When using a 3-phase motor and a bi-metallic overload is it recommended to use the internal Variable Frequency Drive overload protection? Does the internal VFD overload protection change with as the motor speed/amps change? Could a motor overload at 50% speed damage the motor?
Tek-Tips is the largest IT community on the Internet today!
Members share and learn making Tek-Tips Forums the best source of peer-reviewed technical information on the Internet!
-
Congratulations MintJulep on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!
3-Phase motor VFD overload protection 5
- Thread starter webcntl
- Start date
The VFD control generally has built-in overload protection (we're talking 480V type motors) and that is what we generally use.
External overload relays can be used, but may not be as accurate in reflecting actual heating conditions in the motor as the algorithms used in the drive. The heater in the OL relay may not respond to the high-frequency pulses in the same way as an iron core motor.
If the motor is of any size (>50 hp) we generally specify an embedded temperature detector or temp switch in the stator windings. This is the best overload protection.
External overload relays can be used, but may not be as accurate in reflecting actual heating conditions in the motor as the algorithms used in the drive. The heater in the OL relay may not respond to the high-frequency pulses in the same way as an iron core motor.
If the motor is of any size (>50 hp) we generally specify an embedded temperature detector or temp switch in the stator windings. This is the best overload protection.
-
2
- #3
I would suggest discarding the overload block and using the drive software. The overload block calculates heat based on the assumption that the motor is turning full speed. At all other speeds, the block will be in considerable error. On the other hand, most drives provide for derating for reduced speed and some even permit customizing the derate curve.
If the motor has thermal switches, forget the drive software and set up an external fault loop on the drive with the thermal switches in the loop. This is always preferred if the switches are already in the motor. It isn't economical to add them to a motor unless the very last 2% of motor thermal capacity needs to be squeezed out of it.
If the motor has thermal switches, forget the drive software and set up an external fault loop on the drive with the thermal switches in the loop. This is always preferred if the switches are already in the motor. It isn't economical to add them to a motor unless the very last 2% of motor thermal capacity needs to be squeezed out of it.
I am not that familiar with the thermal switches embedded in the motor but it seems these switches would have a temp setting below that of the motor operating temperature range to protect the motor. Otherwise, if the motor is heated up to max operating temp (temp setting of the switch)the damage is already done to the motor. Current monitoring (or power monitoring) is always better, in my opinion, to protect motors from overload.
I would use the drive overload capability before I rely on temperature devices. Temperature devices are used as a back-up protection in the event the overload relay fails. I always said if the temperature rises, the damage is already done. Of course, this is all based on getting the ultimate life out of your motor without concerns for the process.
If the temp switch is set just below the max operating temp then this would be satisfactory (and maybe they are, I don't know because I generally don't use them as primary protection).
Interesting discussion.
I would use the drive overload capability before I rely on temperature devices. Temperature devices are used as a back-up protection in the event the overload relay fails. I always said if the temperature rises, the damage is already done. Of course, this is all based on getting the ultimate life out of your motor without concerns for the process.
If the temp switch is set just below the max operating temp then this would be satisfactory (and maybe they are, I don't know because I generally don't use them as primary protection).
Interesting discussion.
-
1
- #6
Skogsgurra
Electrical
buzzp,
"if the motor is heated up to max operating temp (temp setting of the switch)the damage is already done to the motor. Current monitoring (or power monitoring) is always better, in my opinion, to protect motors from overload"
I don't agree there. It doesn't harm the motor to run at its maximum temperature. It is designed to do just that. But, since the thermal switches (or temperature sensors) probably do not sense the hot spot in the motor (motor manufacturers say that they do, so perhaps they really do that) there might be a five or ten centigrade difference between sensor temperature and hot spot.
Even if that is so, the motor life (insulation life) will be only reduced by 50 percent if run continously at ten centigrades above design temperature. So, there is plenty of head-room if you really want to squeeze the last FHP out of the motor.
Normally, other things limit the temperature. Bearings for instance usually cannot take more than 110 or 120 centigrades and since the rotor/shaft usually is hotter than the winding, it makes operation at 155/180 centigrades (class F/H) hot spot temperature impossible. If the motor manufacturer has fitted temperature sensors or switches, you get the most out of the motor if you avoid any other thermal protection and rely on them.
"if the motor is heated up to max operating temp (temp setting of the switch)the damage is already done to the motor. Current monitoring (or power monitoring) is always better, in my opinion, to protect motors from overload"
I don't agree there. It doesn't harm the motor to run at its maximum temperature. It is designed to do just that. But, since the thermal switches (or temperature sensors) probably do not sense the hot spot in the motor (motor manufacturers say that they do, so perhaps they really do that) there might be a five or ten centigrade difference between sensor temperature and hot spot.
