37 0

상용 스텝모터 가속방법에 대한 연구

Title
상용 스텝모터 가속방법에 대한 연구
Other Titles
A study on comparative analysis of commercial step motor speed control methods
Author
조세형
Advisor(s)
김태환
Issue Date
2013-02
Publisher
한양대학교
Degree
Master
Abstract
여섯 개의 상용 스텝모터를 사용하여 스텝모터 속도제어 실험을 수행하였다. 각 모터의 속도를 일정 초기 속도에서 타겟 속도인 150 mm/s 까지 시간에 비례하도록 가속시켰다. 모터가 사용될 프린터에 탑재된 써멀 프린트헤드의 최대 속도가 약 152 mm/s 이기 때문에 모터의 최대속도를 150 mm/s 로 정했다. 이 실험의 목적은 몇 가지 조건하에서 최대속도에 빠르게 도달하는 모터를 선정하는 것이다. 모터에 공급되는 파워와, 권선 전류 감소시간 (파워 차단시간) 을 가변 시키면서 실험을 진행하였다. 파워는 모터 드라이버 칩 제조사가 권장하는 전류량 (0.6 mA) 과 그 보다 17% 낮은 0.5 mA 를 모터에 공급하여 실험을 수행하였다. 여섯 개의 모터 중에 어떤 모터가 최대속도에 낮은 파워로 도달할 수 있는지 알아보기 위한 실험이었다. 낮은 파워로 프린터를 작동시킬 수 있다면 가격이 낮은 전원 공급장치를 사용할 수 있기 때문에 이 같은 실험을 하게 되었다. 권선 전류 감소시간은 모터 드라이버 칩에 내장되어 있는 파워 트랜지스터를 차단하는데 걸리는 시간을 뜻한다. 권선 전류 감소시간은 드라이버 칩에 연결되어 있는 RC회로의 R값과 C값에 의해 정해진다. 제조사는 50 us 를 권장하지만, 52 us 를 사용해 보았다. 이는 52 us 에 상응하는 R과 C가 더 널리 사용 되어 지고 따라서 더 많이 제조되는 소자들이기 때문이다. 프린터 제조사의 관점에서는 자제수급이 더 원활한 소자를 사용하는 것이 여러 가지 면에서 유리하기 때문에 이 같은 조건을 선정하게 되었다. 실험결과 여섯 개 모터 중 세 개의 모터가 권선 전류 감소시간이 50 us (제조사 권장 사양)이고 낮은 전류를 공급했을 때 최대 속도에 도달하였다. 여섯 개 중 어떤 모터도 전류가 낮고 권선 전류 감소시간이 52 us 일 경우에 최대 속도에 도달하지 못 했다. 최대 속도에 도달한 세 개의 모터 중 가장 높은 속도에 도달한 모터를 선정하여 리니어 가속 방법과 지수함수 가속 방법을 분석하였다. 리니어 가속 방법은 모터의 속도를 미리 정해진 초기속도에서 시간에 비례해서 일정하게 속도를 최대속도까지 높이는 가속 방법이다. 하지만 이 방법은 모터가 처음 회전하기 시작할 때와 최대속도에 도달하였을 때 가속도의 불연속성 (discontinuity)을 유발한다. 이러한 불연속성은 모터에게 심한 부하를 안겨준다. 이러한 이유로 모터가 속도를 높일 때에는 직선보다는 지수함수에 가까운 형대로 가속하다가 최대속도 근방에서 점진적으로 안정화 될 것으로 추정된다. 지수함수 형태의 가속 방법이 프린터 펌웨어에 구현되었고 그 성능을 통상적인 리니어 방법과 비교하였다. 실제 모터의 속도 프로파일을 측정하기 위하여 가로줄을 주기적으로 감열지에 프린트하고 스캔하여 프린트된 가로줄간의 간격으로 모터속도를 예측하였다. 이렇게 측정된 모터속도의 기울기는 시뮬레이션 값보다 높은 것으로 나타났다. 리니어 가속 방법과 지수함수 가속 방법을 비교한 결과 모터의 초기속도에 따라 두 방법의 성능이 세 구간으로 나뉘어 지는 것으로 나타났다. 첫 번째 구간은 초기속도가 높은 구간이다 (Vo = 100 mm/s). 이 구간에서는 리니어 가속 방법이 지수함수 가속 방법보다 더 효과적인 것으로 나타났다. 리니어 방법이 사용될 경우 모터는 여섯 스텝 만에 최대속도에 도달하였다. 지수함수 방법의 경우 최대속도에 도달하기 위해 여덟 스텝이 필요하였다. 지수함수 방법으로 여섯 스텝 만에 최대속도에 도달하도록 모터를 작동시켰을 때 탈조를 유발하였다. 리니어 방법이 더 효과적으로 나타난 이유를 정확히 알기는 어렵다. 일반적으로 지수함수 방법의 최대 속도 기울기는 리니어 방법의 최대 속도 기울기보다 높다. 상기 설명된 가정에 의하면 리니어 방법의 모터속도 프로파일도 실제로는 지수함수에 가까운 것으로 추정된다. 따라서 지수함수 방법은 최대속도 기울기에 도달하였을 때 리니어 방법보다 더 많은 부하를 줄 것이고, 그 때문에 지수함수 방법으로 여섯 스텝 만에 최대속도에 도달하도록 모터를 작동시켰을 때 탈조를 유발되었다고 추정된다. 두 번째 구간은 모터의 초기속도가 아주 낮은 구간이다 (Vo = 10 mm/s). 이 구간에서는 지수함수 가속 방법이 더 효과적인 것으로 나타났다. 지수함수 방법으로 여덟 스텝 만에 모터가 최대속도에 도달했지만, 리니어 방법은 열 스텝을 필요로 하였다. 다시 말해, 지수함수 방법으로 더 빨리 최대속도에 도달할 수 있었다. 일반적으로 모터가 처음 회전할 때의 초기 속도 기울기는 지수함수 방법이 리니어 방법보다 더 낮다. 이러한 차이가 지수함수 방법이 더 효과적으로 나타난 이유라고 추정된다. 왜냐하면 아주 낮은 초기속도에서 짧은 시간 안에 최대속도까지 가속할 때, 높은 초기속도 기울기는 모터에게 더 많은 부하를 주기 때문이다. 따라서 초기속도가 일정 속도 미만일 경우 초기속도 기울기가 최대 속도 기울기보다 더 큰 영향을 주는 것으로 나타났다. 세 번째 구간은 상기 두 구간의 중간 구간이다 (20 mm/s 이상 <= Vo <= 90 mm/s). 이 구간에서는 리니어와 지수함수 방법 모두 여덟 스텝 만에 최대속도에 도달하였고 두 방법 모두 여섯 스텝 만에 최대속도에 도달하지 못했다. 성능 면에서 두 방법의 차이가 없다는 점으로 미루어 볼 때 이 구간에서는 낮은 초기속도 기울기와 높은 최대속도 기울기 (i.e., 지수함수 가속 방법의 특징) 가 복합적으로 작용하여 서로 상호 보완하였고, 리니어 가속 방법의 낮은 최대속도 기울기와 높은 초기속도 기울기도 복합적으로 작용하였다고 추정된다. 