Question about single phase to three phase conversion

[Bararontok]

What type of circuit can convert single phase electric power into three phase electric power? Can the circuit diagram be posted in this thread?

 

[hxtasy]

single phase motor, pulley going to three phase generator



you would need a circuit with capacitors to make up the energy that doesn't exist at the single phases frequencies.

 

[Averagesupernova]

single phase motor, pulley going to three phase generator



you would need a circuit with capacitors to make up the energy that doesn't exist at the single phases frequencies.

I fail to see how the capacitors will 'make up energy' that doesn't exist.
-
Typically converting single phase to 3 phase is done with a rotary phase convertor, or a solid state switching device such as used in variable frequency drives. Both devices have been covered here on PF. Do a search.

 

[NascentOxygen]

What type of circuit can convert single phase electric power into three phase electric power? Can the circuit diagram be posted in this thread?

Are you wanting to do much more than demonstrate the 3 sinusoids on a CRO, or operate a tiny 3ø model motor?

Otherwise, you are talking about a heavy duty DC to AC converter producing 3 phases, and that's just to drive a small domestic 3 phase water pump.

 

[Bararontok]

Can a diagram for a solid state converter be posted here? Variable frequency drives can adjust output frequencies but how can they be used to cause two separate phase shifts in the single phase current to produce a 3-phase output?

 

[NascentOxygen]

You'd basically have 3 DC-AC converters in one big box, all synchronized so they stay perfectly in step and precisely 120° out of phase (one leading, and one lagging, the third).

 

[Bararontok]

And what keeps the three currents out of phase? Is there some kind of electronic timing mechanism that delays currents so that when the first phase is on, the other two phases are off and the same applies to the other two phases?

Perhaps electronic timer switches are connected to each output to time when the outputs will stay on and off.

 

[NascentOxygen]

And what keeps the three currents out of phase?

An oscillator can be designed (using R's, C's and transistors or op-amps) to provide the three low-power sinewaves with correct phase differences. These in turn control high power amplifiers.

Is there some kind of electronic timing mechanism that delays currents so that when the first phase is on, the other two phases are off and the same applies to the other two phases?

I assumed you'd be wanting 3ø sinusoids. If it is good enough to have 3 rough stepwise approximations, then you can use a switching arrangement and this is more efficient. But it is rare that a 3 phase application will accept squarewaves, the most efficient of all to generate.

 

[jim hardy]

take 6 volts from a doorbell transformer
phase shift by 60 degrees with RC as suggested above and invert

now you have two 6 volt sinewaves 120 degrees apart, draw the phasor diagram.
label one A and one C and the circuit common B.

Now you have three phase open delta with phase B "grounded"

run it through a power amplifier (stereo boom box amp) and into a small three phase transformer and you have a few watts. I used that to make a 3 phase 260V source for checking instruments in the power plant. But i used the LM12 power operational amplifier which is obsolete and cost $80 apiece. Use something more modern, one of those fifty watt TDA series audio amps.

 

[Bararontok]

Can the diagram for this type of design be posted so that the exact parts and their location in the circuitry can be identified?

 

[jim hardy]

Can the diagram for this type of design be posted so that the exact parts and their location in the circuitry can be identified?

my old drawings are in DCCAD which no longer runs under windows.
If i can find a hardcopy will scan it into jpg and post.
But it really is as simple as described. I used LM324 for phase shifter and LM12 for power amp. Found a small 3 phase transformer, will post its part number when locate it. I have a couple in the barn.

I used +/- 30 volt supplies for the LM12 opamps, +/-12 for LM324..
Be aware this is a linear approach so it's very inefficient, of absolutely no use for running motors or illumination
But it worked great for fooling a three phase controller circuit into thinking it's powered up.

What is your goal?
If you want it to run motors, search on three phase power for home shop. Machinist supply stores sell them, check Enco. The basic approach is run a fair sized three phase motor on single phase with no load and it will make the other two phases for you. So you parallel your 3 phase machine with that larger free running motor. It's not perfectly balanced but it'll start and run well.
The home shop machinists buy a box with capacitor and relay to get the first motor started. 3 phase motors are cheap at the salvage yard because so few people have 3 phase at home.

old jim

 

[NascentOxygen]

Can the diagram for this type of design be posted so that the exact parts and their location in the circuitry can be identified?

Sure it can, provided it is already in the public domain.

Google will find around 25,000 such circuits. We'll let you have the honor of posting a link to one that you like the look of, then we can discuss some of the finer points of it. After all, you are the one who's doing this homework.

 

[Bararontok]

Here is a diagram of the 1-phase to 3-phase converter:

http://2.bp.blogspot.com/-nyYxFye9_...00/single+phase+to+three+phase+conversion.JPG

The website posting the article about 1-phase to 3-phase conversion stated that a capacitor is used to cause a phase shift for the 3rd phase to be out of phase. But why did it not mention the use of a capacitor to cause the 2nd phase to be out of phase?

