[sdiy] three schematics online

[Czech Martin]

Quick and dirty (because I'm currently
modifying my site): here are the schematics:

http://www.geocities.com/SoHo/Museum/4459/pico.gif

Keithley pA-meter from 1977. Yes, 10x circuit via divider, this saves
a few resistor ranges as well. J.H. was right. Today I would use a 3140
or better for the opamp, saving the dual fet trouble....
The resistor tempco of usually 100 or 50 ppm is a pain if you
want absolute precision, but for the purpose I have in mind (expo
converter evaluation) this is not important. I just want to know
if the converter follows the exponential law and how this following
changes (design, temp). I could even live with non fitting ranges,
because if I switch the range with the same converter setting,
the current will be the same, so I can estimate the resistor network
error for the two ranges and combine the results to a steady graph again.
No auto zero offset, but I think a manuall offset trim every now and then
is not too bad. MOSFET opamps have very low Ib, but they usually drift a
lot.


http://www.geocities.com/SoHo/Museum/4459/prot.gif

Power line protection/filter circuit. Reverse eng. by me.

F: sparc gap (Epcos, Farnell)
V1,V2: Varistors (Siemens? >230V, disk type, Farnell)
S1,S2: thermal fuse (90C ?, Farnell)
G1,G2: neon bulbs

G2 (green) shows correct operation. If V1 or V2 or both are damaged,
the increased power dissipation will heat thermally connected S1 or S2,
they pop open, G1 (red) turns on, indicating varistor fault.
The path to PE is done via F, not with varistor, because little leakage
currents may trigger the ground fault interruptor. The RF filter is a
standard
package.

http://www.geocities.com/SoHo/Museum/4459/headp.gif

Elektor headphone amp. One half snipped, lower rail is of course
-15V. Input par could be changed to japanese duals...

Have fun.

m.c.