Even if that is so, the motor life (insulation life) will be only reduced by 50 percent if run continously at ten centigrades above design temperature. So, there is plenty of head-room if you really want to squeeze the last FHP out of the motor.
Normally, other things limit the temperature. Bearings for instance usually cannot take more than 110 or 120 centigrades and since the rotor/shaft usually is hotter than the winding, it makes operation at 155/180 centigrades (class F/H) hot spot temperature impossible. If the motor manufacturer has fitted temperature sensors or switches, you get the most out of the motor if you avoid any other thermal protection and rely on them.
-
1
- #7
I want to add 3 things to this discussion.
1) Not all VFDs include what most everyone accepts as motor Thermal Overload protection, especially older ones. It is not safe to assume that it is there unless the operations manual specifically describes "inverse time/current" (I^2t) type Overload Protection. Some have a feature that they CALL "thermal overload", but is likely a mistranslation and is just an overcurrent trip feature preset at a fixed value which may allow siginificant motor damage. It is probably for that reason that some motor manufacturers suggest imbedded thermal devices in the motor such as PTC resistors or RTDs, as a backup in case the VFD manufacturer has used a loose interpretation of the word "Overload".
2) If there is a bypass scheme included in the VFD package, the motor Thermal Overload Protection must be a separate protection relay so that it is in the circuit should the VFD be removed for servicing while the motor continues to run full speed.
3) While PTCs are an accepted method of OL protection, I have that opinion that they are more usefull as a backup device than as a primary source of thermal management. PTC protection relays are threshold trip devices. Granted bimetal OL relays are as well, and I will accept that the PTC may provide closer control of the true thermal condition of the windings, but only when compared to bimetal OL relays. IMHO, the best OL protection now comes from the "Thermal Model" algorithms used in many solid state OL protection devices, especially those that can provide biasing from current imbalance. They can provide warnings of impending OL, cooling rate variables and retentive memory of the motor thermal state without the need for devices in the motor windings, which if not ordered at the time of manufacture of the motor may be difficult to retrofit.
"Venditori de oleum-vipera non vigere excordis populi"
1) Not all VFDs include what most everyone accepts as motor Thermal Overload protection, especially older ones. It is not safe to assume that it is there unless the operations manual specifically describes "inverse time/current" (I^2t) type Overload Protection. Some have a feature that they CALL "thermal overload", but is likely a mistranslation and is just an overcurrent trip feature preset at a fixed value which may allow siginificant motor damage. It is probably for that reason that some motor manufacturers suggest imbedded thermal devices in the motor such as PTC resistors or RTDs, as a backup in case the VFD manufacturer has used a loose interpretation of the word "Overload".
2) If there is a bypass scheme included in the VFD package, the motor Thermal Overload Protection must be a separate protection relay so that it is in the circuit should the VFD be removed for servicing while the motor continues to run full speed.
3) While PTCs are an accepted method of OL protection, I have that opinion that they are more usefull as a backup device than as a primary source of thermal management. PTC protection relays are threshold trip devices. Granted bimetal OL relays are as well, and I will accept that the PTC may provide closer control of the true thermal condition of the windings, but only when compared to bimetal OL relays. IMHO, the best OL protection now comes from the "Thermal Model" algorithms used in many solid state OL protection devices, especially those that can provide biasing from current imbalance. They can provide warnings of impending OL, cooling rate variables and retentive memory of the motor thermal state without the need for devices in the motor windings, which if not ordered at the time of manufacture of the motor may be difficult to retrofit.
"Venditori de oleum-vipera non vigere excordis populi"
I am not trying to downplay the importance of embedded protection. I just don't know where the typical trip point is and as you stated, the hot spots within the windings are almost certainly not accounted for. I have seen many failures with embedded temperature devices as well. One question, does the embedded thermal devices meet requirments for OL protection as addressed by the NEC?
buzzp,
"One question, does the embedded thermal devices meet requirments for OL protection as addressed by the NEC?"
I have had several interesting "conversations" with inspectors about that very subject. Apparently it does, by exception to a rule (somewhere in article 430, I need to get my NEC book back from someone to check the number) that says (paraphrased) ... each motor winding shall have inverse time/current overload protection, except when... supplemental protection devices are imbedded in the motor windinds that provide tripping on direct temperature measurment. That is a big paraphrase because I can't actually read it at the moment, but the subject comes up when discussing IEC overload protection devices that only observe current in 2 phases. That scheme was disallowed by the NEC back in the 1970's, but IEC manufacturers get away with using it in equipment that they sell in the US by providing PTC inputs. The fact that 99% of motors used in the US do not have PTCs in them seems to get overlooked by the installers AND inspectors.
"Venditori de oleum-vipera non vigere excordis populi"
"One question, does the embedded thermal devices meet requirments for OL protection as addressed by the NEC?"
