Many a times we find it crucial and handy to possess a true three phase signal for evaluating many different electronic configurations such three phase inverters, three phase motors, converters etc.
Since it's not so easy to incorporate single phase to three phase conversion quickly we find this particular implementation difficult to acquire and enforce. The proposed circuit enables the above discussed well calculated spaced and positioned sine waves outputs to be generated from a single master input source.
Circuit Operation
The circuit functioning of the three phase waveform generator circuit may be understood with the help of the following explanation:
An input sine sample waveform is fed across the point "input" and ground of the circuit.This input signal gets inverted and buffered by the unity gain opamp A1. This inverted and buffered signal acquired at the output of A1 now becomes the new master signal for the forthcoming processing.
The above buffered master signal gets once again inverted and buffered by the next unity gain opamp A2 creating an output with zero degree initial phase across the points "Phase1"
Simultaneously, the master signal from A1 output is phase shifted by 60 degrees via the RC network R1, C1, and fed to the input of A4.
A4 is set up as a non-inverting opamp with a gain of 2 in order to make up for the signal-loss in the RC configuration.
On account of the fact that the master signal is phase shifted 180 degrees from the input signal, and further shifted to an additional 60 degrees by the RC network, the ultimate output waveform gets shifted by 240 degrees, and constitutes the "Phase3" signal.
Now, the next unity gain amp A3 sums up the A1 output (0 degrees) with A4 output (240 degrees), creating a 300 degree phase shifted signal at its pin#9, which is in turn inverted appropriately, shifting the phase to an extra 180 degrees, creating the intended 120 degree phase signal across its output indicated as "Phase2".
The circuit is intentionally wired up to work with a fixed frequency in order to yield better accuracy.
Fixed values are used for R1 and C1 for rendering the intended, accurate 60 degree phase shifts.
For specific customized frequencies, you may use the following formula:
R1 = (√3 x 10^6) / (2π x F x C)
R1 = (1.732 x 10^6) / (6.28 x F x C1)
where:
R1 is in kohms
C1 is in uf
Circuit Diagram

Parts List
All R = 10 kohms
A1---A4 = LM324
Supply = +/- 12vdc
| Frequency (hz) | R1 (kohms) | C1 (nf) |
|---|---|---|
| 1000 | 2.7 | 100 |
| 400 | 6.8 | 100 |
| 60 | 4.7 | 1000 |
| 50 | 5.6 | 1000 |
The above design was investigated by Mr. Abu-Hafss and appropriately corrected for obtaining legitimate responses from the circuit, the following images provide a detailed info regarding the same:
Feedback from Mr. Abu-Hafss:
I needed a 15VAC 3-phase supply to test 3-phase rectifiers. I simulated this circuit the other day but failed to get proper results. Today, I made it work.
IC A2 and resistors connected to pin 6 could be eliminated. The resistor between pin 7 and 9 could be connected between the main input and pin 9. Phase-1 output can be collected from the original AC input. Phase 2 and 3 can be collected as indicated in the circuit.

However, my actual requirement could not be fulfilled. When these 3 phases are connected to a 3-phase rectifier, the wave form of phase 2 and 3 gets disturbed. I tried with the original circuit, in that case all three phases gets disturbed

Finally got a solution! A 100nF capacitor connected in series with each phase and the rectifier solved the problem to a great extent.
Though the rectified output is not consistent but, it is quite acceptable

Update: The following image shows a much simpler alternative for generating 3 phase signals with accuracy and without complicated adjustments:





