Controlling a Single Phase Motor (Refrigerator Compressor) using PWM

[flio098]

Hi, were doing a project which aims to control a Ref Compressor using PWM. From our knowledge the motor is single phase. the PWM comes from our PIC where its input is obtained from our own made thermostat and we aim to control it by varying the frequency of the voltage that is entering the Motor. After the PIC we plan to connect its output to an INVERTER.

From our knowledge the inverter outputs a square wave.

Question.

1. Is the square wave good enough to be our input to the motor?
2. is an INVERTER the best way to go since we want to revert back to 220V after the our PIC output?

any useful suggestion is appreciated Thank You

 

[crutschow]

If your aim is to improve the efficiency of the compressor than a square-wave is not desirable. The harmonics in a square wave will cause the generation of extra heat in the motor and reduce the efficiency.

Why would you use PWM to generate a square wave from an inverter? If you want a square-wave you don't need PWM.

For best efficiency you would modulate the PWM output to generate a sine wave of the desired frequency and amplify that to the motor. The motor's inductance will filter out the high frequencies of the PWM signal and leave the sine wave.

But beware, if you lower the speed of the motor below it's rated input frequency you must also lower the voltage to avoid saturating the motor and causing high currents and motor over-heating.

 

[MikeMl]

AC Induction motors are designed to produce peak efficiency when run on a specific voltage and specific waveform (sine wave). Anything you do to modify the waveshape or the applied RMS voltage will only REDUCE EFFICIENCY!!! Your only hope is to scrap the AC induction motor and replace it with a DC motor.

You will have have to convince me, and everyone else that by the time you convert the line AC to DC, then to PWM, to drive a DC motor, that the overall efficiency could be higher than the original efficiency of the AC motor?????

 

[tcmtech]

Slowing down the speed of the compressor motor is the last thing you want to do if you are trying to gain efficiency with a phase change based heat pump system.

You can build it and it will work but if you run a stock refrigerator against your modified one with identical internal masses to keep cool you will find your design will use more power in the end not less.

 

[wkrug]

Is there really a Motor in your refrigerator?
Most of them has an self interruptor inside like an old door bell.
So the membrane will go forward and back by an Electro magnet steered by an switch to compress the cooling fluid.

In series to that "motor" is an Overload Switch and an interruptor ( PTC ) for starting the process.

So an PWM would not work, also like frequency shifting.

Check this out, before start developing.

Compressors for cooling rooms has a normal motor. That needs the full voltage for startig up. The RPM can be steered by controlling the phase frequency. But i guess that will not work in a wide range, because the phase shifting in such a motor is done by an capacitator.

 

[flio098]

Hi crutschow, How would i amplify the output of the PWM? Is there any other way of amplifying that signal besides a step up transformer?

Thank You

 

[flio098]

Thanks for the comments guys, i would like to add something on what we plan to do. We want to use the PWM as the input in the Gates if IGBT's/Power Mosfets of the Inverter that we intend to make. As a result, the PWM will be acting as the control of the inverter where constant switching of the DC input to the inverter will happen.

If i got the concept of the inverter wrong please correct me as you wish, comments and suggestions are appreciated! Thank you

 

[crutschow]

Hi crutschow, How would i amplify the output of the PWM? Is there any other way of amplifying that signal besides a step up transformer?

You could use high plus and minus DC voltages (equal to desired peak AC voltage) powering high voltage MOSFETs to directly drive the motor. Otherwise you need a transformer.

 

[flio098]

Hm.. im sorry i can't understand, what do you mean by "High Plus and Minus DC Voltage" there are 2 DC inputs in the inverter?

 

[crutschow]

Hm.. im sorry i can't understand, what do you mean by "High Plus and Minus DC Voltage" there are 2 DC inputs in the inverter?

The PWM amplified output is the inverter signal.

The plus and minus voltages would be used to power the MOSFETs that amplify (switch) the PWM signal. You need one MOSFET stage connected to generate the plus peak output and another MOSFET stage connected to generate the minus peak output to provide the plus and minus peaks of the sine-wave to the motor. For example, a 220V AC motor signal would require a positive PWM peak of 308V and a negative PWM peak of -308V.

This is obviously a very high voltage to switch so switching a lower voltage and running that through a transformer may be preferable.

