Elektroda.com
Elektroda.com
X
This content has been translated flag-pl » flag-en View the original version here.
  • I would like to introduce a simple design of a small inverter for a 3 phase motor with a power of about 400W.
    View of the inverter in the photos below
    Simple 1/3 phase inverter with PFC Simple 1/3 phase inverter with PFC Simple 1/3 phase inverter with PFC Simple 1/3 phase inverter with PFC Simple 1/3 phase inverter with PFC Simple 1/3 phase inverter with PFC
    The inverter is powered from one phase, with alternating voltage 230 V. The output gives a 3-phase voltage 3x 230V. The inverter schematic, software, PCB design are in the attachments.
    Falownik_S...ktroda.zip Download (14.75 kB) FNA41560-k...rojekt.zip Download (912.46 kB) schemat fa... 09-27.pdf Download (173.45 kB)
    I think most of the readers who will be interested in this post know how the inverter works and what it is for. The reason why I built the inverter was the price of the main control element FNA41560, which on aliexpress is about 8 PLN. However, as it turned out later, the cost of the remaining elements was about PLN 200. The most expensive components are capacitors and a choke. So it's worth considering before you start building an inverter, because you can probably buy a ready-made one at this price. There is a PFC system at the inverter input, which ensures that the current drawn from the grid is not distorted and is in phase with the voltage. However, the degree of deformation depends on the load as shown in the pictures below
    Simple 1/3 phase inverter with PFC Simple 1/3 phase inverter with PFC
    The voltage on the filter capacitors is approximately 430 V without load and drops to 400 V when the PFC is loaded. In the inverter, the transistors commutate at a frequency of 5kHz, because for this frequency the FNA41560 has been optimized, the dead times for transistors are about 1.2 us as in the pictures
    Simple 1/3 phase inverter with PFC Simple 1/3 phase inverter with PFC

    The inverter has high temperature protection set at 105 ?C, 5 A short circuit, and high voltage on 460 V filtering capacitors. cycle by cycle mode, ie it automatically resets at the end of each PWM cycle (acts as a current limitation). This can be changed in the P1FLTACON microcontroller register to the "latch" mode. In the "latch" mode, the PWM generators turn off and stop until the marker is deleted. The frequency is regulated by a multi-turn potentiometer with a resolution of 0.1 Hz. Adjustable frequency range from 1 Hz to 80 Hz. The useful range starts at 5Hz. The VSM -space vektor modulation algorithm was used to generate the output waveform, which allows the maximum use of DC voltage supplied to the FNA41560 module. More details on this type of control can be found, for example, on the microchip page
    https://microchipdeveloper.com/mct5001:space-vector-modulation

    The differences in the shapes of the waveforms generated by SPWM and SVM are shown in the figures below. In the case of SPWM, the maximum phase-to-phase voltage at the inverter output can be at most ?3 / 2 x Udc in the case of SVM is Udc - if losses in transistors are neglected. In fact, the SVM method gives about 15% higher output voltage compared to the Sine PWM method.
    Simple 1/3 phase inverter with PFC Simple 1/3 phase inverter with PFC
    It should be remembered that the power supply of the system is not isolated from the mains and special care should be taken during commissioning and use. The mass of the system in relation to the ground is potentially life-threatening .
    However, if someone would like to build and run an inverter, below I am giving some helpful tips and attaching videos from commissioning. The sounds that can be heard are not generated by friction of the bearings or the ventilator, I do not know how they arise, I can only confirm that the motor connected to a 3-phase sinusoidal source does not generate them.
    [movie: fdf5ab2e81] https://filmy.elektroda.pl/37_1601565891.mp4 [/ movie: fdf5ab2e81]
    In the second video, between the inverter and the motor, I turned on the LC filter 3x L = 1.5mH and 3 x C = 0.68uF, which in my opinion made the sounds a bit quieter.
    [movie: fdf5ab2e81] https://filmy.elektroda.pl/59_1601566389.mp4 [/ movie: fdf5ab2e81]