Source: http://www.electrotechnik.net/2011/07/running-three-phase-motors-with-single.html

 

[NascentOxygen]

Here is a diagram of the 1-phase to 3-phase converter:

That is not a 1-phase to 3-phase converter, not in the usual meaning of the term. What you are referring to is a dodgy arrangement that allows a small 3∅ motor to run on a single phase supply.

The website posting the article about 1-phase to 3-phase conversion stated that a capacitor is used to cause a phase shift for the 3rd phase to be out of phase. But why did it not mention the use of a capacitor to cause the 2nd phase to be out of phase?

Because they are generating only 2 phases, not 3. It's a rough and ready compromise, but it works because motors can run off 2 phases, though at reduced power.

So are you interested in operating a 3∅ motor off a single phase?

 

[Averagesupernova]

Check this link out: http://www.metalwebnews.com/howto/ph-conv/ph-conv.html

Keep in mind the date so VFD type equipment is not mentioned. A little creative google work will tell you plenty about generating 3 phase off of single phase.

 

[Bararontok]

So are you interested in operating a 3∅ motor off a single phase?

No, just interested in figuring out how to do it. Is it possible to combine 2, 2-phase converters and suddenly end up with a 3-phase converter?

 

[Bobbywhy]

Static Phase Converters

The simplest type of phase converter is generically called a static phase converter and has been in use for nearly one hundred years. This device typically consists of one or more capacitors and a relay to switch between the two capacitors once the motor has come up to speed. These units are comparatively inexpensive. They make use of the idea that a three-phase motor can be started using a capacitor in series with the third terminal of the motor. It then runs with essentially two of the three windings powered.

It is almost guaranteed that a static phase converter will do a poor job of balancing the voltages on the motor. Unless motors operated on static converters run only for short periods or deliver significantly less than half of their rated output, they will be damaged from overheating.

Some manufacturers of static converters make a simplistic statement that since two of the three windings are powered, the motor will be capable of generating two-thirds its rated power. This is misleading and could lead to damage of the motor. If the motor were loaded anywhere near two-thirds its capacity, it would be permanently damaged in short order.

The only good thing that can be said of static phase converters is that they are inexpensive. They can only operate motors, and only single-motor loads. The motors must be limited to light loads and intermittent use.

http://www.phaseconverterinfo.com/phaseconverter_static.htm

Now, if you still want to build and use one of these, you may get more technical details here: http://www.homemetalshopclub.org/projects/phconv/phconv.html

 

[jim hardy]

No, just interested in figuring out how to do it. Is it possible to combine 2, 2-phase converters and suddenly end up with a 3-phase converter?

Look up "open delta"

and do this thought experiment:

Draw a three phase delta connected transformer winding.
Focus your attention on voltages at A and C wrt B.
Observe that if you remove the single winding that's between A and C you do not change the voltage between A and C, it is defined by the other two windings' amplitude and phase.

So it is possible to use two voltage sources to make three phase.
It is in fact done quite often using two single phase transformers. Saves the cost of one transformer. Be observant of power poles in your daily travels and you'll probably see such a setup.

 

[Bararontok]

There are some circuit diagrams here for the open delta converter:

http://www.elec-toolbox.com/images/fig1-8.gif

http://2.bp.blogspot.com/_vfmOyxDCr...IGrAiDA/s1600/3_Phase_Converter_Schematic.png

http://img.photobucket.com/albums/v246/silverdoctor/UKconverter.gif

Sources:

http://www.elec-toolbox.com/usefulinfo/xfmr-3ph.htm

http://www.circuitlab.org/2011/12/schematics-3-phase-converter-miller.html [Broken]

http://www.practicalmachinist.com/v...e-phase-converter-home-united-kingdom-102960/

Are these the correct schematics for such converters?

****

Additionally is this diagram for 3-phase to 1-phase conversion correct:

http://img407.imageshack.us/img407/8593/3phaseto1phaseconverter.png [Broken]

The 3 power phases can be connected to inductors which will be placed in proximity to a larger inductor that is the same size as the sum of the volume of the 3 smaller inductors. The power will then be transferred through electromagnetic radiation and the inductors will thus serve as a transformer with all the energy being fed to one inductor. Because of the overlap of currents coming from the 3-phase power supply, the output inductor can be connected to a rectifier which will then be connected to an inverter to produce a 1-phase output.

 

[Averagesupernova]

Bararontok, where did you get that schematic? I would like to see the smoke pour out of that transformer when hooked up. Incidentally, why is it needed to 'convert' from 3 phase to single phase? Just hook on to a pair of wires from a 3 phase source and you have your single phase. Unless you want to provide more power on the generated single phase than one phase of the original 3 can provide, I see no reason to have any kind of convertor.