이러한 이유로 이 구간에서는 두 방법의 성능 차이가 없게 나타난 것으로 보인다 |Step motor speed control experiments are conducted using 6 commercially available step motors. In these experiments, the speed of each motor was raised from an initial speed to a target speed of 150 mm/s at a constant rate (i.e., the motor speed was kept linearly proportional to elapsed time while accelerating). The target speed of 150 mm/s was selected because approximately 152 mm/s is the maximum speed the thermal printhead installed in the printer can support. The purpose of these experiments was to determine which of the 6 motors can reach the target speed in a reasonable amount of time under different operating conditions chosen for practical reasons. Two different power levels and two different shut-off times were used to test the motors. The two different power levels used include the power level recommended by the motor driver chip manufacturer (24V x 0.6 mA) and a lower power level (24 V x 0.5 mA). The recommended electric current to drive the motors is 600 mA and the motors were also driven with 17% less current (500 mA) to determine if any of the 6 motors can reach the target speed with less power since lower power required to run the printer requires a less expensive power supply. The shut-off time specifies how long it takes to switch off the power transistors embedded in the motor driver chip. It is determined by a parallel RC network connected to the driver chip. The two different shut-off times used in the experiments are 50 us recommended by the driver chip manufacturer and 52 us resulting from using resistors and capacitors that are more commonly available and more widely used. From the printer manufacturer's point of view, these parts will be much easier to obtain and hence it is more likely to produce printers without running out of parts even if the company's production volume increases drastically. It is found that 3 of the 6 motors can reach the target speed at the lower power level if the shut-off time is set to 50 us. None of the 6 motors reached the target speed at the lower power level with the shut-off time set to 52 us. Out of the 3 motors that reached the target speed, the one that reached the highest speed was selected for analyzing the performance of two motor speed control methods: linear acceleration and exponential acceleration. The linear acceleration raises the motor's speed at a constant rate from a predetermined initial speed to the target speed. But this method introduces discontinuities in acceleration at the points where the motor speed begins to increase and the motor reaches the target speed, imposing a heavy burden on the motor. For this reason, it is reasonable to assume that the motor speed gradually increases in an exponential growth-like manner and gradually settles for the target speed in an exponential decay-like manner. This exponential acceleration method is implemented in the printer firmware and its performance is compared against the performance of the conventional linear acceleration method. To measure the physical motor speed