Comments
Hi. Based on the 3-ph signal generation circuit, can you please mail to me a driver circuit for 24v BLDC motor to have enough torque to drive a 2.5 mm wire feeder in a welding machine?
Hi, Is your BLDC motor with sensors or without sensors? I would rather recommend using a dedicated driver circuit for a BLDC motor.
Hi swatgam, please help me with a simple three phase inverter circuit diagram, ill appreciate thank you
Hi Adon, you can get a few 3 phase inverter ideas from the following link:
https://www.homemade-circuits.com/three-phase-inverter-circuit/
Dear Swatgam,
Very interesting topic which covers an issue that I have been struggling with the for the past months. I am trying to design a circuit (with minimum electronics knowledge) to create a triphase 26VAC 400 hz power supply to run an vintage aircraft gyroscope motor. I think I was able to generate a 400hz singal via arduino (sine wave and/or square wave), bu with the rest I am fully stuck. Would you be kind enogh to share a circuit desing for my project ? kindest regards okanSacli
Thank you Okan,
I think you are looking for a 26V 3 phase inverter circuit, which can be perhaps build using a 3 phase driver IC. If you are having a 3 phase signal from an Arduino then you could try applying it to the following circuit for getting a 3 phase AC output. The 600 V can be replaced with 26 V DC. The mosfets can be any suitable ones below 100 V VDS
https://www.homemade-circuits.com/wp-content/uploads/2022/08/3-phase-inverter-circuit.jpg
This four op-amp setup is a well known way to produce a three-phase output from a single phase and known as a phase tripler. Normally R1 is a trim-pot (to adjust for different freq). I have used lab-equipment when teaching since at least 2010 based on this setup. Note that it is important to select high-current op-amps since they can get quite warm especially if driving capacitive loads (op-amps don’t like cap load).
Thank you sharing this useful information. Appreciate it!
Sylvain,
Where is the circuit diagram?
I have a Bendix HSI that uses 26VAC 400 hrz,
I’m not sure if its 2 phase or 3 phase
but really urgently need to build this power supply
driven by 12.0 to 13.8 vdc.
Dear Swatagam,
regarding your Fig. 2, could you explain the function of the Diodes? For me it looks the result would be the same without them. Further are the other Pins of the OpAmp not used?
Thanks for your time.
Dear Sol, the diodes are necessary, and the configuration is called diode shunt feedback-clipping. I cannot fully interpret the diagram, since it was referred from some other site and it is not my design. Other pins except the supply pins are not used.
Dear Swagatam,
in Fig.2 you write U=LF411
I count in you fig. 9 pins, but the LF 411 seems to have 8?
Dear Sol, the circuit uses 3 separate LF411 ICs, however any other op amp can be also tried such as IC 741.
Sorry forgot this question.
If I have a 24 VAC 3 Amp source, will it grill the circuit?
For 24V you can IC LM324 for the op amps
Good your publication, your project gave me an idea to convert from a single-phase network system to a three-phase system but without OPAMPs, only with transformers and electronically controlled, it is phase shift and module according to the load variation of the electric motor… thanks… greetings
Thank you, glad it helped! wish you all the best!
Congrats for your nice job!
Tell me please if this circuit is capable to drive the module PS21245-E.
You see I have one PCB from an old inverter air condition with built in the above mentioned module and I’d like to convert it to a 3-phase power supply unit.
Thanks in advance
Petros
Thank you, yes you can use it for feeding HIN, LIN inputs of the mentioned 3 phase driver module
Because there are six (6) mosfets into this module and I’m not familiar with these deep waters can you give some more help. Can you tell me which are these Hin & Lin on the following Data sheet.
pdf.datasheetcatalog.com/datasheet/MitsubishiElectricCorporation/mXqtqy.pdf
Thanks for your time
Petros
Please look for these pinouts of the IC
HIN = High-side input (PWM)
LIN = Low-side Input (PWM)
Hi,
We have connected the circuit as given in picture3 as done by Mr. Abu-Hafss.
The output for 120degree is not coming , it is only 60degree phase shift. for obtaining 120 degree phase shift i had to connect one more RC circuit at the output of U3. please help with this.
Please explain the working of the summer used.
Also, what is the value (1.732*106) is the equation, R1 = (1.732 x 106) / (6.28 x F x C1)
Thank you in Advance.
Chaithra
Hi, This circuit was referred from another article long ago and not designed by me, therefore regarding the circuit functioning I won’t be able to suggest much.
Regarding the formula, I have updated it in the article with more details. Most of the variables in it are constants.
