Where is the loose connection?

They are thermal detection devices. If a loose connection on the overload itself causes local heating then that will be reflected in the overload tripping point being reduced.

Loose wires anywhere else, no. If a wire elsewhere is so loose as to disconnect a phase it's a crap-shoot as to whether the increased current in only two phases due to single phasing a three phase motor trips or the motor fries.

Keith Cress
kcress - http://www.flaminsystems.com

Dear Mr. Jk1996 (Electrical)(OP)1 Jan 22 21:18
[color ]"...Will a thermal overload relay trip from a loose connection or is it purely overcurrent?"
[/color]
1. The basic principle of operation of most "thermal overload relay" on the market and in the field is based on: the (heat produced by the current) [though the heating element]; directly/indirectly or through CT etc. The (heating element) [heats up another bi-metallic strip], which trips the mechanical contact. BTW: Electronic relays may differ in operation principle.
2. A loose connection
a) on the (wiring, motor etc) which [usually does NOT cause any increment on the current] will NOT cause any tripping. If there is an (increment of current) due to any reasons, [tripping will occur],
b) on the (heating element), may cause [the bi-metallic element to act prematurely]; due to the current + heat produced, due to high contact resistance.
Che Kuan Yau (Singapore)

Loose connection causes single phasing and the thermal overload relays are designed to trip on single phasing. This is true with bi-metal overload relays that have differential bar mechanism. Electronic overload relays are different.

Sadly RRaghunath frequently they don't. In fact very frequently they don't. That's why people buy phase-loss relays.

Keith Cress
kcress - http://www.flaminsystems.com

Dear Mr. RRaghunath (Electrical)2 Jan 22 04:38
"...Loose connection causes single phasing and the thermal overload relays are designed to trip on single phasing. This is true with bi-metal overload relays that have differential bar mechanism...."

1. My experience is (loose connection) does NOT causes [single phasing]. The loose connection phase current, and voltage wrt neutral are lower than the other two healthy phases. But NOT zero in value. (Single phasing) is when [one phase is cut=open circuit], where current and voltage are =0.
2. When the (motor is lightly loaded), all phase currents are [lower than the set/trip value]. The loose connection phase with current lower than the other two healthy phases. Under this very common occurrence at site , the thermal overload even with differential bar, may not trip.
Che Kuan Yau (Singapore)

IEC style bimetal thermal OL relays will all have the differential trip mechanism that MIGHT trip from single phasing the motor, but that depends on the LOAD on the motor. The differential trip mechanism BIASES the trip point to be lower than it would be under a balanced 3 phase application, but if the motor is lightly loaded, it may not trip, yet the motor may still suffer thermal; damage from negative sequence currents.

To explain what Keith said, NEMA style thermal OL relays (either bimetal or eutectic melting alloy type) used in North America DO NOT have a differential trip mechanism, so they will not bias the thermal tripping point at all, so they will not pick up a single phasing condition until is it usually too late, hence his comment on using a Phase Loss Relay.

90% of the Solid State OL relays on the market have some form of phase loss protection built in, it's just firmware for them. Usually they function on looking at a current imbalance of 20 to 30% depending on brand, then more expensive versions are adjustable.

So will an OL relay trip if a connection is loose? maybe, maybe not. You can't count on it.

" We are all here on earth to help others; what on earth the others are here for I don't know." -- W. H. Auden

Dear Mr. jraef (Electrical)3 Jan 22 05:54
"...IEC style bimetal thermal OL relays will all have the differential trip mechanism that MIGHT trip from single phasing the motor, but that depends on the LOAD on the motor. The differential trip mechanism BIASES the trip point to be lower than it would be under a balanced 3 phase application, but if the motor is lightly loaded, it may not trip, yet the motor may still suffer thermal; damage from negative sequence currents...".
1. In Europe, Bimetallic thermal OL relays are covered by IEC 60947-4-1 Contactors and motor starters....The words "overload relay" is used. The Standard covers both non-single-phasing and single-phase designs. Both types are available in the European market and used in the field. Note: NOT all have the differential trip mechanism.
2. a)Limits of (time delay) [when all poles are energized] shall comply with Table 3.
b)Limits of (time delay of three-pole thermal overload relays) [energized on two poles shall] comply with Table 4. Note: the test is conducted with 2 poles at 1.15 and 1 pole at 0 multiple of current setting. The tripping shall occur in less than 2h.
3. There is NO guide line on the (tripping delay) [when 2 poles are < and 1 pole is << then the setting].
Che Kuan Yau (Singapore)

jraef,
comprehensive reply as always..