I have had several interesting "conversations" with inspectors about that very subject. Apparently it does, by exception to a rule (somewhere in article 430, I need to get my NEC book back from someone to check the number) that says (paraphrased) ... each motor winding shall have inverse time/current overload protection, except when... supplemental protection devices are imbedded in the motor windinds that provide tripping on direct temperature measurment. That is a big paraphrase because I can't actually read it at the moment, but the subject comes up when discussing IEC overload protection devices that only observe current in 2 phases. That scheme was disallowed by the NEC back in the 1970's, but IEC manufacturers get away with using it in equipment that they sell in the US by providing PTC inputs. The fact that 99% of motors used in the US do not have PTCs in them seems to get overlooked by the installers AND inspectors.
"Venditori de oleum-vipera non vigere excordis populi"
Thanks Jraef. I always wondered about that but never investigated it.
I looked it up, quickly, in the NEC and it looks like they call these devices "thermal protectors", integral to the motor. They are allowed but they give trip settings based on the currents (not temperature)in tables 430.148-150.
Motor FLA <9A = 170%
Motor FLA 9.1A to 20A = 156%
Motor FLA >20A = 140%
The tables mentioned are the FLA tables as assigned by the NEC. So that means they allow a higher current to pass through the motor if the overload protection (not the device that opens the circuit) is integral to the motor versus an overload that trips by measuring the current directly. This seems to be a contradiction to the manufacturers recommendations. However, if the manufacturer uses these within the motor, I would imagine they would set them to trip above FLA but less than service factor amps (at least I would hope).
This was only a quick investigation so there maybe other exceptions.
I looked it up, quickly, in the NEC and it looks like they call these devices "thermal protectors", integral to the motor. They are allowed but they give trip settings based on the currents (not temperature)in tables 430.148-150.
Motor FLA <9A = 170%
Motor FLA 9.1A to 20A = 156%
Motor FLA >20A = 140%
The tables mentioned are the FLA tables as assigned by the NEC. So that means they allow a higher current to pass through the motor if the overload protection (not the device that opens the circuit) is integral to the motor versus an overload that trips by measuring the current directly. This seems to be a contradiction to the manufacturers recommendations. However, if the manufacturer uses these within the motor, I would imagine they would set them to trip above FLA but less than service factor amps (at least I would hope).
This was only a quick investigation so there maybe other exceptions.
buzzp
What I was refering to was Table 430.37, where at the bottom calls for 3 overload devices for 3 phase motors, one in each phase. There is an asterisk next to that which allows fewer than 3 devices "...where overload protection is provided by other approved means." That exception then can harken back to what you saw in the section above. This is the technical glitch that some IEC style device manufacturers are exploiting by providing PTC input capability. The fact that the motors do not have PTC resistors imbedded seems to go largely unoticed by inspectors, so they (the mfgrs) are getting away with a cheaper system than is technically allowed.
The soft starter sold by Baldor is a prime example. It is made by Fairford Electronics in the UK, and has only 2 CTs looking at motor current. When Baldor sells it in the US, they tactfully advise the user (only in the manual) that they must provide an externbal OL relay in order to meet code. When Fairford sells it direct or through some of the other "energy saver" outfits hawking it, no such notation exists. It bears a UL label so everyone thinks it is fine to install, but in fact it technically does NOT meet code unless an additional OLR is added (or the user uses PTCs). The same is true for several VFD manufacturers. They will only have 1 or 2 Hall effect transducers on the output of the VFD, which technically is not allowed by the NEC, but because a PTC input exists it can pass UL and get labeled.
"Venditori de oleum-vipera non vigere excordis populi"
What I was refering to was Table 430.37, where at the bottom calls for 3 overload devices for 3 phase motors, one in each phase. There is an asterisk next to that which allows fewer than 3 devices "...where overload protection is provided by other approved means." That exception then can harken back to what you saw in the section above. This is the technical glitch that some IEC style device manufacturers are exploiting by providing PTC input capability. The fact that the motors do not have PTC resistors imbedded seems to go largely unoticed by inspectors, so they (the mfgrs) are getting away with a cheaper system than is technically allowed.
The soft starter sold by Baldor is a prime example. It is made by Fairford Electronics in the UK, and has only 2 CTs looking at motor current. When Baldor sells it in the US, they tactfully advise the user (only in the manual) that they must provide an externbal OL relay in order to meet code. When Fairford sells it direct or through some of the other "energy saver" outfits hawking it, no such notation exists. It bears a UL label so everyone thinks it is fine to install, but in fact it technically does NOT meet code unless an additional OLR is added (or the user uses PTCs). The same is true for several VFD manufacturers. They will only have 1 or 2 Hall effect transducers on the output of the VFD, which technically is not allowed by the NEC, but because a PTC input exists it can pass UL and get labeled.