 

[Bararontok]

Perhaps it is possible that the transformer can be designed with a wattage rating that is higher than the input power of the 3-phase supply to avoid overloading and enable all the power from the 3 phases to be converted into a combined 1-phase output but it is correct that a pair of wires can be connected to a single pair of terminals from a 3-phase circuit and 1/3 of the power can be used as the 1-phase output. The 3-phase to 1-phase converter circuit diagram was made by the thread originator purely for curiosity's sake and not for any purpose.

 

[Averagesupernova]

My point is that you cannot just throw 3 windings, one from each phase, on the same core. They will fight each other.

 

[jim hardy]

Are these the correct schematics for such converters?

****

The first one is what i was describing.

The rest are more complex devices that start with single phase and make a third phase by shifting angle with capacitors. Note the first one starts with three phases and ends with xame three phases.
Point being, two vectors define three points and if those points are vertices of an equilateral triangle they are same geometry as three phase phasors.

Second one appears to employ a motor , and that scheme actually works fairly well. My neighbor has one in his home workshop. It actually uses the motor as an induction generator for part of each cycle to make third phase.

Third one is similar to second but with improvement of a step-up transformer. Works quite well for running motors. I've done it myself. Draw the phasor diagram and you'll see it.

 

[NascentOxygen]

Additionally is this diagram for 3-phase to 1-phase conversion correct:

http://img407.imageshack.us/img407/8593/3phaseto1phaseconverter.png [Broken]

The 3 power phases can be connected to inductors which will be placed in proximity to a larger inductor that is the same size as the sum of the volume of the 3 smaller inductors. The power will then be transferred through electromagnetic radiation and the inductors will thus serve as a transformer with all the energy being fed to one inductor. Because of the overlap of currents coming from the 3-phase power supply, the output inductor can be connected to a rectifier which will then be connected to an inverter to produce a 1-phase output.

It represents in principle a method to convert from 3ϕ to single phase: you rectify the 3ϕ and use that DC to power an inverter. But I would need to examine his "transformer" arrangement very closely (at a low testing voltage) before anything else, as it is unusual. You might consider using an AC→DC→AC inverter like this if your need was for a single phase frequency that was different from that of the mains 3ϕ, or if it were essential that you must load all 3 phases equally (this could be a power authority condition). Otherwise, why not just connect your single phase motor (or its associated transformer) between any two of the phases from your 3ϕ supply?

 

[SunnyBoyNY]

There are a few ways to convert power from a single phase AC to three phase AC. Nowadays, the most popular way is to use a single phase inverter/rectifier that converts the AC voltage and current waveforms into DC. The DC link voltage is usually regulated by the rectifier. A typical rectifier that is capable of bidirectional power flow is a simple H-bridge. A full bridge is also needed to obtain high power factor -- the drawn current is also sinusoidal.

Now, we have the DC link to which we connect a three phase inverter. The simplest topology are three phase legs. The middle of each leg is connected to one phase of the utility via an inductor. It is not possible to connect two voltage sources without a magnetic element such as inductor.

The three phase currents are regulated via current control algorithms by controlling the duty ratios of the three legs (PWM scheme). This topic has been rather mature for the past ten years. To obtain maximum efficiency and power factor, each phased current is a scaled version of its phase voltage. This way there is no reactive current in the system.

These two systems are called back-to-back AC/DC/AC system and it is a voltage sourced system.

To convert power directly from AC to AC you would use a cycloconverter. These guys can handle high currents and voltages as they are mostly based on SCRs and triacs - current sourced converters.

A picture of a three phase inverter is here: http://techno-fandom.org/~hobbit/cars/training/800/m112.jpg

Single phase inverter (bottom part): http://solar.smps.us/grid-tie-inverter-schematic.png

And here are two back-to-back VSC systems: http://ars.sciencedirect.com/content/image/1-s2.0-S0960148111006550-gr1.jpg [Broken]

Source: power electronics engineer

 

[Bararontok]

As a potential system for 1-phase to 3-phase conversion, would it also be possible to use two capacitors to change the phase angle of the two other phase terminals where C2>C1 so that the third phase terminal will be out of phase to a greater degree than the second terminal? The diagram is below:

http://img37.imageshack.us/img37/3186/1phaseto3phaseconverter.png [Broken]

 

[NascentOxygen]

Would it also be possible to use two capacitors to change the phase angle of the two other phase terminals where C2>C1 so that the third phase terminal will be out of phase to a greater degree than the second terminal?