Hello. for the updated circuit, could you tell me what the resistance and capacitance is to make the 50 Hz frequency? And how to change the frequency with the resistance?
Hello, I assume that it is originally calculated for 50Hz, nonetheless you can practically test it and tweak the results if required by changing the R or C values accordingly…
what software simulator you used to show the graph ?
Hi Amir
The software is LT Spice IV, it free to download at http://www.linear.com/designtools/software/
Hi Swagatam,
Can the variable frequency signal [ from 1 to 60 Hz] circuit from timer 555 to feeds into its controlled board to run a salvage 3phase induction washer motor from 120VAC ?
Hi Cong, if you change the frequency in IC 555, it will also proportionately vary its PWM, it's better to use IC 4060 for getting a variable and perfect 50% duty cycle frequency…but the output will be a square wave which might not suit your induction motor
Can I use Motor of 3-Phase Motor 2 HP with above sequence . ?
yes you can
Hello Mr. Majumdar, I would like to ask you how did you get the 60 degrees phase shift to sum up with the 240 degrees from A4? thanks for your time
thanks for updating your experience!
Oh ok, hehe. I have to check then. The thing is that I used this network for my work. Works fine and I understand all the process except that part I asked you.
Thanks anyway for your time.
Hello Rich This design is not mine, rather taken from another source, I have not yet investigated the calculations so far, so can't comment regrading it..
Hie swagatam plz can you help me build a three dc to ac inverter 48/240
Hi Adonis, I already have a 48V inverter circuit in this blog, do you want 3 separate inverters or tied up together.
the request
sir,need pwm coding for three phase multilevel inverter @Swagatam Majumdar
sorry Basit, I have no idea about it….
Dear Majumdar,
what is the input value of the ac in the above circuit..? and how much will be the out put ac?
Actually i need 3 phase 24 or 12V ac at60Hz out put. and i have 12 and 24 v ac 60Hz.
Dear Basheer,
the output will be equal to the peak to peak value of the IC supply voltage, if the supply is 12-0-12V then the output will be also in accordance with this value,
The input could also be equal to this value but not above this to avoid voltage clipping.
sir good day…sir have you design a much simple circuit for a single phase to three phase system…
you can refer to the following article:
https://www.homemade-circuits.com/2014/12/3-phase-signal-generator-using.html
Hi Genius,
I want a circuit that can convert single phase AC( 220v 15 amps 50-60 hz) to three phase 120 deg phase shift ( 440 v phase to phase 50-60 hz) that can run a 3hp induction motor.
I am an Automation engg. from Mech. background and I have been looking for this circuit from a long time. it happens in VFD but they are too costly for the product that we are planning. Will really appreciate your help in it.
Thanks Debu,
I have already published a few related circuits in this website, you can refer to those in the below given links, if you have problems understanding the concepts you can feel free to interact with me:
https://www.homemade-circuits.com/2014/12/3-phase-vfd-circuit.html
https://www.homemade-circuits.com/2013/10/three-phase-inverter-circuit.html
Just got a solution! A 100nF capacitor connected in series with each phase and the rectifier solved the problem to great extend.
https://dl.dropboxusercontent.com/u/20969135/Screenshot%202014-11-22%2001.23.53.png
Though the rectified output is not consistent but, it is quite acceptable.
Thanks for uploading my findings.
However, this line
"Though the rectified output is not consistent but, it is quite acceptable."
should be mentioned at the end.
Thank you Abu-Hafss,
That's very impressive and this will be quite handy for all the readers out there who may be trying to build the above circuit successfully.
I'll update the corrected images in the above article soon
Hi Swagtam
I needed a 15VAC 3-phase supply to test 3-phase rectifiers. I simulated this circuit the other day but failed to get proper results. Today, I made it work.
IC A2 and resistors connected to pin 6 could be eliminated. The resistor between pin 7 and 9 could be connected between the main input and pin 9. Phase-1 output can be collected from the original AC input. Phase 2 and 3 can be collected as indicated in the circuit.
https://dl.dropboxusercontent.com/u/20969135/Screenshot%202014-11-22%2000.58.38.png
However, my actual requirement could not be fulfilled. When these 3 phases are connected to a 3-phase rectifier, the wave form of phase 2 and 3 gets disturbed. I tried with the original circuit, in that case all three phases gets disturbed 🙁
https://dl.dropboxusercontent.com/u/20969135/Screenshot%202014-11-22%2001.00.40.png
Any suggestions??
Will u pls post or mail the original article link