Try swapping some of the same IC's around (note their original locations) and see how it might throw the Cal out.
They were clever guys back in the day...............
The original socketed chips (not the one I put a socket in for):
CA3046 Quad NPN transistors (3 chips)
74LS132 NAND gate
UA723 Vreg
LM324 Quad Op AmpNot socketed:
74S74 D flip flop
MC1514 differential comparator
LF356 JFET
741 Op Amp
Numerous transistors ...
What Tautech mentioned is intriguing. The only place where I might think they put in sockets for swapping would be the sine shaper circuit. It uses 3 quad NPN transistor arrays to make the sine wave from a triangle wave. It's possible they wanted to swap the arrays I guess, for the least distorted sine wave (but there are adjustments for that too). However that would not explain the others, which aren't really that critical, as far as a generic part. I mean a quad NAND gate swap? But I might buy it for the NPN arays.
wow, Xrunner. I remember something about sockets from long, long ago. I asked the very same question to a computer system designer/CEO/manufacturer named Bill Godbout. (late 1970's, I believe) I was wondering why he had some ic's in sockets and some soldered on a I/O board for the S100 buss that he manufactured. His reply was two fold: that it allowed for critical parts to be upgraded to more reliable and faster ones in the future if needed, and the sockets provided more decoupling capacitance in the circuit. I did not understand that last part for many years. I'm still not sure I understand it completely now.
Possibly, but I don't think a quad NAND gate is going to be upgraded.
![Smiley :)](https://www.eevblog.com/forum/Smileys/default/smiley.gif)
The decoupling capacitance - maybe, but I don't think a socket contributes that much. It would depend on the application - maybe in a high frequency RF circuit it would contribute enough to make a difference.