If You Think a Temperature Sensor Will Always Protect a Servomotor from Overheating Think Again

If You Think a Temperature Sensor Will Always Protect a Servomotor from Overheating Think Again Richard Welch Jr. Consulting Engineer (welch022@tc.umn.edu) Introduction Consult the data sheet for a typical Brushless DC (BLDC) Servomotor and one finds the manufacturer normally publishes both a Continuous or Safe Operating Area torque-speed curve (SOAC) plus an intermittent PEAK curve for each motor [1, 2, 3 & 4]. The correct interpretation for the SOAC is it defines a torque-speed boundary within which the motor can operate safely and indefinitely without exceeding its maximum continuous operating temperature when powered by a specified Drive and subjected to a specific ambient condition [2]. Regarding the PEAK torque-speed curve most servomotor manufacturers specify at least 2:1 peak to continuous torque ratio [1] while some allow an even higher 4:1 or 5:1 ratio [2, 3 & 4]. However, commanding a servomotor to output peak torque greater than it s 1X maximum continuous value for too long a time will definitely cause the motor to overheat. For example, 4X peak torque corresponds to 16X power dissipation in the motor s electrical winding since torque output increases linearly with current while the electric resistance power dissipation amounts to I 2 R where I is the current (amp) and R (ohm) is the winding s resistance. Hence, commanding a servomotor to output peak torque is normal and allowed but the Duty Cycle must be kept below 100% or the motor s winding can overheat and possibly even burn up! In trying to protect a servomotor s electrical winding from overheating, during all possible modes of operation, the manufacturer typically places a temperature sensor inside the motor and, space permitting, attaches this sensor directly to the winding [5 & 6]. The main purpose of this temperature sensor is to inform the Drive when the winding s dynamic temperature reaches its maximum allowable value and in turn the Drive is supposed to shut off the power being supplied to the motor and protect it from overheating. In some motors the manufacturer goes so far as to place a temperature switch in series with each phase of the motor s multi-phase electrical winding [6] in compliance with the UL 2111 overheating protection standard [7] to make sure the motor won t overheat. However, after extensive research I ve determined that even a temperature sensor attached directly to a servomotors electrical winding [6] won t always protect the motor from overheating and the purpose of this paper is to show you graphically why this can happen in the real world of servomotor operation. 1

Two-Parameter Thermal Model For over 50-years and even today electric motors are thermally characterized using what s generally called the two-parameter thermal model and I first encountered this model in the Electro-Craft Engineering Handbook [8]. The two-parameter thermal model assumes the motor has one, dynamic operating temperature plus one value for it s winding to ambient thermal resistance, R th ( C/watt) in parallel with it s thermal capacitance, C th (joule/ C) analogous to a simple R-C electrical circuit. Solving this two-parameter thermal model for both constant power dissipation heat-up and zero power dissipation cool-down one finds, analogous to the electrical R-C circuit, the motor both heats up and cools down in a well-known exponential manner with a thermal time constant, τ(sec), such that τ = R th C th. Hence, consulting a servomotor s data sheet one generally finds the manufacturer specifies values for both R th and τ allowing you to calculate the motor s thermal capacitance thereby completing the two-parameter thermal model. Although this two-parameter thermal model is still used extensively to calculate dynamic winding temperature during all possible modes of servomotor operation, experimental measurement shows its NOT very accurate in predicting winding temperature when greater than 1X continuous current is supplied to a servomotor motor. Hence, to overcome this inaccuracy the much more accurate four-parameter thermal model has been developed [9]. Four-Parameter Thermal Model After considerable research and comparing measured winding temperatures against those calculated using the two-parameter thermal model, I concluded that the two-parameter thermal model isn t accurate enough in predicting dynamic winding temperature during typical servomotor operation. The problem is this two-parameter model assumes the entire motor has one value (including the winding) for its dynamic operating temperature while actual measurement shows this isn t true. In fact, actual measurement proves that even within the winding there can be significant temperature differences and the two-parameter model doesn t account for these differences. Furthermore, thermodynamics teaches that for heat power to flow from within the motor towards its outer exposed surface area, and ultimately into the surrounding ambient environment, there must be a temperature gradient within the motor plus between the motor and the ambient environment. Depending on motor size and operating temperature there can be as much as a 30 C 50 C temperature difference between the motor s electrical winding and its exposed outermost surface area and this difference can t be ignored. Hence, after much research I concluded a higher order [i.e., 4, 6, 8, parameter] thermal model was needed and this model must allow the motor s electrical winding to have its own dynamic operating temperature along with its own thermal resistance and thermal time constant that differs from the rest of the motor that I call the Case. Ultimately, after more research I concluded that a fourparameter thermal model provides sufficient accuracy to explain all the measured temperature data plus it is end user friendly and fairly easy to obtain the necessary four parameter values [9]. Using my four-parameter thermal model along with the four measured parameter values for a particular 60mm diameter servomotor, Figure 1 shows the dynamic temperatures, of both the Winding and the Case [i.e., rest of the motor], during 1X constant power dissipation heat up. 2

