Thanks Ron!!
email (option): gino_kerekes@yahoo.co.uk
Ron
Who says random parts are universally interchangeable? I worked in a machine shop for years and found that bearing parts are not necessarily interchangeable. In fact ball bearing manufacturers routinely sort machined balls by measured dimensions specifically for individual race components. Bearing assemblies are sold as within spec assemblies, not parts. Same with firearms. Manual fitting is still needed to tune the assembly for proper functioning.
Hence my experience in having to fit the m20 clutch roller assembly. If your experience is otherwise, then count yourself lucky. I was taught to hand fit and test all mechanical assemblies before use. So my experience indicates that craftsmanship is still valued in a world of CNC machining and laser precision.
Dave W.
email (option): Dwdiak4@gmail.com
Dave firstly I think I missed that you were installing brand new parts? Never the less, BSA machined parts with an upper and lower tolerance as stated in the standards book. I agree that BSA did select parts on assembly to get the best fit and would have measured them to determine this.
The difference between the upper and lower tolerance for these clutch parts is just 0.0005" (1/2 thou) If both components were at the upper tolerance for instance, it would still give 0.001" (1 thou) clearance overall for the 1/4" rollers.
It concerned me that you that you removed several thou with a honing stone without measuring anything and suggested it as a remedy. I do agree that fitting the two together with the tighter tolerance can be tedious and I've know a roller to inadvertently tip over, which you can't see once assembled......So you take it off to check and start over again:persevere:
Ron
email (option): ronpier@talk21.com
any thoughts? tricks?
You say that the clutch hub is new, is that NOS or is it one of the newly made clutch hub. If so when fitting to an original shaft then this will be your problem. If you can assemble it and the unit feels clicky as you turn it then the clearance if any is too tight. This will need honing if possible, if not and its not too tight you can assemble it and add i very small amount of very fine grinding past and turn by hand. Wash all parts out and try the fit again with a bit of oil. Only try this if your fit is only a little on the tight side. If too tight then you will have to get it honed.
Tim W
email (option): t.j.walker@btinternet.com
Must admit, I wasn't considering after market parts with a poorer tolerance threshold. :thinking_face: Ron
email (option): ronpier@talk21.com
I'm pretty sure that in some cases (timing gears for example) BSA 'selectively assembled' critical parts, matching tight upper, mid and lower limit parts to produce the optimum fit...This method can produce fits that are incorrect when the required final tolerance is very small and the parts are not 'selected' to match each other...
I have had this problem more than once with timing gears when I have had a full set of NOS components for the timing side but the final gear mesh was too tight at assembly...
Equally it would have been the case with less critical parts that they would have been toleranced to produce a dimension that would have fallen between a minimum and maximum clearance. They would have then been selected randomly from within that range of sizes for assembly, producing a final clearance that would vary slightly but would still fit together in every case, being neither too tight or too loose for the application...
So, the manufacture of replica parts can be problematic when it comes to parts that were originally selectively assembled if all the tolerancing information is not available and only one of a pair or group of matching parts is being produced...This may well apply to the clutch sleeve, rollers and clutch bearing track which, being a bearing, may have been matched for a close and constant fit...
These days with CNC manufacture where machines can even adjust themselves to maintain very accurate dimensions over a production batch these tolerancing methods are less prevalent but in the past where it was actually physically impossible for a machine operator to produce everything to exactly the same size it was a subject in itself...Tight tolerances took more time to produce, required more complicated inspection procedures and resulted in more reject parts that fell outside the limits...
When, as part of my apprenticeship, I was being trained to do job layouts that detailed the manufacture of a part to guide the machinists on the production floor I was constantly told...'Don't over tolerance when it isn't needed!'.....Ian
email (option): ian@wright52.plus.com