need help: speed control of 3 phase induction motor using triac

[malik_zohaib]

hi guys i need help to control the speed of 3 phase induction motor. i tried v/f method but failed, now i am trying to achieve this by using the triac. the problem is i can't able to trigger the triac with moc3021. triac scheme is simple as shown in diagram 1 of the below link.

http://www.datasheetcatalog.org/datasheet2/0/00epgehqsl8fx7e9s69413k4wrwy.pdf

i need your help.

thanks in advance.

waiting for your replies.
regards.

 

[malik_zohaib]

guys little help please......... i am in deep s**t.

 

[mtwieg]

Do you at least see the output of the MOC3021 trigger? The MOC3062 used in that document is a very different device, which contains a zero crossing detector. That may be the root of your issue.

 

[malik_zohaib]

moc3021 also have zero crossing detector. i just able to on/off the test device (bulb) but not able to control the phase.

 

[mtwieg]

moc3021 also have zero crossing detector.

The datasheet says nothing about a zero cross detector.

i just able to on/off the test device (bulb) but not able to control the phase.

Are you using the MOC3021 to directly drive a load in this test, or are you using it with a secondary triac?

At this point you should probably show a detailed schematic, for both test setups.

 

[malik_zohaib]

oh sorry basically it is moc3031 not 3021 , my bad. using same schematic as on page 7 of moc3031 datasheet. also me using triac bt136.

 

[kak111]

Due to working principle of 3-phase induction motor there is no practical way
to adjust the rotation speed by changing the supply voltage.
Only useful way is to use a variable frequency power supply.

Induction motor / Speed control

The theoretical unloaded speed (with slip approaching zero) of the induction motor is controlled by the number
of pole pairs and the frequency of the supply voltage.

When driven from a fixed line frequency, loading the motor reduces the rotation speed.
When used in this way, induction motors are usually run so that in operation the shaft rotation speed is kept above
the peak torque point; then the motor will tend to run at reasonably constant speed. Below this point, the speed tends
to be unstable and the motor may stall or run at reduced shaft speed, depending on the nature of the mechanical load.

Before the development of semiconductor power electronics, it was difficult to vary the frequency, and induction motors
were mainly used in fixed speed applications. However, many older DC motors have now been replaced with induction
motors and accompanying inverters in industrial applications.

Induction motor - Wikipedia, the free encyclopedia

Motor Control 3-Phase AC Induction Motor Solutions By Freescale

 

[malik_zohaib]

i tried v/f method but failed. i think the problem was in ibgt driver that i designed using ir2136 but i cant able to locate the error.
for that i used pic18f452 and my inverter design was form the microchip notes an889.
i wanna know can i directly give my pwms to igbts?? or first i pass them to buffer and pull them up to 12-15v and then supply to igbt inverter??? please answer this. thanks.

 

[mtwieg]

oh sorry basically it is moc3031 not 3021 , my bad. using same schematic as on page 7 of moc3031 datasheet.

You mean this datasheet?: https://www.fairchildsemi.com/ds/MO/MOC3042-M.pdf
There are several datasheets...

Are you certain you're triggering the device right after the zero crossing? As I understand it the zero crossing locks you out after the terminal voltage reaches a certain point. In fact, doesn't that make it useless for speed control, unless you actually want to control it by completely skipping cycles?

Have you tried continuously driving the input? And that the triacs are not being destroyed by excessive currents in the gate or terminals?

edit: and yes kak111 is correct, getting decent speed control from AC mains with an induction motor isn't feasible, due to their basic operating principle. I thought you were doing a variable frequency inverter from a DC bus...

---------- Post added at 14:59 ---------- Previous post was at 14:50 ----------

i tried v/f method but failed. i think the problem was in ibgt driver that i designed using ir2136 but i cant able to locate the error.
for that i used pic18f452 and my inverter design was form the microchip notes an889.
i wanna know can i directly give my pwms to igbts?? or first i pass them to buffer and pull them up to 12-15v and then supply to igbt inverter??? please answer this. thanks.

For all the IGBTs you will need a driver to supply more current than microcontroller pins can do alone. For the high side IGBTs you will need some kind of high side driver circuit, like the ir21xx series stuff. Or a fully isolated driver with its own isolated power supply.

 

[malik_zohaib]

that is the problem. i cant able to design the appropriate driver. that is why me asking for direct pwm supply to igbt or through some buffer.

 

[mtwieg]

that is the problem. i cant able to design the appropriate driver. that is why me asking for direct pwm supply to igbt or through some buffer.

The ir2136, along with all the devices like it, is that buffer. And it's generally the simplest possible solution (though all the bootstrapping high side drivers do seem to give many people a lot of headaches. A quick search of these forums will reveal tons of threads on the topic). The only simpler solution I can think of is if you use P-channel high side switches and use some discrete level-shifting circuit to drive them. But it will give you poorer performance and will overall probably be more difficult than boostrapping drivers.