60mm Motor - Winding and Case Heat-Up with 1X Constant Power Dissipation Solid Red = Winding Temperature Solid Blue = Case Temperature Figure 1 As shown in Figure 1, with 1X constant power dissipation the Winding s temperature (solid Red line) begins to rise immediately from its initial 25 C ambient temperature. However, looking at the Case temperature (solid Blue line) there is a time lag in its temperature rise and this is a key point to understand when we later discuss why a temperature sensor won t always protect the motor from overheating and especially if the sensor isn t attached directly to the winding. Also notice for this particular 60mm servomotor the Winding s temperature ultimately stabilizes at its rated130 C maximum continuous value while the Case temperature stabilizes at 100 C when surrounded by 25 C ambient air temperature. Hence, given this 30 C winding to case temperature gradient one issue for a motor manufacturer is where to locate the motor temperature sensor and what type of sensor does one use such that the motor is always protected from overheating but you also don t want nuisance shut-down of the motor? 3

Next, in Figure 2 we directly compare the Winding s temperature rise only for this same 60mm servomotor, with 1X continuous power dissipation, as calculated by both the two-parameter and four-parameter thermal models. 60 mm Motor, Winding Heat-Up with 1X Constant Power Dissipation Solid Red = Four-Parameter Model Dash Black = Two-Parameter Model Figure 2 As shown in Figure 2, the winding s temperature rise calculated by the four-parameter model (solid Red line) does indeed rise faster than the temperature rise calculated using the twoparameter model (dash Black line). However, as one might expect, both curves converge at the rated 130 C maximum continuous winding temperature and this feature proves to be consistent between the two models for 1X continuous power dissipation. Hence, the much simpler twoparameter thermal model provides reasonable accuracy in calculating dynamic winding temperature so long as the power dissipation inside the motor doesn t exceed its 1X maximum continuous value. Although, that s not the way a servomotor typically operates. Instead, servomotors are often commanded to produce a dynamic motion profile that typically contains time periods calling for 2X or even up to 4X peak torque output if allowed by the manufacturer [2, 3, & 4]. Therefore, Figure 3 again compares the winding s dynamic temperature rise for both thermal models only this time we assume 2X peak torque output corresponding to 4X power dissipation in the motor s electrical winding. 4

60 mm Motor, Winding Heat-Up with 4X Power Dissipation Solid Red = Four-Parameter Model Dash Black = Two-Parameter Model Figure 3 As you can see in Figure 3, with 2X peak torque output, 4X power dissipation, the fourparameter model shows the winding temperature rises to its rated 130 C value in 140-seconds while the two-parameter model lags way behind and shows the winding s temperature should only be less than 80 C which is a significant, and unacceptable temperature difference I have verified experimentally on this particular servomotor. Hence, as you can see graphically once the power dissipation becomes greater than the 1X maximum continuous value the two-parameter model becomes very inaccurate and this dynamic temperature difference between the two thermal models gets progressively worse with increasing power dissipation. To verify this last statement, Figure 4 compares the dynamic winding temperature during heat up for both models only this time with the motor producing 4X peak torque corresponding to16x power dissipation in the motor s electrical winding. 5

