Welding onto a gearbox input shaft
Discussion
My Noble uses a mildly breathed on ford mondeo ST220 gearbox, its held up very well with 580+ftlb torque but there is an issue with it.
Basically the input shaft is a floating design, it comes within 20mm of the crank but doesnt actually locate in it... There is room in the crank for a bearing (Jaguar version of the engine has a different box were the input shaft locates in a bearing in the crank).
The issue I have is that the huge torque of the engine vibrates the input shaft by over 1mm and thus my clutch plates catches on the inner casing of the clutch basket.
I wish to eliminate this by machinging up a 2mm section that will locate 5mm into the input shaft and then into the crank bearing... question is can I weld onto the end of an input shaft?
Basically the input shaft is a floating design, it comes within 20mm of the crank but doesnt actually locate in it... There is room in the crank for a bearing (Jaguar version of the engine has a different box were the input shaft locates in a bearing in the crank).
The issue I have is that the huge torque of the engine vibrates the input shaft by over 1mm and thus my clutch plates catches on the inner casing of the clutch basket.
I wish to eliminate this by machinging up a 2mm section that will locate 5mm into the input shaft and then into the crank bearing... question is can I weld onto the end of an input shaft?
What about drilling into the shaft and creating a small locating dowel that way ?
Although companies do weld dog teeth onto gears after machining them, so I'd say if that can be done, the input shaft shouldnt pose a problem other than making sure everything is perfectly aligned before welding.
Maybe get them heat treated or cryo'd afterwards...whichever the most appropriate.
Although companies do weld dog teeth onto gears after machining them, so I'd say if that can be done, the input shaft shouldnt pose a problem other than making sure everything is perfectly aligned before welding.
Maybe get them heat treated or cryo'd afterwards...whichever the most appropriate.
You could but obviously it will need to be very straight which is far from easy when welding. I would also avoid any torque being passed through the welded section (or more accurately the heat affected zone), which would be a recipe for failure e.g. can you put the welded section 'behind' the friction plate, on the engine side?
andygtt said:
My Noble uses a mildly breathed on ford mondeo ST220 gearbox, its held up very well with 580+ftlb torque but there is an issue with it.
Basically the input shaft is a floating design, it comes within 20mm of the crank but doesnt actually locate in it... There is room in the crank for a bearing (Jaguar version of the engine has a different box were the input shaft locates in a bearing in the crank).
The issue I have is that the huge torque of the engine vibrates the input shaft by over 1mm and thus my clutch plates catches on the inner casing of the clutch basket.
I wish to eliminate this by machinging up a 2mm section that will locate 5mm into the input shaft and then into the crank bearing... question is can I weld onto the end of an input shaft?
Certainly if it's done properly. However you are unlikely to be able to weld an extension on such that it's perfectly aligned without a final machining op. I'd suggest making the extension from EN 24T otherwise known as 817M40T which is fairly hard and strong in the T condition but also easily machinable.Basically the input shaft is a floating design, it comes within 20mm of the crank but doesnt actually locate in it... There is room in the crank for a bearing (Jaguar version of the engine has a different box were the input shaft locates in a bearing in the crank).
The issue I have is that the huge torque of the engine vibrates the input shaft by over 1mm and thus my clutch plates catches on the inner casing of the clutch basket.
I wish to eliminate this by machinging up a 2mm section that will locate 5mm into the input shaft and then into the crank bearing... question is can I weld onto the end of an input shaft?
Drill the end of the input shaft to 2/3 of its maximum end diameter by 5mm or so deep and chamfer at 45 degrees to make a weld recess. Make the extension to fit into this hole also with a 45 degree chamfer and leave the spigot bearing diameter 1mm oversize. Tap the extension in, TIG weld into the chamfer and then finally turn the spigot extension true and to size and trim up the weld. It shouldn't need further hardening. The lateral loads are not high.
If it's just a pilot bearing then it's not taking any torque. I'd do as stevie says and drill a quality hole in the end of the shaft and shrink in an extension. No need for any welding.
Much of the middle core of the shaft isn't doing much in the way of torque transmission so you can probably get rid of the inner 6mm or so to locate the extension without detracting from the torque capacity of the shaft to any huge degree. In fact if you keep it shallow enough, there might not be any loss at all, since the torque load starts where the clutch engages with the splines.
