Problem with starting sensorless PMAC for small wind turbine with IHM009V1

[labartheg]

Posted on January 08, 2009 at 07:26

Problem with starting sensorless PMAC for small wind turbine with IHM009V1

[labartheg]

Posted on May 17, 2011 at 10:15

We are using a 1,8 kW Ginlong Permanent Magnet Generator. It is three phased and has 16 poles. The voltage can reach 500 V at 600 RPM.

Since we are designing a Vertical axis wind turbine, we need to start it thanks to the generator. We would like to bring it up to at least 100 RPM. Since our generator has no hall sensors, we are using your sensorless reference design kit for 3 phase PMAC motors (reference design kit GUI - 3PH PMDC/AC or BLDC/AC Motor Sensorless). ControlDB-7FMC2(UM0121-UM0122) and power board semitop 3 (UM0252). I have also read AN1129-AN1130-AN1276. We fed the three phases bridge at the same time as 300VDC(115V rectified). We obtain: We observed the phase alignement, noise due to accelerating voltage (2rpm/s), but the generator does not turn and stays in the same phase alignment and we feel the torque. I have tried to select RV2 and RV3, and when i turn RV3 potentiometer, the motor is turning right a little bit when i turn RV3 in one direction and left in other direction. If i let RV3 pot fixed the motor doesn't turn at all. So No BackEMF can be read. it seems that the timing for the starting phase is to short at the beginning. Do you have any idea, why it doesn't turn at all. Thanks Guillaume

[muratterzi]

Posted on May 17, 2011 at 10:15

i think your purpose is to work your generator as a motor to get sufficient rpm's at start phase. Though your generators power rate is 1.8kw i investigated this typr of generator from

http://www.ginlong.com/wind-turbine-pmg-pma-permanent-magnet-generator-alternator.htm

but there is no sufficient data for electrical parameters of motor. for example you need phase to phase dc resistance and inductance paramaters to obtain actual needed power to drive this generator as motor. in case you measure these parameters i think it is needed to modify power stage and slow down pwm frequency that will suits your generator response time. Because generators response time and frequency band is very slow this is needed due to there is no a control ability purpose. they need only to produce electrical power. but in nature they can be driven as brushless dc motors. There are some examples to drive car alternators as brushless dc motor using some mosfets and simple visual basic lpt port controller on web (youtube):) http://www.youtube.com/watch?v=Cqf4ICwFBIU but they use very low pwm frequency as you can easilysee led blink . sincerely

[labartheg]

Posted on May 17, 2011 at 10:15

Thank a lot for your rapid answer, which encourages me a little bit.

The phase resistance is 5.5 Ohms per phase. I agree that the electrical spec are very poor. I guess Inductance could be also usefull, but as the frequency is changing a lot it is difficult to have a such value for large range of frequency. Nevertheless i will do some measurement to have it at 60Hz. This can help for the rotating phase and backEMF control. To reduce the frequency of the PWM, i am not able to do that with RDK GUI, but i think i can change it into the software directly? I also wonder if i can use another interface which is AK-ST7FMC Motor control starter Kit V2.0 instead of ST7 Toolset V3.12? What's the difference? Is it usefull to use MP8011A user interface? I have also downloaded :BLDC Sensorless Motor Stand Alone Library rev 2.1 but i don't know how to use it. It's an executable already installer, but in the Softec software in don't know how to used it? Thank you for your help Regards Guillaume

[muratterzi]

Posted on May 17, 2011 at 10:15

i still offer you to use ST7 Toolset V3.12. But i have not sufficient motor control experience with st7 series.Though there is no sufficient bldc library iam working on bldc posiiton control using stm32 series. i dont offer you to use MP8011A user interface that was only how that can be achived. So st7 or stm32 series are more powerfull with respect tou adaptability and control features. But it must be easy to decrese pwm freq values using software or manipulating header files before flashing mcu. incase stm32 series when using focgui utility header files can be adapted easily for your project like below. i send you some simple motor equations to understand your system. But i think your power stage is not sufficient to drive 1.8kw generator as motor and pwm freq ishigh than your motor response time. if you use formula to find how a current source you need .

if we eliminate Ke parameter(we dont know) assuming that Vs=300 v dc Vs=I.R+Ke.W 300=I.5+0 I=60 A (max) this shows your power source and drive circuitry must have sufficient current to drive motor. And that power stage you use may nat be sufficient for this purpose. we need more parameters to estimate what occurs. sincerely

[labartheg]

Posted on May 17, 2011 at 10:15

Thank for you advise on FOCgui which seem very interesting, but i'm not sure i can use it with my ST7(because of 8bit micro instead of 32bit).

Perhaps i can check for buying such platform i there no other solution. I need to first determine from where the starting problem comes. As the PMG is triphase, for 300V and 5.5Ohms per phase, it give with the same formula: I=300/(2*5.5)=27 Amp, because always 2 phases are energized. Actually i was thinking that the maximum current was the average over a PWM period. So if i have 10% duty cycle of PWM. i should give me an average current of 27/10=2.7Amp. I need to verify that point and be sure to not override the maximum PMG current rating in order to avoid deenergization of the magnets. Also i need to verify if 10% duty cycle of 27Amp is not dommageable for the PMG. I will buy an autotransformer to be able to adjust the AC Voltage and so DC voltage. What i wonder is if the Power board as current limiter, which can block the starting process, because of 27Amp(but only 10%)? Although i can hear the switching of the accelerating phase of 3phases inverted voltage. Thanks Guillaume

[muratterzi]

Posted on May 17, 2011 at 10:15

According tou your correct calculation of current as max 27 A , in case %10 duty cycle beleiving that mean average current is 2.7a is a mistake.(this calculation does not guarantee the max current drain by igbt).we must look igbt power stage out. circuit schematic is shown below for single phase. used igbt is STGF7NC60HD and datasheet information says max continuous current is only 10 A at 25 degree . though pwm looks like easy to drive your motor as decresing current limits. i think you must get a drive circuitry at least cabable to source 27 A.

sincerely...