60 mm Motor, Winding Heat-Up with 16X Power Dissipation Solid Red = Four-Parameter Model Dash Black = Two-Parameter Model Figure 4 As you can clearly see in Figure 4, with 4X peak torque output, 16X power dissipation, the fourparameter model shows the winding temperature reaches its rated 130 C value in only 25- seconds while the two-parameter model lags way behind and shows the winding temperature should only be 62 C which is a huge and unacceptable temperature difference. Again, I have verified this temperature difference between the two models with actual measurement on different size motors and the four-parameter model has much greater accuracy in calculating dynamic winding temperature when more than 1X power dissipation occurs in the motor. Why a Temperature Sensor Won t always protect the Motor from Overheating Before completing our discussion why a temperature sensor won t always protect the motor from overheating we must consider one more heat-up curve. Back in Figure 1 the dynamic heat-up of both the Winding and the Case was calculated using the four-parameter model with 1X power dissipation. Next, in Figure 5 we again see dynamic heat-up of both the Winding and Case only this time with the motor outputting 4X peak torque corresponding to 16X power dissipation. 6

60mm Motor - Winding and Case Heat-Up with 16X Power Dissipation Solid Red = Winding Temperature Solid Blue = Case Temperature Figure 5 As you can see in Figure 5, the four-parameter model calculates that with 4X Peak torque and 16X power dissipation the winding s temperature rises rapidly from it s initial 25 C ambient value to it s 130 C rated value in only 25-seconds. However, during this same time the case temperature hardly changes as Figure 5 shows it only rises to 30 C. In effect, for 16X power dissipation, winding heat-up is adiabatic with little heat power being transferred to the case during the first 25-seconds. If we next look at 70 seconds elapsed time we see the winding is approaching 280 C while the case has barely reached 40 C. In contrast, if one uses the twoparameter model to calculate the entire motor s (including the winding) dynamic temperature rise you find the winding temperature (graph not shown) is supposedly less than its130 C rated value at 70-seconds but again actual measurement shows the winding is actually at or above 280 C and in the process of burning up! As mentioned earlier, one practical issue for every servomotor manufacturer is what type of temperature sensor do you use (i.e., thermocouple, thermistor, temperature switch etc.) and where inside the motor is the best place to locate this sensor [5]? In combination with the Drive you want this temperature sensor to protect the motor from overheating during all possible 7

modes of operation but you also don t want nuisance motor shutdown. Furthermore, since a servomotor can only operate in combination with a Drive one must be aware of what type of temperature sensor the Drive will interface with? Many commercial Drives only allow use of a temperature switch and not a thermistor or a thermocouple. In addition, most of today s Drives use the Pulse Width Modulation (PWM) technique to produce their output voltage and current and PWM Drives are electrically noisy [11]. Based on my experience of trying to measure dynamic winding temperature using a thermocouple it s extremely difficult to obtain accurate temperature values due to all the electrical noise emitted by the Drive [11]. Hence, I typically find many servomotors contain either a temperature switch [5] or a thermistor [2 & 5] mounted inside the motor since most Drives aren t designed to interface with a thermocouple. The final question is where s the best location for a motor temperature sensor? Now that we have the four-parameter thermal model showing us graphically how fast the motor s winding dynamically heats up compared to the case it seems logical the best location for a temperature sensor is attached directly to the motor s electrical winding [6]. Furthermore, in reviewing the advertisements from several different servomotor manufacturers I find many of them proudly announcing their motors are UL and/or CSA recognized under the UL 1004 and/or CSA 22.2 motor standards. As part of the UL/CSA recognition process, the motor s electrical insulation system must be constructed in compliance with the UL 1446 Insulation System Standard [12]. As described in Section 4 and displayed in Table 4.1 of UL 1446, the winding s maximum allowable Hot Spot temperature, occurring at any point and at any time, is determined by the Class of the insulation system used to construct the winding. Hence, to comply with UL 1446 the winding s insulation system must have a maximum Hot Spot temperature that s at least equal to or greater than the maximum continuous winding temperature. In addition, to provide the motor with an over-temperature safety margin during peak torque output, it makes engineering sense to construct the winding using a higher Class insulation system such that the winding s maximum Hot Spot temperature isn t exceeded at any time in violation of UL 1446. Therefore, to make sure the servomotors stays in compliance with UL 1446 the temperature sensor should also be placed at the point in the motor s electrical winding where its maxim Hot Spot temperature occurs. Although it s best to locate the temperature sensor at the point in the winding where the Hot Spot temperature occurs it s not always possible to do this especially in the smaller size 20-90mm diameter servomotors. As mentioned earlier, many servomotor Drives will only interface with a temperature switch such as a Thermik SO1 [10]. Given the physical size of a temperature switch in combination with the packing density of the motor s electrical winding the manufacturer often attaches the switch to the winding s end turns, as shown in reference [4]. However, the end-turn location doesn t always correspond to the winding s Hot Spot location. Furthermore, in many 20mm to 60 mm diameter servomotors the manufacturer locates the temperature switch inside the motor but due to physical size of both the switch and the winding this normally closed switch isn t attached to the winding and the winding s dynamic temperatures are not being measured directly. Besides not attaching the temperature switch directly to the winding and/or not placing it at the winding s Hot Spot location, some servomotor manufacturers specify their motors as having a Class B (130 C) or Class F (155 C) insulation system while they correspondingly specify 130 C 8