profiles, horizontal lines were printed on thermal paper. The printed sheets were then scanned and the motor speed was estimated based on the distance between each pair of adjacent lines. Motor speed measurements support the assumption that the rate at which the motor speed increases is higher than the theoretical rate. It is found that when the initial speed is set to 100 mm/s, the linear method performed better than the exponential method. The linear method allowed the motor to reach the target speed in 6 steps, while the exponential method allowed the motor reach the target speed in 8 steps. When forced to reach the target speed in 6 steps using the exponential method, the motor stopped rotating in the middle of acceleration. The precise reasons why the linear method performed better are unknown. The maximum acceleration, i.e., the maximum rate at which the motor speed increases with respect to elapsed time, of the exponential method is greater than that of the linear method. Under the aforementioned assumption, the linear method induces the motor to physically accelerate in an exponential manner. Therefore, it is likely that the exponential method imposes an even heavier load on the motor at the maximum acceleration than the linear method. As a result, the exponential method is presumed to fail to raise the motor speed to the target speed in 6 steps. On the contrary, when the initial speed is set to a much lower level of 10 mm/s, the exponential method performed better than the linear method. The exponential method allowed the motor to reach the target speed in 8 steps, while the linear method allowed the motor reach the target speed in 10 steps. In general, the initial acceleration (i.e., the rate at which the motor speed begins to increase from rest) of the exponential method is lower than that of the linear method. This difference in initial acceleration is likely one of the primary reasons why the exponential method performed better because when the motor has to reach the target speed from a very low speed in a short period of time, a higher speed slope at the beginning imposes a heavier burden on the motor. Thus, when the initial speed is below a certain level, the initial acceleration appears to have more dominant effects on printer performance than the maximum acceleration. When the initial speed is set to a level between 10 mm/s and 100 mm/s (20 mm/s <= Vo <= 90 mm/s), both methods allowed the motor to reach the target speed in 8 steps, but both failed to raise the motor speed to the target speed in 6 steps. The fact that there is no difference in performance between them suggests that in this region a low initial acceleration may compensate for a relatively high maximum acceleration (the characteristics of the exponential method) and a low maximum acceleration may compensate for a relatively high initial acceleration (the characteristics of the linear method), which is likely the main reason why both methods produced the same performance test results.