This is a thought experiment, is it? You want to take one of the phases from 3ɸ and mess with its angle? Sure you can do that. Simply adding a capacitor to the phase from the mains won't do it, but if you use a transformer-inductor and capacitor as illustrated in earlier schemes, you could change the angle. It probably wouldn't have any practical use, as far as I can see -- I presume 120° is the all round optimum for rotating machinery.

 

[Bararontok]

This is a thought experiment, is it? You want to take one of the phases from 3ɸ and mess with its angle? Sure you can do that. Simply adding a capacitor to the phase from the mains won't do it, but if you use a transformer-inductor and capacitor as illustrated in earlier schemes, you could change the angle. It probably wouldn't have any practical use, as far as I can see -- I presume 120° is the all round optimum for rotating machinery.

No, the intention was to use the device to convert 1-phase to 3-phase current. The supply would be split into 3 sets of terminals, 1 with no capacitor and the other 2 with capacitors of different ratings to give each one a different phase angle.

 

[SunnyBoyNY]

No, the intention was to use the device to convert 1-phase to 3-phase current. The supply would be split into 3 sets of terminals, 1 with no capacitor and the other 2 with capacitors of different ratings to give each one a different phase angle.

Do not forget that virtually all three phase machines - PM, induction motors and VRM have their torque dependent on the current.

For example, if a PM machine has sinusoidal windings, you need to excite those windings with current that is exactly in the phase with the back EMF. Since the stator winding inductance is constant and your capacitors values are constant, the intended phase shift would work for one frequency only.

Why don't you buy a flux drive and hook it up to a full-bridge rectifier to get 1ph->DC->3ph ac variable speed?

 

[Bararontok]

Do not forget that virtually all three phase machines - PM, induction motors and VRM have their torque dependent on the current.

For example, if a PM machine has sinusoidal windings, you need to excite those windings with current that is exactly in the phase with the back EMF. Since the stator winding inductance is constant and your capacitors values are constant, the intended phase shift would work for one frequency only.

Why don't you buy a flux drive and hook it up to a full-bridge rectifier to get 1ph->DC->3ph ac variable speed?

But each of the two capacitors will have differing values, is that not enough to produce three phase angles?

 

[SunnyBoyNY]

But each of the two capacitors will have differing values, is that not enough to produce three phase angles?

First, if the two capacitors are connected to a DC link (which is a voltage source) then they are slaved to it. Voltage on them would not have any phase shift. There would need to be an inductive element between the voltage source and these capacitors to produce phase shift.

Also, the phase shift would appear only at a certain frequency - so you would need to pull current at a certain frequency to have the two cap voltages phase shifted 120d to each other.

At zero frequency the caps would resonate with their phase inductors. Provided non-zero resistance, it's a second order damped system. When the initial ringing dies out, both caps will have the same DC voltage as the source.

 

[Bararontok]

First, if the two capacitors are connected to a DC link (which is a voltage source) then they are slaved to it. Voltage on them would not have any phase shift. There would need to be an inductive element between the voltage source and these capacitors to produce phase shift.

Also, the phase shift would appear only at a certain frequency - so you would need to pull current at a certain frequency to have the two cap voltages phase shifted 120d to each other.

At zero frequency the caps would resonate with their phase inductors. Provided non-zero resistance, it's a second order damped system. When the initial ringing dies out, both caps will have the same DC voltage as the source.

The source is a 1-phase AC source, not a DC source.

 

[NascentOxygen]

No, the intention was to use the device to convert 1-phase to 3-phase current. The supply would be split into 3 sets of terminals, 1 with no capacitor and the other 2 with capacitors of different ratings to give each one a different phase angle.

A capacitor in series with the mains is a dodgy arrangement that allows a motor to see a phase with a leading angle, but that won't be anywhere near 120° as far as I can see, though I'm no expert. It relies on a characteristic of the motor winding. If you use a different motor, you'll need a different capacitor or performance will suffer even more.

You won't generate 120° lead just by adding a series capacitor to your single phase supply. It may be possible to generate 120° lead and lag using a more complex passive network, carefully designed, but it will fall apart as soon as you try to draw anything but miniscule current from it. You may be able to power a thimble-sized miniature model 3ɸ motor as a demonstration, but nothing of any use, is my thinking.

If it was as simple as you picture it, then there would be no need for the big, heavy duty complex circuits that are employed to do the task.

 

[Bararontok]

The capacitors of the design are in parallel to the load. Though it may be correct that the capacitor would have to have a different value for different motors but the device can have a wattage rating label placed on it to ensure that the supply is not overloaded since other types of power supplies already have these labels anyway.

 

[NascentOxygen]

The capacitors of the design are in parallel to the load.

If the mains see the capacitors in parallel with the load then those capacitors can do nothing more than PF correction. The 1ɸ mains has to see them in series with something to achieve a phase lead.