Much of the middle core of the shaft isn't doing much in the way of torque transmission so you can probably get rid of the inner 6mm or so to locate the extension without detracting from the torque capacity of the shaft to any huge degree. In fact if you keep it shallow enough, there might not be any loss at all, since the torque load starts where the clutch engages with the splines.
I think this is quite a bad idea.
The gearbox itself maintains axial alignment of the input shaft centerline and the crankshaft centerline by having two pretty stout bearings on each casing half and two dowel positions on the casing that peg it to the engine. If those dowels are not in place then you will not have good alignment.
At the extremes of tolerance based on the ability to machine every single hole to true nominal position there will be crank to input shaft misalignment. This can destroy the driven plate hub on cars that idle a great deal - not untypical on many US market cars where owners start the car, go in and have breakfast, then go to work when the interior is warm. To counteract this, there is often a misalignment feature in the clutch driven plate hub that allows the centre to float relative to the true centre defined by the lining position.
I would suspect that the phenomenon the OP is seeing is symptomatic of a machining tolerancing stack issue combined with a floating clutch hub. if the misalignment is as bad as he reports then machining the end of the input shaft and welding on a pilot will result in so much bending stress in the weld area once assembled that the new part will fatigue and fail very quickly.
I would fix the issue with corrections if required to the doweling and a non-floating driven plate hub.
The gearbox itself maintains axial alignment of the input shaft centerline and the crankshaft centerline by having two pretty stout bearings on each casing half and two dowel positions on the casing that peg it to the engine. If those dowels are not in place then you will not have good alignment.
At the extremes of tolerance based on the ability to machine every single hole to true nominal position there will be crank to input shaft misalignment. This can destroy the driven plate hub on cars that idle a great deal - not untypical on many US market cars where owners start the car, go in and have breakfast, then go to work when the interior is warm. To counteract this, there is often a misalignment feature in the clutch driven plate hub that allows the centre to float relative to the true centre defined by the lining position.
I would suspect that the phenomenon the OP is seeing is symptomatic of a machining tolerancing stack issue combined with a floating clutch hub. if the misalignment is as bad as he reports then machining the end of the input shaft and welding on a pilot will result in so much bending stress in the weld area once assembled that the new part will fatigue and fail very quickly.
I would fix the issue with corrections if required to the doweling and a non-floating driven plate hub.
I know its a compleatly different type of car but on my saxo turbo we made an adaptor spline that was a tight fit onto the gearbox input shaft and incorporated a nose that went into a bearing in the crank then used clutch plates with larger splines obviously
This cured 2 problems we had using the ap twin plate horrid vibration and unpredictableity when taking up drive and ripping out the tiny splines in the clutch plates cos of the torque lol
Si
This cured 2 problems we had using the ap twin plate horrid vibration and unpredictableity when taking up drive and ripping out the tiny splines in the clutch plates cos of the torque lol
Si
Cheers for the responses
To give a little more detail,
The box is designed for 200ftlb I'm running 3 times that although I suspect it's not the torque that causes the deflection in the input shaft.
All upgraded nobles do this when they have an upgraded clutch, probably due to the extra clamping force, mine is worse as I have a triple plate and thus higher clamping force, it reduces when I put the lesser clutch in and wasn't there at all when the car was stock.
The noise that's clearly this deflection happens on downshift or pulling away, never on up shift.
The other easier option is to machine 1mm from my end friction disc as I will still have plenty of clamping force... It just seemed like a bit of a patch it for the real issue?
To give a little more detail,
The box is designed for 200ftlb I'm running 3 times that although I suspect it's not the torque that causes the deflection in the input shaft.
All upgraded nobles do this when they have an upgraded clutch, probably due to the extra clamping force, mine is worse as I have a triple plate and thus higher clamping force, it reduces when I put the lesser clutch in and wasn't there at all when the car was stock.
The noise that's clearly this deflection happens on downshift or pulling away, never on up shift.
The other easier option is to machine 1mm from my end friction disc as I will still have plenty of clamping force... It just seemed like a bit of a patch it for the real issue?