or 155 C as the maximum continuous winding temperature. In addition, they also specify 4:1 or even 5:1 as the peak to continuous torque ratio and this doesn t provide any safety margin between the winding s maximum continuous and its maximum Hot Spot temperature. As we have already seen graphically the four-parameter thermal model shows a servomotor s electrical winding heats up much faster than predicted by the still widely used two-parameter model. As also shown, as the motor s peak torque output increases above the 1X continuous value the dynamic temperature difference between the winding and the case becomes increasingly greater and above 2X Peak torque the initial winding heat-up is adiabatic with the rise in case temperature lagging behind. Therefore, with no safety margin between the winding s maximum continuous and its Hot Spot temperature actual measurement shows for greater than 2X Peak torque output it s extremely difficult, if not impossible, for a motor temperature sensor, not attached directly to the winding, to react fast enough such that in combination with the Drive the winding s maximum Hot Spot temperature is exceeded in direct violation of UL 1446. Adding to this problem is a reality that both servomotor manufacturers and motor users still use the oversimplified two-parameter thermal model to make all their Duty Cycle calculations as evidenced by the fact most manufacturers publish only one value for the motor s winding to ambient thermal resistance (R th ) along with its thermal time constant. Hence, motor users are forced to make Duty Cycle calculations using the over simplified two-parameter thermal model unless they measure the motor s four parameter values themselves as shown in reference [9]. Therefore, just because the two-parameter calculation says the maximum Hot Spot temperature can t be exceeded during a specified dynamic motion profile isn t necessarily true and the end result is the servomotor overheats and the Drive-temperature sensor combination won t always prevent this from happening. Furthermore, even if the temperature sensor is attached directly to the winding [4], but not at the winding s Hot Spot location, the measureable temperature gradient, normally occurring within many servomotor windings, can also prevent the temperature sensor from detecting the dynamic rise in Hot Spot temperature fast enough to prevent the maximum allowable value from being exceeded which again is in direct violation of UL 1446. Therefore, just because a servomotor contains a temperature sensor, it can t and won t always protect the motor from overheating or possible even burning up! References [1] (http://www.exlar.com/prod_slm_st_curves.html) [2] (http://kollmorgen.com/website/com/eng/download/document/200608011600510.akm_selection_guide.pdf) [3] (http://www.baldor.com/support/literature_load.asp?litnumber=br1202-e) [4] (http://www.hurst-motors.com/ntdynamo.html) [5] Thermik Corp., (http://www.thermik.de/index.php?id=produkte&l=1) [6] (http://gearmotorblog.wordpress.com/2009/09/21/thermal-overload-protectors/) 9

[7] Underwriters Laboratories, UL 2111 - Overheating Protection for Motors Go to: (http://ulstandardsinfonet.ul.com/scopes/2111.html) [8] Electro-Craft Corp., DC Motors Speed Controls Servo Systems An Engineering Handbook, First Edition October, 1972 [9] R. Welch, Continuous, Dynamic, and Intermittent Thermal Operation in Electric Motors (http://www.smma.org/motor_college_thermal.htm), 52-page Tutorial Book available from (welch022@tc.umn.edu) [10] Thermik Corp., (http://www.thermik.de/index.php?id=s01&l=1) [11] (http://en.wikipedia.org/wiki/pulse-width_modulation) [12] Underwrites Laboratories, UL 1446 Systems of Insulating Materials General (http://ulstandardsinfonet.ul.com/scopes/scopesnew.asp?fn=1446.html) 10