    The assembly suggests soldering and then starting the PFC part, soldering the MC33262 integrated circuit, rectifier bridge, diode D11, Q1 transistor and choke, filtering capacitors C17 and C22 to which you need to solder 470 k? resistors to each one as in the picture. Winding (3 turns) should be wound on the choke, which will power the MC33262 integrated circuit. I used a cable from a computer network cable (so-called path cord) for winding it. The end and beginning of the windings are important and should be connected as in the application note. I used an inductance meter to determine the start and end of the windings. I connected the wound winding to the choke winding in series and compared the measured inductance values. In the case of obtaining a higher value of inductance, I obtained a connection between the end and the beginning of the windings. The power supply to the inverter should be led through starting thermistors to limit the current flowing through the D11 diode, and for which there is not enough space on the board, or to use another solution that will limit the inrush current. I used 2 NTC6D-15 thermistors with a maximum current of 5A. Direct connection to the mains may damage the D11 diode. After soldering the PFC circuit elements, I soldered two 100W / 230V incandescent bulbs connected in series to the capacitors, thus I checked whether the PFC works. The voltage on the bulbs should be 400V
    The next step I made was soldering and checking the operation of the transformerless power supply built on the LNK306. At the output of which you should measure the voltage, which should be 15 V. At the very end, I soldered the FNA41560 and the microcontroller that should be programmed in the system. The J3 compatible Pickit3 connector is used for programming.

    In order to start the system, apart from turning on the power, give logical number 1 from the RA2 output to the RA3 input, then the inverter starts up to the speed set by the potentiometer with the ramp for about 5 seconds. Connecting RA3 to ground causes the inverter to stop - the transistors are turned off.
    It is important to control the starting or stopping of the inverter from the RA2 output (pin 1 on J4 as shown in the diagram) because in the event of too high temperature or other disturbances the state on R2 changes to low and the inverter turns off.

    Noteworthy is the FVO output (pin 11) of the FNA41560 chipwhich is shorted to ground when the supply voltage is less than 12 V and also when the system is not powered. This is indicated by the fault diode, during the startup and initial tests, when the power supply from the pickit3 programmer is connected to the board, do not worry about it - you just have to remember that when this diode is lit on the microcontroller output, no PWM waveforms are generated.
    To get rid of this error and get PWM waveforms at the output of the microcontroller, temporarily disconnect the power from the programmer and connect 15 V to the FNA41560. Of course, we only perform these activities when the inverter is disconnected from the power grid.

    The maximum output voltage of the inverter is obtained at a frequency of 60 Hz.
    For low frequencies, the voltage between 1 Hz - 5 Hz is constant. Above 5Hz, U / f = constant increases.
    The system was assembled on a board with dimensions of 100 x 100 mm so that the cost of the tiles was $ 2.
    The program was written in C in the MPLABX environment.
    Finally, I would like to add that, apart from the starting thermistors, the inverter should be powered by an anti-interference filter.

    Cool? Ranking DIY
    About Author
    Tatalek
    Level 12  
    Offline 
    Tatalek wrote 27 posts with rating 240. Live in city Gdansk. Been with us since 2008 year.
  • #2
    szymon122
    Level 38  
    Is this type of inverter suitable for the construction of an inverter for an electrospindle whose speed reaches 20k RPM?

    What are these capacitors and a choke that cost PLN 200?
    Did you wind the choke yourself?
  • #3
    Tatalek
    Level 12  
    The cost of all elements is around PLN 200, capacitors and a choke are around PLN 70
    the choke is ready from the DTMSS-47 / 0.22 / 15-V feristor, I only wound 3 coils to power the MC33262
    I do not know what the electro-spindle is, at what frequency it works
  • #4
    VaM VampirE
    Level 21  
    Spinning spindles up to 24,000 rpm, have a power supply of 3 * 220V 400Hz.
    High-speed spindles working at e.g. 60,000 rpm require 600Hz inverters, or even up to 1000Hz they come.

    Certainly a niche would be an inverter working with two-phase spindles, they are quite exotic and inverters for them are even more exotic.

    As for power, it is a fact, buying a 400Hz inverter with a power of up to 0.55kW is often around PLN 200. It is enough to post an advertisement on some forum.
  • #5
    DJ_KLIMA
    Level 20  
    Just wait for the Chinese module, if it does not already exist :)
  • #6
    neo_84
    Level 15  
    The engine makes a strange noise because something is wrong. It can be seen in the 2nd film, it "walks" squarely at low speed and it should be smooth.
  • #7
    pawelr98
    Level 39  
    A very nice inverter.
    Although I would get + 15V using L2, adding another winding and another voltage regulator behind it, working on a lower DC voltage.

    PS. A friend from the university of technology?
  • #8
    Tatalek
    Level 12  
    pawelr98 wrote:
    A very nice inverter.
    Although I would get + 15V using L2, adding another winding and another voltage regulator behind it, working on a lower DC voltage.

    PS. A friend from the university of technology?


    In fact, with this + 15V, I will remember a very good idea and I will probably use it in the future
    I used to study there, now seagoing ships, i.e. from the shipyards :)

    Added after 5 [minutes]:

    neo_84 wrote:
    The engine makes a strange noise because something is wrong. It can be seen in the 2nd film, it "walks" squarely at low speed and it should be smooth.


