It surely sounds like the op-amp is not operating. Make sure pin 8 is grounded, and that the .1 uF cap appearing after the isolation diode is mounted close to terminals 4 and 8 of the chip. If so, then likely the op-amp is bad, or has been damaged.
The two 47K resistors create a voltage divider whose output at their junction point is equal to 1/2 of the supply voltage appearing after the isolation diode. Accounting for the load of your meter, it sounds like all is good with that. The output of the voltage divider is then used to bias the output of the op-amps at 1/2 of the supply voltage (by way of the 1 MΩ resistors), which is not happening: pins 1 and 7 will normally show a voltage of about 1/2 the supply voltage, and apparently they don't.
If all is good with your wiring and connections, then the chip is suspect. As suggested, this should be connected into the circuit by way of a socket, so that no direct soldering is made to its terminals. It also makes for easy change out if that becomes necessary.
I hope this helps!
Dave
The two 47K resistors create a voltage divider whose output at their junction point is equal to 1/2 of the supply voltage appearing after the isolation diode. Accounting for the load of your meter, it sounds like all is good with that. The output of the voltage divider is then used to bias the output of the op-amps at 1/2 of the supply voltage (by way of the 1 MΩ resistors), which is not happening: pins 1 and 7 will normally show a voltage of about 1/2 the supply voltage, and apparently they don't.
If all is good with your wiring and connections, then the chip is suspect. As suggested, this should be connected into the circuit by way of a socket, so that no direct soldering is made to its terminals. It also makes for easy change out if that becomes necessary.
I hope this helps!
Dave