So when you remove the box, is the input shaft loose, or tight like it should be ?
If the input shaft was ever loose, and the friction discs got clamped in a non central position, not only would you feel this as the assembly would no longer be in balance, but the trans itself probably wouldnt last long as the input shaft would be orbiting, as opposed to spinning.
If the input shaft was ever loose, and the friction discs got clamped in a non central position, not only would you feel this as the assembly would no longer be in balance, but the trans itself probably wouldnt last long as the input shaft would be orbiting, as opposed to spinning.
The input shaft isn't loose when removed, in fact it has the same defer election as when it was built.
Wear marks on the clutch plates show the friction plate nearest the engine has deflected off center 1mm and touched the inside of the clutch basket and the friction plate itself has chips, the middle plate has deflected just over half and the gearbox side plate hardly at all.
I can't do pictures as I'm not home, but the wear marks are clear as to what's happening.... I am the only one running a tripple plate and this has tighter tolerances to the single plate unit which has more clearance in the same areas hence the noise is louder when using the tripple over the single.
The noise heard on down shift is clearly this contact and it's as though the shaft vibrates then centralises as the noise is brief on initial clutch engagement and goes on full engagement. So the clutch must center properly.
Must admit I hadn't considered that the gearbox may have this deflection built into in... I'm now leaning towards machining down the end friction plate to allow more clearance.
Wear marks on the clutch plates show the friction plate nearest the engine has deflected off center 1mm and touched the inside of the clutch basket and the friction plate itself has chips, the middle plate has deflected just over half and the gearbox side plate hardly at all.
I can't do pictures as I'm not home, but the wear marks are clear as to what's happening.... I am the only one running a tripple plate and this has tighter tolerances to the single plate unit which has more clearance in the same areas hence the noise is louder when using the tripple over the single.
The noise heard on down shift is clearly this contact and it's as though the shaft vibrates then centralises as the noise is brief on initial clutch engagement and goes on full engagement. So the clutch must center properly.
Must admit I hadn't considered that the gearbox may have this deflection built into in... I'm now leaning towards machining down the end friction plate to allow more clearance.
So how well engaged onto the splines is that disc ?
in general a thin hub as used in a multiplate will tend to have more slop in fit, just because it is looser.
So maybe it's a little too loose ?
Also, is there any chance that disc is falling off the input shaft slightly when the pedal is pressed ?
Some of the more modern multiplates with a single hub, and toothed outer friction discs is a much better design. Although it's maybe only carbon ones come like that
ie this sort of design
in general a thin hub as used in a multiplate will tend to have more slop in fit, just because it is looser.
So maybe it's a little too loose ?
Also, is there any chance that disc is falling off the input shaft slightly when the pedal is pressed ?
Some of the more modern multiplates with a single hub, and toothed outer friction discs is a much better design. Although it's maybe only carbon ones come like that
ie this sort of design
The discs are a tightish fit to the input shaft, there isnt any chance it could move by 1mm.... the clutch basket cant be moving as its the friction plate at the flywheel end thats touching and so it cant flex there as its clamped to the flywheel with a return.
the friction plates splines have 100% engagement on the input shaft splines in both instances, its close but I have checked carefully many times so there is no chance they are falling off slightly.
Also the stock clutch does this if you upgrade it (they reinforce the springs and fit a more abrasive friction plate). so its not unique to mine, only more prevalent due to tighter tolorences IMO
For reference the exact same clutch is used in mid engined cars with a different box and massive V8's with similar torque and they are silent in those cars.... the box's they use DO have a spiggot bearing that teh input shaft registers into (like most cars).
the friction plates splines have 100% engagement on the input shaft splines in both instances, its close but I have checked carefully many times so there is no chance they are falling off slightly.
Also the stock clutch does this if you upgrade it (they reinforce the springs and fit a more abrasive friction plate). so its not unique to mine, only more prevalent due to tighter tolorences IMO
For reference the exact same clutch is used in mid engined cars with a different box and massive V8's with similar torque and they are silent in those cars.... the box's they use DO have a spiggot bearing that teh input shaft registers into (like most cars).
Gassing Station | Engines & Drivetrain | Top of Page | What's New | My Stuff