标签: repair

Electronic engineers may love to play with small mechanical 'toys'. DC Servo Motor is one of these. Most of the time, if we do not overload the servo, the lifespan of the motor can be very long, in spite of the brushed design. However, some DC servos are out of order not because of the motor is burnt (or the brushes are ruined), but due to some components on the PCB. I have bought a brand-new DC analog servo from somewhere, it costs HKD $90. I write some codes on my STM32 platform to test it. The program is simple, which makes it to run back and forth in small steps. Fig.1 Newly bought RC servo After testing it (not continuously) for some time which was less than an hour, the servo motor suddenly became out of order... As usual, what wass going to be next would be a screwdriver with the servo motor! Haha   Fig.2 Ready to disassemble the motor with a screwdriver   Then the casing was opened, there were a control PCB assembly and the motor.   Fig.3 The casing is opened   Then the motor was powered on and PWM signals were given to the servo motor. I traced the signals from the motor. The signals inputted to the two Dual NMOS and PMOS ic, which formed a H-bridge, did not show any waveform, given the message that the later part of the circuit did not receive the control signal.   After that, I started tracing from beginning... I found that there are one of the two small SOT-23 transistors had some problems. The base terminal (pin 1) of the PNP transistor was LOW (turned on) and pin 3 (emitter) was 5V (supply voltage), but pin 2 (collector) was 0V... So the transistor was not functioning!   A soldering iron and utilities were prepared to replace the transistor   Fig.4 Prepare a solder iron to replace the transistor   I replaced the original transistor with a SS8550, keeping the other components unchanged. If it did not work, I would have tested other components.   Fig.5 Before and after transistor replacement   To my surprise, the servo motor worked normally again!! Yeah!  

Sometimes it is good to keep our development tools and utilities up-to-date, since updates may have the bugs fixed and have more functions. I just replaced my MDK4.6 with MDK4.7a (verson Mar 2013). Fig.1 MDK updated to 4.7a   Surprisingly, I was unable to download the firmware (to a STM32 MCU) with my ULINK2 anymore... The IDE went out of response during program download... Fig.2 A ULINK2 to download program   Fig.3 The IDE went out of response... download program failed On the other hand, the 'COM' LED keeped on... No further response for a long time... Fig.4 The COM LED seemed to be on forever To solve this problem, many people on the Internet could think of downloading older firmware versions to the ULINK2 through the UART port by opening the casing. Fig.5 Pins soldered for UART firmware download   I did not download the old firmware successfully, the LPC2000 flash utility was not working stably after trying for several times, using CP2102 as the USB to TTL interface. Probably the CP2102 had baud rate timing problems... I don't know...   At the end, I found that increasing the 'Max Clock' in the configuration would totally solve the problem!! The default "Max Clock" was 1MHz, if we plug the ULINK2 into the USB, then we can choose "10MHz" as the clock.   Fig.6 Choosing "Max Clock" to be 10MHz, after plugging in the ULINK2   Everything was working fine since then... however, old projects still have to be re-configured before downloading them again to MCUs...