Step motor speed control experiments are conducted using 6 commercially available step motors. In these experiments, the speed of each motor was raised from an initial speed to a target speed of 150 mm/s at a constant rate (i.e., the motor speed was kept linearly proportional to elapsed time while accelerating). The target speed of 150 mm/s was selected because approximately 152 mm/s is the maximum speed the thermal printhead installed in the printer can support. The purpose of these experiments was to determine which of the 6 motors can reach the target speed in a reasonable amount of time under different operating conditions chosen for practical reasons. Two different power levels and two different shut-off times were used to test the motors. The two different power levels used include the power level recommended by the motor driver chip manufacturer (24V x 0.6 mA) and a lower power level (24 V x 0.5 mA). The recommended electric current to drive the motors is 600 mA and the motors were also driven with 17% less current (500 mA) to determine if any of the 6 motors can reach the target speed with less power since lower power required to run the printer requires a less expensive power supply. The shut-off time specifies how long it takes to switch off the power transistors embedded in the motor driver chip. It is determined by a parallel RC network connected to the driver chip. The two different shut-off times used in the experiments are 50 us recommended by the driver chip manufacturer and 52 us resulting from using resistors and capacitors that are more commonly available and more widely used. From the printer manufacturer's point of view, these parts will be much easier to obtain and hence it is more likely to produce printers without running out of parts even if the company's production volume increases drastically. It is found that 3 of the 6 motors can reach the target speed at the lower power level if the shut-off time is set to 50 us. None of the 6 motors reached the target speed at the lower power level with the shut-off time set to 52 us. Out of the 3 motors that reached the target speed, the one that reached the highest speed was selected for analyzing the performance of two motor speed control methods: linear acceleration and exponential acceleration. The linear acceleration raises the motor's speed at a constant rate from a predetermined initial speed to the target speed. But this method introduces discontinuities in acceleration at the points where the motor speed begins to increase and the motor reaches the target speed, imposing a heavy burden on the motor. For this reason, it is reasonable to assume that the motor speed gradually increases in an exponential growth-like manner and gradually settles for the target speed in an exponential decay-like manner. This exponential acceleration method is implemented in the printer firmware and its performance is compared against the performance of the conventional linear acceleration method. To measure the physical motor speed profiles, horizontal lines were printed on thermal paper. The printed sheets were then scanned and the motor speed was estimated based on the distance between each pair of adjacent lines. Motor speed measurements support the assumption that the rate at which the motor speed increases is higher than the theoretical rate. It is found that when the initial speed is set to 100 mm/s, the linear method performed better than the exponential method. The linear method allowed the motor to reach the target speed in 6 steps, while the exponential method allowed the motor reach the target speed in 8 steps. When forced to reach the target speed in 6 steps using the exponential method, the motor stopped rotating in the middle of acceleration. The precise reasons why the linear method performed better are unknown. The maximum acceleration, i.e., the maximum rate at which the motor speed increases with respect to elapsed time, of the exponential method is greater than that of the linear method. Under the aforementioned assumption, the linear method induces the motor to physically accelerate in an exponential manner. Therefore, it is likely that the exponential method imposes an even heavier load on the motor at the maximum acceleration than the linear method. As a result, the exponential method is presumed to fail to raise the motor speed to the target speed in 6 steps. On the contrary, when the initial speed is set to a much lower level of 10 mm/s, the exponential method performed better than the linear method. The exponential method allowed the motor to reach the target speed in 8 steps, while the linear method allowed the motor reach the target speed in 10 steps. In general, the initial acceleration (i.e., the rate at which the motor speed begins to increase from rest) of the exponential method is lower than that of the linear method. This difference in initial acceleration is likely one of the primary reasons why the exponential method performed better because when the motor has to reach the target speed from a very low speed in a short period of time, a higher speed slope at the beginning imposes a heavier burden on the motor. Thus, when the initial speed is below a certain level, the initial acceleration appears to have more dominant effects on printer performance than the maximum acceleration. When the initial speed is set to a level between 10 mm/s and 100 mm/s (20 mm/s <= Vo <= 90 mm/s), both methods allowed the motor to reach the target speed in 8 steps, but both failed to raise the motor speed to the target speed in 6 steps. The fact that there is no difference in performance between them suggests that in this region a low initial acceleration may compensate for a relatively high maximum acceleration (the characteristics of the exponential method) and a low maximum acceleration may compensate for a relatively high initial acceleration (the characteristics of the linear method), which is likely the main reason why both methods produced the same performance test results.
URI
http://dcollection.hanyang.ac.kr/jsp/common/DcLoOrgPer.jsp?sItemId=000000066185https://repository.hanyang.ac.kr/handle/20.500.11754/134452
Appears in Collections:
GRADUATE SCHOOL OF ENGINEERING[S](공학대학원) > ELECTRICAL ENGINEERING AND COMPUTER SCIENCE(전기ㆍ전자ㆍ컴퓨터공학과) > Theses (Master)
Files in This Item:
There are no files associated with this item.
Export
RIS (EndNote)
XLS (Excel)
XML


qrcode

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.

BROWSE