Retorquing

[ModbusMan]

A properly torqued connection should not need to be retorqued.
IR scanning is a better tool to locate suspect connections.

This has been our practice... correctly torque during installation and mark with grease pen so you can visually identify if it's working loose. After that, annual IR scanning.

 

[Pinnie]

I have heard this as well. The manufacturer takes into account the metal relaxation after you torque. When they give a specified torque value you torque to that value and the metal relaxes to the actual torque value they want the termination to be. You should not "re-tourque" terminations as you would then be over torquing the conductor. If you want to verify torque value cut the conductor, strip, and re-terminate and torque. (IMO)

Ryan Jackson video that elaborates more. its 2020 but still applies:

 

[Fred B]

Here is what I see as the break down why retorque is only to 90%
The CU or AL conductor is softer than the screw or pressure connector it is being attached to and will after a very short time continue to slacken after initial compression. (Called cold flow) The steel alloy set screw does not loosen and the marking is just an indicator you have torqued.
Mark the screw or lug after initial torquing, and sometimes in as little as 10 minutes if were to go back at full torque setting you will get a slight additional turn, this can be done over and over again (each time it may take a little longer) until you finally compressed the wire and it breaks. The 90% won't over stress the wire or over torque but will indicate you are tightened and shouldn't move the connector. Add, over time, the heating of the connection and wire even minimally, will further softens the conductor or the expansion will increase the effective torque and added potential cold flow. Copper's tensile strength decreases by about 5% at 100°C while steel only looses tensile strength by about 3% at 200°C.
Retorqueing of the heating effected connection at 90% will only bring back to a minimum connection pressure, not over torque provided no additional overheating was to occur

 

[LarryFine]

Plus the pre-deformed conductor has greater contact area than a new conductor, so slightly less pressure is necessary for a low-resistance connection.

 

[pipe_bender]

Here is what I see as the break down why retorque is only to 90%
The CU or AL conductor is softer than the screw or pressure connector it is being attached to and will after a very short time continue to slacken after initial compression. (Called cold flow) The steel alloy set screw does not loosen and the marking is just an indicator you have torqued.
Mark the screw or lug after initial torquing, and sometimes in as little as 10 minutes if were to go back at full torque setting you will get a slight additional turn, this can be done over and over again (each time it may take a little longer) until you finally compressed the wire and it breaks. The 90% won't over stress the wire or over torque but will indicate you are tightened and shouldn't move the connector. Add, over time, the heating of the connection and wire even minimally, will further softens the conductor or the expansion will increase the effective torque and added potential cold flow. Copper's tensile strength decreases by about 5% at 100°C while steel only looses tensile strength by about 3% at 200°C.
Retorqueing of the heating effected connection at 90% will only bring back to a minimum connection pressure, not over torque provided no additional overheating was to occur

Nice explanation thanks.
The facility we worked on had a firm IR scan everything, and we only re-torque the AL terminations that are the lug type and its to 90%.
I never understood why just the aluminum got retorqued, but everything got scanned.

 

[winnie]

What is the chance that when retorquing is required, the connections have lugs crimped to the wires, and the lugs are bolted to bus bars?