using the bs2sx to control large 3 phase motors

remmi870

Im new to the forums but ive used bs2 for the past year or so, study EE and i want to build a large 3 phase motor controler. I know that the pwm and the bs2 instruction execution is too long to properly use pwm to control 6 individual IGBT's or mosfets that would be needed to create the 3 phases to feed a motor.

I have worked on big cnc machines using inverters and 3 phase motor drivers, the normal block diagram is to bring in the three phase power, rectify it to DC, clean it up with caps, run it into bipolar transistors or isolated gate bipolar transistors to control the frequncy and breaking of the motor with a master control board. Some of the boards alow serial port control to troubleshoot.

I wanted to know if anyone knows of a driver chip that i can use that either will create one phase·or hopefuly 3 phase modulation from pwm or even digital control that i can program a bs2 to control and alow me to later connect to with serial port to diagnose or adjust, idealy i want to make a softstart motor starter that has a slow ramp·that can be adjusted with dip switches or alow me to plug into and monitor.

Post Edited (remmi870) : 11/20/2008 8:43:05 AM GMT

Bruce Bates

remmi -

Soft-start motor controllers are readily available from all the major motor control companies. Above and beyond that, it's not clear to me exactly what you're looking to do.

One additional item. There are single phase to 3-phase rotary converters available for taking a single phase input, and creating a 3-phase output. Is that what you were looking to do? If so, I can probably find a couple of resources for you - just ask.

Regards,

Bruce Bates

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
When all else fails, try inserting a new battery.

remmi870

thanks bruce, but i dont think you quite get it, the motor im looking to use this on is around 10 hp, 7.5 kw. a commercial soft start that has limited ramp capabilities cost around 2000-3000 dollars, i can use around 250$ of the major hardware components and use a bs2 to control speed and full ramp.For the same reason that people build thier own cnc machines at a fraction of the cost, one could buy commercial cnc machines for 10k dollars that could be made for a fraction of that.

im looking for a chip that is used to control IGBT's that can accept digital input to control it, most use individual PWM for each phase, unfortunetly the bs2 cant get the speed to control each of the 6 sides of the phase, the program execution is simply too long.

Bruce Bates

remmi -

Why are you limiting yourself to the BS-2? There are a number of MUCH faster PBASIC Stamps. Attached please find a Stamp Comparison document which lists all of the Stamps, and their relative speeds. I've no idea whether any of the other Stamps will be fast enough, that's something that you will have to ascertain.

As to the other question of multi-channel PWM chips, I've never used one, so I really don't know whether they exist or not. I would THINK that they do. Sorry I have no resource for you on that score.

Regards,

Bruce Bates


▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
When all else fails, try inserting a new battery.

$WMc%

remmi870

Have You thought of using more then one $stamp for Your App.....What I'm geeting at is the $stamp can only do one thing at a time...,But 6 $stamps and 1 controller $stamp can do A lot.,In the same time frame!!!!

It's just a tought,But take a look at the "OEM BS2"!!!!

I have done what Your lookin to do....It's the wave forms that worry Me the most...This is the hardest part.. If the output wave is'nt rite,The motor will fail

______________________________$WMc%_________

Carl Hayes

Your real design problem is the motor controller itself.· At that power level you're probably going to use reverse-parallel SCRs and a pulse-generating circuit of some kind to trigger them twice each cycle.· When, each cycle, is what your circuit must decide, for that's how you throttle the motors.· Basically you must design a really big 3Φ lamp dimmer.

A Stamp could conceivably do that (provide the timed pulses).· For smaller motors you'd use triacs, but reverse-parallel SCRs will turn off automatically·with perfect reliability each half-cycle, while turn-off is a problem sometimes with triacs.

Incidentally, I'd love to have a source for 10hp 7.5 KW motors, which would sell well, since one horsepower is exactly 745 watts.· That's almost 98% efficiency in the motor.· And here I thought 90% was really, really good.·

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
· -- Carl, nn5i

Post Edited (Carl Hayes) : 11/21/2008 3:39:44 AM GMT

remmi870

I work for a wood working company, my father owns it, i do all of our machine maintence and i just installed a belt driven regennative vaccum pump, the belts keep breaking so the first step would be to create a soft start motor starter, we have 3 phase running all over and one of our machines had a 7.5 kw steper motor driver in it that i had to replace, after taking it apart and seeing what broke (the voltage regulator that controls the regulator blew up the caps) i got to study this design.

