Frequency converter for single-phase motor

Hi-tech electronic and computer equipment and Internet. Better use of electricity, help with the work and specifications, equipment selection. Presentations fixtures and plans. Waves and electromagnetic pollution.
User avatar
Forhorse
Econologue expert
Econologue expert
posts: 2485
Registration: 27/10/09, 08:19
Location: Perche Ornais
x 359

Frequency converter for single-phase motor




by Forhorse » 07/09/10, 10:59

At the chance of a surf last night I came across a mircrochip application note that interested me a lot.

It is the plans and the firmware to build a small frequency converter (speed controller) for single-phase motor.
At first glance the components are findable in France and the realization does not seem complicated too much.
In addition to the advantage of being able to easily vary the speed of a single motor this assembly also removes the conventional capacitor that serves to turn it and is very often a cause of failure.
From what I understand a small modification of the firmware allows to power a three-phase motor.
It is limited to a maximum of 2.2Kw engine, but it's already quite good and I think that it may interest some do-it-yourselfers.

I do not have the time currently to look more at this project, I keep it warm for later. But I wanted to share with you this little discovery.
https://www.econologie.com/fichiers/partager2/1283849662vKhTYN.pdf

This application note and firmware source codes can be found on the website of microchip

As soon as I have time I look seriously on the project because the construction of a small frequency converter has always interested me. And why not propose here a small kit with components and circuit board to those who would like to build a kind of dimmer but do not necessarily have the knowledge or the material to do it even.
0 x
User avatar
quartz
I posted 500 messages!
I posted 500 messages!
posts: 640
Registration: 20/01/07, 14:38
Location: Parisian region




by quartz » 07/09/10, 11:28

Hello everybody :D

Just a little warning about the system,
In the majority of cases asynchronous single-phase capacitor motors do not have two identical coils.
The use of a capacitor on one of the two phases makes it possible to give, plus an indication of direction of rotation at startup rather than a real torque.
The maximum current on the phase supplied by the capacitor is practically half that of the main phase,
to a cow, it depends on the engines and their uses.

Regarding the phase capacitor it will actually be removed,
on the other hand you will need smoothing capacitors after recovery of the sector.
Not cheap either, on average it takes about 500μF for a nipple of power.
Rounded to 470μF which is a normalized value, all under 400 to 450V, inform you of the costs.

That said the idea is still relevant, thanks for the datasheet that is well done.

A+

PS: Page 15 bottom right of the document, there is a small diagram that shows the two phases one is Main Aux, so they take into account it is a good point.
Last edited by quartz the 07 / 09 / 10, 11: 37, 1 edited once.
0 x
Every minute that passes is an opportunity to change the course of your life.
YouTube page, Dailymotion page, Picasa album, personal Pages, Viktor Schauberger
User avatar
chatelot16
Econologue expert
Econologue expert
posts: 6960
Registration: 11/11/07, 17:33
Location: Angouleme
x 264




by chatelot16 » 07/09/10, 11:36

interesting, but there is no information on the power transistor and even less on the main: interface between peak and transistor

the 3 transistor bottom control is easy: just amplify a little

for the top transistor it takes a circuit of isolation because it strolls completely in tension: for me it will be 3 transformer

Some prefer to completely isolate the peak: it demands 6 transistor control transformer and also isolated system for measuring voltage and current

I prefer to put the mass of the peak at the mass of the transistor bridge


this note of application is interesting but does not really give the principal
0 x
User avatar
quartz
I posted 500 messages!
I posted 500 messages!
posts: 640
Registration: 20/01/07, 14:38
Location: Parisian region




by quartz » 07/09/10, 11:58

OUPS !!
Last edited by quartz the 07 / 09 / 10, 12: 07, 1 edited once.
0 x
Every minute that passes is an opportunity to change the course of your life.