    I think you are right and I "trumpeted success" too soon
    I checked that I have a constant voltage component at the inverter output, it should not be so. It seems I haven't tested everything thoroughly.
    I have to find out why. I'll let you know when I find it, and I'll work on it over the weekend.
  • #9
    pawelr98
    Level 39  
    Tatalek wrote:

    In fact, with this + 15V, I will remember a very good idea and I will probably use it in the future
    I used to study there, now seagoing ships, i.e. from the shipyards :)


    Well, if the PFC chip is powered, why not the rest?
    It's worth using what you have.

    And I asked about the polytechnic because the panel with the derived voltages looked a bit familiar.

    Even when comparing the costs, there is the fact that inverters with 3x230V power supply are very cheap and can be adapted to a single phase supply at a low cost (connecting an additional capacitor to the DC bus and possibly an external rectifier if the power is high).
    I bought one myself and I didn't do anything, I just turned off the no-phase alarm. But the PFC would be useful if there was a higher voltage available at the output.
  • #10
    Anonymous
    Level 1  
  • #11
    coberr
    Level 20  
    A very interesting, quite complicated and rare design on the electrode.

    As for the constituents and any circuses at the inverter output:
    I think the first thing you should verify are the bootstrap capacitors.

    Contrary to appearances, the values here are very critical and exaggerating the capacities may, under certain unfavorable conditions, even damage the power stages - and thus - directly also the control stages. It is about fluctuations and instability of the voltage supplying the drivers. (depending on the carrier frequency of the PWM waveform and the pulse filling for the modulating frequency) The DEAD TIME parameter generated by the processor for the IGBT module is also important here

    In my opinion - instead of the 10uF value - you should use values in the range 100-220nF and only then verify the inverter's operation and any "cribs" at its output at different frequencies.
    Capacitors - good ceramics or polypropylene - or foil ones.
    Zener diodes - although the manufacturer recommends a value of 22V - I would give 20V or even 18V of power, of course 1W. (the more that the module's power supply is 15V)

    (although I can see that now - that the datasheet even has a value of 22uF - but everything will depend on the parameters of the PWM waveform). Very often, in any applications with bootstraps, the values of these capacitors were exactly as described above.

    Legs 16-17 - what was the idea of inserting the diode there? "protection of the system against possible accidental high voltage" coming from the module?
    The LED in this place seems to be completely redundant. (and maybe - it can lead to problems with the system's operation) The protection here should be a zener diode (e.g. 16V) and a resistor with a small value from the supply side)
    An additional error is the lack of a blocking capacitor in this place - the value of which should be 0.68-1.5uF (good polypropylene or foil ceramics).

    module leg 3 and point GND-R2 - as close to these points should also be a good quality polypropylene or foil capacitor 0.1-0.22uF.

    it's like "fast" - probably a few things still could be improved.

    Added after 2 [minutes]:

    Awyrdonyt wrote:
    Such an inverter is suitable for long-term change of engine speed?


    what does the colleague mean exactly?
  • #12
    Anonymous
    Level 1  
  • #13
    coberr
    Level 20  
    it depends - what power is this engine, what drives it and how much it should be slowed down.
  • #14
    Tatalek
    Level 12  
    neo_84 wrote:
    The engine makes a strange noise because something is wrong. It can be seen in the 2nd film, it "walks" squarely at low speed and it should be smooth.



    I have to find out why. I'll let you know when I find it, and I'll work on it over the weekend. [/ Quote]

    I was able to find the cause of the engine noises. For a short while the FLTA input of the microcontroller was activated, which caused accidental only the transistors (hence the DC component at the output). When this input was disabled in the handler, the sounds stopped. I soldered a capacitor between ground and the input 1.5nF capacitor and turned on the support again - it helped. Simple 1/3 phase inverter with PFC
    and updated the schematic
    Falownik_e...te_sch.pdf Download (174.47 kB)
    and I made a video
    [movie: e69b769815] https://filmy.elektroda.pl/22_1601739798.mp4 [/ movie: e69b769815]
  • #15
    żarówka rtęciowa
    Level 37  
    Hello

    Can the cheaper L6562 IC with PFC correction be used in this design?
  • #16
    hubiwit
    Level 25  
    Hello, how does this PFC MC33262 work in the system of this inverter - in the datasheet it controls a mosfet that turns on the inductance to compensate for cutting the mains voltage at the peaks, generated by the increase in current consumption by the device, here in the inverter, the mosfet turns on resistance?
  • #17
    Janusz_kk
    Level 37  
    After all, the author writes that the choke, you even have the type given. The fact that the diagram is a bit poorly drawn does not change anything, there is an explanation in the text, you have to read and not just watch the pictures :)
  • #18
    hubiwit
    Level 25  
    Thanks for the answer, I compared the schematic with the schematic from the datasheet and I thought I didn't know myself :D
  • #21
    Tatalek
    Level 12  
    pustan wrote:
    I often browse Russian-language websites and sometimes I find something interesting.
    https://radioskot.ru/publ/3_faznyj_invertor_ot_220_v/1-1-0-1563
    Who's Who ...


