pretty low. Started raising it (dropping the tool) and quickly figured out I was prolly short on adjustment. Finally just raised the adjustment all the way up so it's not touching anything and let the AXA block bottom out.
Sumbitch if I ain't but just 5 or 10 thou high.
Here's the adjustment wheel all the way up...

View attachment 365658

And here's a dead center stuck in the axle hole...

View attachment 365659

And here's a view through the hole. Looks to be maybe 5 or 10 thou high. I figure I'll get a good 5 to 10 thou flex out if the Atlas when the wheel's under load, so I could actually be dead on. I think I'll shave a good 10 thou off the bottom of the block just in case... On the other hand... might just give it a go as it sits. But.... goin' on the assumption I've used up my share of luck for the day, I'm callin' it.

View attachment 365661

View attachment 365660
Thanks for taking the lead on this! Looked up the knurling tool you bought and am going to pull the trigger myself. I don't have an immediate need for it, but have always preferred straight knurls (rather than diagonal) for thumbscrews, knobs etc
 
I'll be watching this!
Me too. I watched the video posted by @Jim and was wondering about the necessity of all the machining/press-fitting etc to create a new blank to machine splines on. Is there a reason why one wouldn't simply build up the damaged area of the shaft with layers of TIG-weld and then re-machine the shifter-shaft? That approach would be my first attempt.
 
Me too. I watched the video posted by @Jim and was wondering about the necessity of all the machining/press-fitting etc to create a new blank to machine splines on. Is there a reason why one wouldn't simply build up the damaged area of the shaft with layers of TIG-weld and then re-machine the shifter-shaft? That approach would be my first attempt.
Yeah, I wondered at that too. Only reason I could see to do all the machining is if the shaft were broke off on the end. Mine's not, so I'm gonna build it back up with the tigger and machine it down to the correct size. Hell, with all the welding the guy did in the video, he coulda just done the same.
 
Me too. I watched the video posted by @Jim and was wondering about the necessity of all the machining/press-fitting etc to create a new blank to machine splines on. Is there a reason why one wouldn't simply build up the damaged area of the shaft with layers of TIG-weld and then re-machine the shifter-shaft? That approach would be my first attempt.
I have an answer... sort of. :doh:

First thing I need to do is find out if the shaft's straight.
It's not. :rolleyes:

Stuck it in the lathe and measured the shaft... it's about 2-3 thou wobble just behind the splines. I can live with that. After all, it ain't like this shaft is pullin' a gazillion revs. It just goes down a quarter turn and back up when you start it. As they say, close enough.

1770515722148.png



But... I needed to know that so I could see how much the splined area is bent. Just eyeballin' it tells me it's bent...


1770515955872.png



Took some head scratchin' to figure out how to measure how much. I finally came up with this contraption.
Ground a point on a brass rod, then stuck that in the machined hole in the end of the shaft and put the runout indicator against the rod. Didn't get a pic of the actual measurement, but here's the general idea... Indicator hasn't been moved to the rod in this pic.

1770516383065.png



Anyway, it worked. There's .028" runout at the end of the splines. :doh:
So, I marked the high side on the shaft, and stuck it in Big Red.


1770516591862.png



It's a good thing the lathe and press are only a few steps apart, 'cause it took at least a dozen trips back an forth. Measure runout, bend... measure, bend. About 45min into that ordeal, the answer hit me like a lightning bolt. I coulda done like the guy in the vid... make and press on a cap to machine down to size. That would have been a fair bit quicker. Anyway, finally got it down to about 2 thou wobble... close enough.

The answer to the question? If the shaft runs true, go ahead and weld it up. If the shaft is bent (most likely), go ahead and machine a piece to press on as a replacement like the guy in the vid.

After that, I stuck it in the hydrochloric acid tank to remove the zinc plating, ground down the buggered splines and she's ready for welding.


1770517130769.png



Welding a shaft is much easier when you can rotate it as you weld, so I made up another Rube Goldberg contraption.


1770517260623.png


1770518026939.png


1770517302717.png



Used the disc sander to knock off the big pieces...


1770517362305.png



... and stuck it in the Atlas.


1770517422929.png


1770517468494.png



Still need to take a fair bit more off, but I'm on hold until the kicker boss shows up. I measured the diameter of the splines as best I could considerin' the state they were in, but I want to see if I can verify the correct dia before I try to knurl the splines.


1770517651446.png
 
Last edited:
Once I know the right diameter, I'll turn down a test piece and knurl that. Go from there. :shrug:
Interested to know how you are going to determine that. Google AI gives a reasonably good description of how minor/major diameters are used to determine shaft size. At the end of the day, all you really need is enough interference + the clamping of the pedal boss to ensure the shaft-to-pedal joint does not move as you are kicking the starter shaft through its stroke, right?
 
Since knurling moves/displaces metal, my first instinct would be to simply turn the test bar down so it fits inside the kicker boss (or use pin gages to determine the max diameter the boss will accept and turn the bar to that pin gage's size), and then knurl it till it locks into the kicker boss.

My other instinct would be to try to research the appropriate standard (not ANSI but the Jap equivalent) and machine the bar to those specs. The ANSI standards I use tell me the min and max lathe diameter, the spline count, and what my micrometer should say if I use 2 pin gages between 2 splines.
 
Last edited:
Back
Top