YouTube page, Dailymotion page, Picasa album, personal Pages, Viktor Schauberger
User avatar
quartz
I posted 500 messages!
I posted 500 messages!
posts: 640
Registration: 20/01/07, 14:38
Location: Parisian region




by quartz » 07/09/10, 12:06

For this type of application the best is to use IGBTs, control by optocouplers.
An important info about the optocoupler is that their driving mode must be by short circuit of the input diode,
I explain myself in a low voltage application, it is enough to inject a current into the input diode of the opto, it is controlled at the output,
then we cut the input current and it goes off.
When one is in high voltage and high current applications,
Lord of the setting off of the opto there occurs a train of waves producing harmonics of quite important commutations.
These harmonic are sufficiently powerful to put the optocoupler back in conduction by internally capacitive coupling of the IC.
And here is the drama ..... the two totors lead at the same time, so courjus.
The trick is to bypass the input diode to put the opto off.
see here on page 13 you will understand better, figure 22 shows you what not to do.
Figures 21 or 23, on the other hand, make it possible to stop any incipient voltages at the terminals of the input diode,
naturally extinguished by the command which is a short circuit.

A+
0 x
Every minute that passes is an opportunity to change the course of your life.

YouTube page, Dailymotion page, Picasa album, personal Pages, Viktor Schauberger
User avatar
chatelot16
Econologue expert
Econologue expert
posts: 6960
Registration: 11/11/07, 17:33
Location: Angouleme
x 264




by chatelot16 » 07/09/10, 12:38

optocouplers transmit information but not power

it is therefore necessary also for each transistor an isolated supply circuit to power the amp that follows the photocoupler

that's why I prefer the good old transformer, who do all the work reliably ... well what makes it easier is that I do my winding, so it allows me to do exactly that that I want

for low power, I prefer the real mos fet: with a rds it is weak enough we have less voltage drop than the igbt and it avoids putting large radiators

Another note: in my depanage of industrial equipment I see that the photocoupler are weak point often defective: the impulse transfo are eternal ...
0 x
User avatar
Forhorse
Econologue expert
Econologue expert
posts: 2485
Registration: 27/10/09, 08:19
Location: Perche Ornais
x 359




by Forhorse » 07/09/10, 12:42

Regulate the scheme better ... :?
The power transistors used are IGBTs in the form of a three-phase bridge and its control logic.
IRAMS10UP60A (35 € HT at farnell, 26 € at Radiospares)
There is really no point in doing either, everything is indicated.
0 x
User avatar
chatelot16
Econologue expert
Econologue expert
posts: 6960
Registration: 11/11/07, 17:33
Location: Angouleme
x 264




by chatelot16 » 07/09/10, 12:49

quartz tone optocoupler contains the amp to directly control the igbt

I prefer to separate the functions: a simple optocoupler to isolate and some transistors to amplify
Last edited by chatelot16 the 07 / 09 / 10, 13: 52, 1 edited once.
0 x
User avatar
quartz
I posted 500 messages!
I posted 500 messages!
posts: 640
Registration: 20/01/07, 14:38
Location: Parisian region




by quartz » 07/09/10, 13:07

Personally I work with optos that 2A spike to order 70A IGBT 600V and it works for a long time.
The datasheet which I put the link shows an opto driver of this type precisely.
That said the transformer works very well as well, may be less easy to find for the individual.

The mosfets are historically more fragile than the IGBTs it is besides of the IGBTs which are used at the moment in the railway of last generation.
The voltage drop across the terminals is getting smaller and the dissipation is very low.
The frequency changer up to 5KW have no heatsinks.

The mosfets are becoming stronger with rdsON in free fall.
No heatsink also up to relatively high power.
After, everything is a matter of cost and availability of components.
but the technologies are worth it.

A+
0 x
Every minute that passes is an opportunity to change the course of your life.

YouTube page, Dailymotion page, Picasa album, personal Pages, Viktor Schauberger
User avatar
chatelot16
Econologue expert
Econologue expert
posts: 6960
Registration: 11/11/07, 17:33
Location: Angouleme
x 264




by chatelot16 » 07/09/10, 13:10

I did not understand this: everything is in the same module isolation, command, and 6 IGBT

how awful ! how do we go about troubleshooting this?

I refuse to mix power and control ... of course 26 euro at radiospare, it spoils the job a little, while separate component will come back maybe at the same price, but 20 in the circuit a separate component will still be reparable, while the module will not be found

I regularly deprive thyristor dimmer built in 1975: we can always put compatible component ...

and there is no need for 2kw: with separate components we can choose cheaper mosfet for low power
0 x

 


  • Similar topics
    Replies
    views
    Last message

Back to "Electricity, electronics and computers: Hi-tech, Internet, DIY, lighting, materials, and new"

Who is online ?

Users browsing this forum : No registered users and 163 guests