I should have specified that im using a bs2sx chip that i designed from oem components, basicaly i made a module on a breakout board. However the PWM meathod that i was planing to use would be to create the sine wave, however i dont think a basic stamp could create the proper wave via pwm, even using a sx chip, the only real viable solution would be to use a propeller or maby sx technology chips.

I was just curious if anyone has done anything like this and may know of a chip that is programed or designed to run a IGBT in a sine wave patter using pwm.

Jesse Hasty

You can vary the pwm in a sinusoidal manner. You should be able to achieve slow speed with this.

remmi870

the problem is going into a high speed, 60 hz, a full wave is 16.6 ms long, so a half wave is 8 ms, but to control a full sinusoidal waveform the basic stamp couldnt realy get precise enough to control a full wave in ms, it would need to be in us or microseconds delay periods between the pulses to achive 60 hz total.
this is to control a IGBT to create a sine waveform, one could use pwm to create a square wave but that would be dirty and cause alot of heat in the motor.
but the problem is if there is a hold in the program like a do untill statement it may put a delay in the program that would hinder the total speed of the diodes, the other problem is that there needs to be a continuous loop and any delay would cause a missfire that would throw the phases out of sync for instance (P1PWM-P2PWM-P3PWM-P1LPWM-P2LPWM-P3LPWM-program delay-P1PWM) this would be catastrophic on a motor.

stamptrol

Unless your time is worth nothing, this will be a huge project to end up with only an average result.

One way that would give you a reasonable product is to buy a small 3 phase drive ( say 1 or 2 HP). Then buy the IGBT's in a size to control your 10 HP motor. Replace the small drives's transistor package with the bigger transistors.Use the Stamp to control the drive front end (usually a 0 - 10 volt analog signal.

But, on the other hand you can get a new 10HP (575 volt) drive for $762, or a 10HP (460 volt) for $679. Sounds like not too much to keep your business going. See http://web4.automationdirect.com/adc/Shopping/Catalog/AC_Drives/GS2_(115_-z-_230_-z-_460_-z-_575_VAC_V-z-Hz_Control)/GS2_Drive_Units_(115_-z-_230_-z-_460_-z-_575_VAC).

Cheers,

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Tom Sisk

http://www.siskconsult.com
·

Carl Hayes

If I were doing it I would likely go low-tech -- a three-phase Variac with a great big knob on the front.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
· -- Carl, nn5i

remmi870

thanks stamptrol, ive used automation driect before, they have decent prices, but even these drives state that they dont have a good sine wave and they need a line reactor to smooth out the drive signal.

the other benifit to using my meathod is if i were to use a 3 phase motor for a cnc axis drive using a encoder as feed back and letting the bs2 control the IGBT bridges. these small motor drives only go to 10hp, depending on the type of motor. if i build my own driver, i could increase the drivers, however if i were to modify a comercialy available driver, i would have no way to control the curent since the smaller modules are meant to run smaller motors.

one of the other problems is if the manufacture uses a proprietary driver i may have problems building a circuit fast enough to control larger drivers, the IGBT modules i plan on using need two isolated dc power sources per module that are individualy fired and opticaly isolated, because of this it is realy hard to interface a diferent control.

i was able to use pulsout to create 6 pulses around the legnth that would be needed to create a squarewave enough to give me a 3 phase waveform, however with the square waveform and the cycle of my program i get a .6ms delay at the end of the waveform that may cause a problem.

stamptrol

I wouldn't get too worried whether the drive output is pure sine wave (almost none are). The motor is perfectly able to smooth things sufficiently to run safely. Most times, the line reactor is only needed on long cable runs. And, its chief use is to help control voltage spikes which can puncture motor insulation. Most inverter duty motors have better wire insulation than in the old days and if you can use a 460 volt motor (instead of 600v) you'll have that much better margin.

Anyway, good luck to you. Keep us informed of your progress!

Cheers,

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Tom Sisk

http://www.siskconsult.com
·

remmi870

the problem i got now is most of the IGBT modules need seperate isolated power supplies for each gate, im still workin on that, but if i can figure that much out, i could program the bs to run the waveform but thats the key is to get the driver pack set up. i have a attached a datasheet of the type of IGBT module that i plan on using for my project, anyone got any cost effective solutions to build a isolated power source for each gate?