On my Herbert high speed drill, I’m just in the process reassembling the belt pulley support assembly that drives the spindle. I have found that the three clamp nuts are rather tight on the shaft threads making the unit clumsy to assemble. Worse still, it will make changing the pulley when the unit is installed in the drill rather challenging. (A previous owner had clearly faced this problem got around it by holding the pulley with pliers!)  On the threads there is no obvious thread damage that can be cleaned up by hand so I thought I might look for a cheap die/tap for the job, as the cause of the tightness must be a thread diameter or form issue.

Here is the assembly:-

Identifying the threads is proving difficult however. I have measured the TPI as 22.01.  When I check against various taps, 20 and 24 TPIs don’t mesh. I have an ancient Helicoil tap marked 22TPI and this fits perfectly. Clearly therefore, 22TPI is what my two threads are.

The OD of the threads measures 0.4358″ which is pretty close to 7/16″ (0.4375″). The microscope shows the threads are a little flat-topped and this could account for the difference.

When I consult the popular thread data table to determine what I might have, there is no obvious match with the above.

The nearest, in terms of TPI are the UNF, Cycle thread and at the other end, the ‘Admiralty’ thread. But no 22 TPIs in sight!

Does anyone have any thoughts as to what Herbert have done here, and why ?

The curiosity is that the two shaft support bearings are 1201’s. (Confirmed correct by the parts manual.) These are double-row ball self-aligning types. This is a very odd choice because the there is no need for a self-aligning capability in this set-up, and this type have very poor load capacity, by nature of their essentially flat raceway in the outer ring. (Infact one of these bearings is already showing raceway fatigue.) I will therefore replace them with conventional 6201’s for a much lower cost.

Gerry