HP ‘Bill West’ switch springs – simple process for making replacements

[Bill Woodbridge]

In the course of my HP 8566B repair / refurbishment, while cleaning the keyboards I noticed that a leaf spring from one of the famous ‘Bill West’ switches was missing.  This was a surprise, but it was a fairly obscure key on the IF/Display unit that I hadn’t yet used in my initial playing-about with the instrument, so I hadn’t previously noticed it.
 
I thought I’d post my successful recipe for making a new leaf spring, as most of the online comments I’ve found regarding BW switches and their leaf springs seem to start by lamenting their non-availability as spares but without giving further details as to how the problem could be solved (other than by cannibalisation of other units), or else go into metallurgical deep-dives.
 
I started by removing a good spring which was easy enough with tweezers.  The best online tip I found was a suggestion of using a 0.002” feeler gauge as raw material, and (having extracted my ancient Imperial feeler gauge set after 40 years sitting in an old car toolkit) this did indeed look very promising.  In fact I am pretty sure the original spec for the springs must have been 0.002” stock, and without getting too metallurgical the feeler gauge’s metal seemed to have a good match to hardness, springiness and looking ‘stainless but not too stainless’.  The problem was how to cut the resilient and springy feeler gauge using basic garage-level tools into a small part with sufficient accuracy.  As expected, various attempts with scissors, tin snips and wire cutters (even my excellent Lindstroms) either did not cut at all or left mangled and twisted edges (and with such small parts, a mangled edge is effectively a mangled part).

The penny dropped when I was marking out another area of the feeler gauge for yet another attempt, using the original spring as a template.  I applied a bit more pressure to my tungsten-carbide tipped scriber this time, and noticed the metal starting to bend upwards at the scribed line in a rather neat and controlled fashion.  So, having replaced the original spring with a small engineering steel rule clamped down to the feeler gauge to give a better edge for the scriber to run down, I carried on scribing the line with moderate pressure and after about twenty passes, the metal completely parted with a very neat, precise edge and only microscopic distortion.

So there’s the answer: a 0.002” feeler gauge, just repeatedly scribe out to the correct dimensions with a tungsten-carbide tip, and no need for any subsequent cutting or even fatigue-bending of the metal to achieve the final parting.  And of course the bonus is that scribing is far more precise than cutting anyway, so I’ve been able to reproduce the dimensions to well within 0.5mm accuracy – and I found it needed to be this good for them to sit properly in the tiny plastic slots in the white plunger and red body.  Slotting my new spring into the BW switch gave perfect results with just the same snappy ’S’ action as the originals.

The only remaining worry is the whereabouts of the remains of the old failed spring, presumably floating about somewhere in a unit which boasts 18kV amongst other exciting voltages … I’ll take all the panels off (again) and try the endoscope, but if there’s no joy there I’ll just have to let sleeping dogs lie as I don’t want to embark on a full disassembly.