Strange Phasor Reading

Your toys are too fancy for your understandings.

you are right, that is why I ask for some opinion from someone who have experience something like this. I better ask for something I dont know rather than stay quiet to hide my stupidity.

Which wiring was changed? Did the pump run OK before the test? Does the pump run under load after the wiring was changed? Or was only the wiring of the measurement changed?

Its a bit "wiring". One thing for sure its not who has the biggest car its also who can drive it!

Do you or your contractor have a test procedure for this?

...and what sort of load is on the motor? Is it fully loaded or not?

The power factor of your motor is correct. I have done some test on one of the factory induction motor and the PF is register at 0.5. So we use capacitor to improved the PF of the motor to 0.85. This is because the induction of the motor when running.

Then what is the meaning of nameplate? because power factor in nameplate shows 0.8

If we consider actual [sinfi]=registered p.f.[cosfi] and recalculate the actual p.f.=sqrt(1-sinfi^2) on each phase then we get 87.1 kw total active power supplied and if 75 kw is rated then efficiency will be 75/87.1=86% [reasonably].

No possibility of wrong exterior connection[switching phases or "in" to "out" or supply voltage R-Y instead of R-N -in this case the total power would be sqrt(3) times more.

In my opinion the wattmeter circuit has an additional large reactance parallel with the measuring circuit and turn it into varmeter[measuring sin(I,V) instead of cos(I,V)].

If you would use voltage transformers for ARN Power Meter of 400/230 V a wrong wiring could occur!

The PT from 400/110. Dyn5 connection.

"...because portable generator nameplates shows Power factor is 0.8."

Your generator plate shows n.

Now you have loaded it with an inductive load.

Your power factor has decreased from an ideal load to a real-world load.

You may adjust your PF as stated by another, using capacitance.

If the voltage at terminal of the meter is supplied by voltage transformers of 400/230 the supply voltage will lag the current with 90-fi and the meter will show sin(fi) instead of cos(fi).

Instead of R-N [the same phase as current IR it is]it is supplied through transformer Y-B[or S-T] then the angle measured is 90-f .

Only 230/230 transformer will solve the problem.

Yesterday we bypass the PT and the reading is correct (same with our 3phase multimeter) there is possibility it was caused phase shifting by PT. The PT vector group is Dyn5, so the phase would leading 150゜. Contractor will try using Yyn0 for the PT.