    Respect for Russians, they work quickly, I think they have not copied the most important one yet
    that you need to solder a capacitor to the FLT input and that the short circuit protection does not activate because the engine will make sounds
    as if something is spawning in it - I think they will catch up soon :)
  • #22
    zbyszkok
    Level 20  
    If this were a Russian design for L, N would probably be Cyrillic :-) .
    It does not appear from the text that the USSR :-) wrote that it was his project.
  • #23
    szelus
    Level 34  
    On the contrary, there is even a link to this thread at the end.
  • #24
    zbyszkok
    Level 20  
    Shelus you are right I did not read the Orginal
  • #25
    pustan
    Level 19  
    But the note Original was placed after my information appeared on the Electrode. If it had been there, I would not have written
    "Who's Who ..."
  • #27
    Bojleros
    Level 16  
    One of the methods of fighting noise is blurring the spectrum of the generated current. There are several methods for this, but in general, if you use a strong enough DSP modulator, you can pseudo-randomly vary the pulse frequency (carrier) within a certain range. And the cycle ... instead of tonal noise you will have something approaching white noise which is less bothersome for the ear (at least for a short while).

    Added after 11 [minutes]:

    Tatalek wrote:
    The maximum output voltage of the inverter is obtained at a frequency of 60 Hz.
    For low frequencies, the voltage between 1 Hz - 5 Hz is constant. Above 5Hz, U / f = constant increases.


    neo_84 wrote:
    The engine makes a strange noise because something is wrong. It can be seen in the 2nd film, it "walks" squarely at low speed and it should be smooth.


    Maybe too little voltage boost in the low frequency range. By the way, the question is whether this inverter is intended for a 50Hz or 60Hz motor? If the high V / f point is at 60Hz and we close a 50Hz motor, it will always work suboptimally.
  • #28
    111lisu
    Level 16  
    Some time ago I came across this very interesting thread. My friend bought an external unit of an LG air conditioner with approximately 7kW of heating power, for which there is no internal unit and there is no way to turn it on (it is only supposed to "give" heat). Inside there is an FSAM20SM60A ONSEMI and the IGBT is controlled by a TLP251 optocoupler, probably the same CPU, but I haven't thought about it thoroughly yet. I have a question: if I added MC332623 and DSP (dsPIC33F), would there be a chance to control this compressor? I apologize for not preparing for the topic due to the late hour, but I did not analyze the differences between FSAM20SM60A and FNA41560

    Simple 1/3 phase inverter with PFC
    Simple 1/3 phase inverter with PFC
  • #29
    Tatalek
    Level 12  
    In my opinion, the main difference is the signal controlling the IGBT module. In the case of the FSAM20SM60A system, the low level turns on the IGBT transistor located in the module. In the case of the FNA41560 system, a high level at the microcontroller output turns on the transistor in the IGBT module.
    To use the system from the description, you need to negate the signals coming from the micro controller and then connect them to the inputs of the FSAM20SM60A module.
    The levels can also be negated in the program when configuring the PWM module of the micro controller. I understand that the intention of the question is to obtain information whether it is possible to use the micro controller from the article to control the IGBT module that is located on the board in the photo.
    As for the PFC module from the article, it is only up to about 400 W and it is difficult for me to say how much power the system in the photo needs to produce 7 kW of heat.
  • #30
    111lisu
    Level 16  
    So I want to replace the microcontroller that controls the FSAM20SM60A and the IGBT which is actually controlled by optocouplers, from what I see there are 2 different ones, I will take care of the rest with ESP, i.e. controlling the fan and gas switch. The compressor is 5CS130XCC03 R410 MATSUSHITA, i.e. 5.2kW of cooling power, a bit more heating power, but realistically it will be 2kW, something here: https://www.elektroda.pl/rtvforum/posting.php?mode=quote&p=10239503 they write about it but I can't find any specific pdf. Thanks for the information, I'm away from home today, I'll have to take a closer look at it. However, my knowledge about inverters is minimal, so I will be grateful for any information/suggestions.

    Simple 1/3 phase inverter with PFC

    Simple 1/3 phase inverter with PFC