Correct part.
Obsolete. No replacement.
Only way to buy these now is through Chinese questionable sources which is why I don't stock them.
It's possible Marco got a bad batch but it is also possible that you are comparing two different die revisions to each other.

Measuring between pins 4 and 5 is actually measuring the difference between two independent outputs. That measurement won't tell you anything.

Actually a solid state relay with internal control circuit and logic - no easy way to test these.

Probably yes.
Do as Chris said -- swap in a known good one there to be sure that the problem is with the new SSRs you are getting are good or bad.
Or safer -> Pull a known good one out, put it into static shielding bag, put new SSR in that location and make sure that column still works. If not then the new SSRs are bad. If new SSR works for this column then the column driven by U15 may be shorted.
Important - these have FET transistors on the outputs. These are VERY susceptible to ESD (static) damage. These should always be stored within static shielding bags (silver or black) and not anti-static bags (pink) and never regular bags (clear).

Going from your information:
The differences were:
1. Voltage across 47ohm resistors (Pin 1 to board ground) was something like 0-.3vdc on U17 but almost zero on U15
This says the part is consuming nearly no power... but it consumes very little power anyways.

2. Output voltage was 6-11vac on U17 and .6-1 on U15
Points to U15's internal FET not turning on (bad)

3. Output voltage when unplugged was slightly different on U15 .03-.05vac
Points to U15's internal FET not turning on (bad)