go to Vyper Interface
The picture looks better in full size (but look at Dave Cordes site - the same plan just better look).
Source lost - I only changed R1 to1MOhm because 560kOhm didn't work from start. U1 has to be a CMOS type.
Dave Cordes has a very good homepage about this topic. There is also a description for mechanic contacts.
I started with another way and finished that. It looks very homemade which it is - but is works.
from above and below
and from the side
The basis is a piece of persplex I had left from another project. 4
bolts (M4) give a lead to the computer case. 3 bolts (M3) give contacts.
Because of the transparent material it is very easy to find the correct
position for the contacts.
First I soldered the bolt heads as contacts. But I got better results with conductive moss rubber I fixed with wire to the bolts. It is normally used to protect CMOS ICs against static charges and every electronics shop should have it. It is smooth like the material of the original interface.
Another plate shields the electroni and cables. The cable is fixed with cable fix.
In contrast to the original I don't have to remove the protective cover.
The good software can be downloaded at Suunto .
The circuit above seems to have problems with some PCs, because the voltages are often too low. Ralph Behrens has made some modifications and I present here his new plan (click at it to get a better picture):
Patrick L'huillier sent me CAD data for that: eon_if.zip
The long awaited plan for a Vyper Interface finally came up: http://subaqua.web.cern.ch/subaqua/techniques/suunto/suunto.htm
And here my idea for the mechanics:
I screwed M3 bolts to a small plastic case, which holds the PCB. The contacts are from PCB connectors like the ones found on every PC motherboard. I soldered them to the bolts. The boltheads are also filed, to get a good fit into the opening at the Vyper. The other two bolts give a good fit. To get some contact force, just close the strap.
Of course a machined plastic part which fits neatly inside the opening with spring contacts would be the better solution but it works my way too.
there is another variation of the circuit and a plan for a casing for Suunto