 

[Mr RB]

Is there really a Motor in your refrigerator?
Most of them has an self interruptor inside like an old door bell.
So the membrane will go forward and back by an Electro magnet steered by an switch to compress the cooling fluid.

In series to that "motor" is an Overload Switch and an interruptor ( PTC ) for starting the process.

So an PWM would not work, also like frequency shifting.

Check this out, before start developing.

Compressors for cooling rooms has a normal motor. That needs the full voltage for startig up. The RPM can be steered by controlling the phase frequency. But i guess that will not work in a wide range, because the phase shifting in such a motor is done by an capacitator.

I can't say for other countries but in Australia the compressors generally use a rotary motor. Well older fridges do, I haven't opened up a compressor from a brand new fridge.

However I agree that trying to slow the motor is a bad idea! Generally these motors have a very low torque and low momentum so they barely have the power to operate the piston against the pulse load even when running at full voltage and full speed.

Many fridges are destroyed in brownouts when the voltage drops to 70% of the mains voltage etc for a few seconds and the motor stalls, then when mains voltage is increased again it doesn't have the torque to restart the motor and the motor burns out. So you shoudl always turn the fridge off at the wall when you get a brownout.

 

[Diver300]

Many fridges are destroyed in brownouts when the voltage drops to 70% of the mains voltage etc for a few seconds and the motor stalls, then when mains voltage is increased again it doesn't have the torque to restart the motor and the motor burns out. So you shoudl always turn the fridge off at the wall when you get a brownout.

In my experience, if you get a brownout, or a short supply interruption, the motor stalls, as you said. However, the thermal overloads detect the heat and trip the motor out before it does too much damage. By the time the motor has cooled off, the pressures have equalised and the motor restarts OK.

It is certainly kinder to turn the fridge off for 10 minutes or so if there has been a brown out or a short interruption.

The modern microprocessor controlled fridges have timers to stop the overload.

 

[flio098]

The PWM amplified output is the inverter signal.

The plus and minus voltages would be used to power the MOSFETs that amplify (switch) the PWM signal. You need one MOSFET stage connected to generate the plus peak output and another MOSFET stage connected to generate the minus peak output to provide the plus and minus peaks of the sine-wave to the motor. For example, a 220V AC motor signal would require a positive PWM peak of 308V and a negative PWM peak of -308V.

This is obviously a very high voltage to switch so switching a lower voltage and running that through a transformer may be preferable.

Hi, we were able to generate the positive and negative square waves by using IGBT in our simulation. We had a 220V DC input and PWM in the Gates of the IGBT's. What happened was that the output peak of the square wave only ranged from 5V to 10V even though the DC that we were trying to switch was 220V. What could be the reason for this?

Thank you very much

 

[debe]

Refrigeration compressors that are designed for variable speed use 3 phasemotors. Danfoss 12/24V DC compressors are 3phase low voltage for portable fridge freezers. Airconditioners such as inverter types use 3phase motors. Here in Australia the 240V AC single phase is changed to DC 350V, then electronicly made into 3phase & the freq is altered to change the comp speed. These motors are all especialy designed for variable speed. Daryl

 

[crutschow]

Hi, we were able to generate the positive and negative square waves by using IGBT in our simulation. We had a 220V DC input and PWM in the Gates of the IGBT's. What happened was that the output peak of the square wave only ranged from 5V to 10V even though the DC that we were trying to switch was 220V. What could be the reason for this?

Well it could be the phase of the moon, but without a schematic we can't help you much.

 

[tcmtech]

Did you at least use high and low side driver IC's or similar devices to run the switching devices?

 

[flio098]

Thanks for your answer crutschow.

This might be off topic but do you know if its possible to amplify a "Square Wave" signal about 5Vpp to lets say 20Vpp?

 

[crutschow]

Thanks for your answer crutschow.

This might be off topic but do you know if its possible to amplify a "Square Wave" signal about 5Vpp to lets say 20Vpp?

Certainly you can. But how you do it depends upon the amount of output current you need.

 

[flio098]

well lets say about 1A to 2A of current? what circuit will i use? my plan is PWM -> "amplifier of the PWM signal"

Thanks

 

[crutschow]

well lets say about 1A to 2A of current? what circuit will i use? my plan is PWM -> "amplifier of the PWM signal"

If you are doing PWM then why do you need the variable amplitude square wave? PWM is a fixed amplitude square-wave with varying duty-cycle.