 

[malik_zohaib]

then can you suggest me how to construct the working driver circuit??

 

[mtwieg]

Which driving circuit..?

First, give some specs on things. What is the DC bus voltage, the load current, the motor speed, and the switching frequency you want? Depending on that some driver methods may not be practical. But I'm telling you that using high side drivers will almost certainly be the easiest solution, if you can find out why you couldn't get them working before.

 

[malik_zohaib]

about 220-230v dc; motor current is 2 amp; speed is about 2900rpm and motor operating frequency is 50hz and pwm frequency is 20khz. and also is my dead time generator circuit is ok??? and i am using the same design for igbt inverter as given in microchip note an889.

 

[FvM]

and also is my dead time generator circuit is ok???

AN889 uses drivers with built-in deadtime. Your circuit is delaying the wrong pulse edge and generating negative deadtime, the inverting action of the CMOS gate must be considered.

 

[Sergio16463]

Regarding drivers for the IGBTs, a good place to start is the IR2110 from International Rectifier: is is quite simple to use, and the datasheet schematics are quite clear. As for the pwm command, instead of the PIC 18f4552, a more suitable would be the dsPIC30F2011, as the latter has dead-time timing and complementary control, all via software. All you have to do is choose which method is more suitable to control the motor (full wave, 2 level SPWM, 3 level SPWM, SVM, etc).

Another idea would to use an isolated driver, like the A3120 for the IGBTs on the upper side, and connect via a resistor/zener to the pic (ideally, through an opto, like the 4N26 or the 9030).

You can still use a 3 triacs (one for each phase) to control the stator voltage, but all you can control is the torque, not the rotor speed.

 

[FvM]

Two additional comments on triac control of asynchronous AC motors.

- The Onsemi circuit quoted in post #1 is explicitely intended for on-off and rotation control. For the same reason it uses zero-crossing detector opto tricas.

- The principle limitations of asynchronous motor speed control by triacs and phase angle variation hav ebeen mentioned. There are however a few applications where it can work though. It's the case with fans and similar loads that have a strongly speed dependent torque, at least linear or even quadratic.

 

[malik_zohaib]

AN889 uses drivers with built-in deadtime. Your circuit is delaying the wrong pulse edge and generating negative deadtime, the inverting action of the CMOS gate must be considered.

yes but i am using pic18f452 because of its availability. and also my driver ir2136 inputs are out of phase that is why i did this. is this wrong???

---------- Post added at 10:21 ---------- Previous post was at 10:20 ----------

Regarding drivers for the IGBTs, a good place to start is the IR2110 from International Rectifier: is is quite simple to use, and the datasheet schematics are quite clear. As for the pwm command, instead of the PIC 18f4552, a more suitable would be the dsPIC30F2011, as the latter has dead-time timing and complementary control, all via software. All you have to do is choose which method is more suitable to control the motor (full wave, 2 level SPWM, 3 level SPWM, SVM, etc).

Another idea would to use an isolated driver, like the A3120 for the IGBTs on the upper side, and connect via a resistor/zener to the pic (ideally, through an opto, like the 4N26 or the 9030).

You can still use a 3 triacs (one for each phase) to control the stator voltage, but all you can control is the torque, not the rotor speed.

but the problem is i cant able to find ir2110 , the isolator driver and other pic18. so me using 18f452 and ir2136.

---------- Post added at 11:17 ---------- Previous post was at 10:21 ----------

hi guys i just able to manage the ir 2110. now i want to ask two things.!\
1) i want to test it for low power mosfet like irf540n. should i use the same scheme like in the datasheet of ir2110??? and what should be the value of resistors and bootstrap capacitor???

2)will it work on 50hz (50% DC)??? or i have to generate 20khz and then update it with 50hz rate????

please clear me, i am very confuse.
thanks.

 

[FvM]

The deatime generation polarity is correct for IR2136.

 

[mtwieg]

hi guys i just able to manage the ir 2110. now i want to ask two things.!\
1) i want to test it for low power mosfet like irf540n. should i use the same scheme like in the datasheet of ir2110??? and what should be the value of resistors and bootstrap capacitor???

For the bootstrap capacitor you want it big enough that it won't droop during a switching cycle due to either leakage current (in the ir2110 or the capacitor itself) or due to charging the gates. If your PWM frequency is high (like 20KHz) and your FETs/IGBTs aren't huge, then something in the 100nF-1uF is generally enough. For the resistor, I'm not sure what resistor you're talking about.

2)will it work on 50hz (50% DC)??? or i have to generate 20khz and then update it with 50hz rate????

You'll want to keep the PWM frequency high; 20KHz is nice since it shouldn't generate audible noise, though depending on the power you want to deliver you may want to adjust it up or down. For the motor speed that's variable and you should only be limited by the motor itself and your bus voltage, and maybe your control system.