"Venditori de oleum-vipera non vigere excordis populi"
-
1
- #12
The rationale for 3-pole 3ø motor-overload protection seems like something that electric utilities may have ‘legislated’ in the early 1970s. A scenario is motor(s) served by a transformer set of the wye-delta or delta-wye configuration. If a single-phase {e.g.; open-fuse} incident occurs on the transformer primary, it causes an increase in negative-sequence current in the stators of served 3ø induction motors.
The phase currents can theoretically and practically become severely unbalanced, with two phases running at 1 per-unit current, but another operating at 1.73 per-unit current. Because it is possible that any primary-side fuse could operate, there is a 1 in 3 chance that 2-pole overload relays will not respond to hiside single-phase conditions.
In an attempt to keep things basic and simple, here is what I would advise for motor thermal protection on VFD's in the US:
1. An overload block is generally unnecessary on modern drives because the drive will always provide overload protection software. This software is preferred over con-ventional overload blocks because a derate for slower speeds can be built in whereas an overload block assumes 60Hz operation and will calculate too little heat at slower speeds due to motor shaft fans turning slower.
2. If motor thermal switches are available, always use them. In the vast majority of cases, you can turn the software protection off. I have never lost a motor to overload heat with thermal switches of the proper temp for the insulation class installed. If you want to be extra conservative and risk some nuisance overload faults, run with the thermal switches and the software both.
3. If a motor without thermal switches is to be fitted with thermal switches, be sure the motor shop installs switches rated for the proper insulation class. Then, use the motor as in #2 above.
The above guidelines are for NEMA motors. In sizes above NEMA, if often becomes necessary to add bearing temp detection as well as winding protection. Since most over-NEMA motors are built to application by the motor manufacturers, their input on proper protection is preferred since they are providing the warranty. Go with their recommendations.
1. An overload block is generally unnecessary on modern drives because the drive will always provide overload protection software. This software is preferred over con-ventional overload blocks because a derate for slower speeds can be built in whereas an overload block assumes 60Hz operation and will calculate too little heat at slower speeds due to motor shaft fans turning slower.
2. If motor thermal switches are available, always use them. In the vast majority of cases, you can turn the software protection off. I have never lost a motor to overload heat with thermal switches of the proper temp for the insulation class installed. If you want to be extra conservative and risk some nuisance overload faults, run with the thermal switches and the software both.
3. If a motor without thermal switches is to be fitted with thermal switches, be sure the motor shop installs switches rated for the proper insulation class. Then, use the motor as in #2 above.
The above guidelines are for NEMA motors. In sizes above NEMA, if often becomes necessary to add bearing temp detection as well as winding protection. Since most over-NEMA motors are built to application by the motor manufacturers, their input on proper protection is preferred since they are providing the warranty. Go with their recommendations.
SquareD's pamphlet on VFDs recommends using the internal software protection, a conventional overload relay that responds to motor current, AND the thermostats that are inside of the motor. If the motor has separately powered cooling the overload relay for the auxiliary cooling motor also needs to be rigged to shut of the bigger motor AND you should also use a sail switch of other flow switch to confirm that cooling air really is flowing.
In 1 application where I had a gearmotor rewound with 120 degree celsius thermostats ( to protect the lamination insulation which was still only class B) I ended up having to set the conventional overload relay at 5/6 of the motor rating to protect the GEARBOX!
SquareD's concept is that software is not fully reliable, you need the conventional relay both for backup protection and to protect mechanical components from overload, and the motor thermostat should be used only for protection against lack of cooling.
For most applications a bimetallic relay with single phase differential will also protect the motor from partial single phasing such as what happens with a small amount of copper oxide in a wiring joint. The high frequency carrier should not change the calibration of the relay particularly if the relay is immediately downstream of an inductance that is used to suppress the high frequency carrier and reduce wiring capacitance problems.
Mike Cole, mc5w@earthlink.net
In 1 application where I had a gearmotor rewound with 120 degree celsius thermostats ( to protect the lamination insulation which was still only class B) I ended up having to set the conventional overload relay at 5/6 of the motor rating to protect the GEARBOX!
SquareD's concept is that software is not fully reliable, you need the conventional relay both for backup protection and to protect mechanical components from overload, and the motor thermostat should be used only for protection against lack of cooling.
For most applications a bimetallic relay with single phase differential will also protect the motor from partial single phasing such as what happens with a small amount of copper oxide in a wiring joint. The high frequency carrier should not change the calibration of the relay particularly if the relay is immediately downstream of an inductance that is used to suppress the high frequency carrier and reduce wiring capacitance problems.
Mike Cole, mc5w@earthlink.net
- Status
Similar threads
- Question
- Locked
- Question
- Question
- Locked
- Question