Tahmid said:
Hello
Sorry for late reply but there was urgent repairs to do on my house.
I did take the double sides pcb out and I've seen that it is much difficult to figure out since there is kind of parallel wires mix with serials one, and there is others traces that I can't see who go between the pcb and the transfos and there is also some kind of sensor into the windings of the transfos.
So I will forget to change or double the transfos.
Thank a lot for trying to figure out the transfos types and winding.
Bye
Gaetan
zigzagflux said:
Hello
Not free shipping to canada and our money are now lower than US money.
I did find some kind of big diy PWM inverter on the Net, I will post the link wen I will have look a bit at it.
Thank
Bye
Gaetan
Hello
Here is a diy 2kw PWM true sine inverter I've found on the Net, it would need less filtering to clean the noises.
The only problem for me are the PIC16C84 MCU , I dont have any experiences or knowledges about those PIC and no hardware to program it.
I can see on this web page that he is using two ferroresonant transfos in parallel, that's ok since I have some big transfos here at home.
Here's the link;
http://www.technology.niagarac.on.ca/people/mcsele/i2k.htm
Anybody did seen or built that 2kw PWM sine wave inverter, any ideas or comments ?
Thank
Bye
Gaetan
Here is a diy 2kw PWM true sine inverter I've found on the Net, it would need less filtering to clean the noises.
The only problem for me are the PIC16C84 MCU , I dont have any experiences or knowledges about those PIC and no hardware to program it.
I can see on this web page that he is using two ferroresonant transfos in parallel, that's ok since I have some big transfos here at home.
Here's the link;
http://www.technology.niagarac.on.ca/people/mcsele/i2k.htm
Anybody did seen or built that 2kw PWM sine wave inverter, any ideas or comments ?
Thank
Bye
Gaetan
WHOA !!!
DO NOT MODIFY THE INVERTER !!!! It has been designed to operate properly as it is. If you "double" the power transistors then you will be doubling the load of the signals that drive those transistors. This will most likely overload them and the results will be worse than it is unmodified.
Trying to change the inverter's output is pointless. Inverters work by generating a 60 Hz square wave to approximate a 60 Hz sine wave. There is nothing you can do to make it more "sine-like".
WHY do you want to get a cleaner (more sine wave-like) output ??? If inverter noise is getting into your audio (analog) circuitry then you need to do one or more of the following:
1) Shield the audio circuitry;
2) Shield the audio signals going into and coming out of the audio circuitry;
3) Properly filter or regenerate the DC power supply voltages powering the audio circuitry;
DO NOT MODIFY THE INVERTER !!!! It has been designed to operate properly as it is. If you "double" the power transistors then you will be doubling the load of the signals that drive those transistors. This will most likely overload them and the results will be worse than it is unmodified.
Trying to change the inverter's output is pointless. Inverters work by generating a 60 Hz square wave to approximate a 60 Hz sine wave. There is nothing you can do to make it more "sine-like".
WHY do you want to get a cleaner (more sine wave-like) output ??? If inverter noise is getting into your audio (analog) circuitry then you need to do one or more of the following:
1) Shield the audio circuitry;
2) Shield the audio signals going into and coming out of the audio circuitry;
3) Properly filter or regenerate the DC power supply voltages powering the audio circuitry;
Hello
I will not modified the inverter, but it have mosfets output, so it is much less load to signals that drive those transistors.
Have you take a look at the link for a diy 2kw PWM true sine inverter I've found on the Net ?
http://www.technology.niagarac.on.ca/people/mcsele/i2k.htm
Any ideas about it ?
Thank
Bye
Gaetan
I will not modified the inverter, but it have mosfets output, so it is much less load to signals that drive those transistors.
Have you take a look at the link for a diy 2kw PWM true sine inverter I've found on the Net ?
http://www.technology.niagarac.on.ca/people/mcsele/i2k.htm
Any ideas about it ?
Thank
Bye
Gaetan
That inverter design is very similar to that found in some APC UPSes. The PIC drives two MOSFETs through a discrete driver circuit which drives a 60Hz transformer. A simple resistive voltage divider is used for voltage regulation. It is a partially bidirectional design. In normal operation, the transformer is connected to the input through a relay. The MOSFETs stay off in that mode and the circuit reconfigures into a charger. That design works nicely as long as you don't mind the weight.gaetan8888 said:
Many modern inverters including the one you have step up the voltage at a higher frequency and generate a 60Hz wave with a H bridge.
star882 said:
Hello
I would use my T-MOS mosfet, MTE215N10E or MTP60N06HD
I have few big transformers that I can use there. I presume that the guy use two transformers in parallel because he allready have them.
The only problem for me are the PIC16C84 MCU , I dont have any experiences or knowledges about those PIC and no hardware to program it.
Thank
Gaetan
Hello
For the i2K diy 2kw PWM true sine inverter from the link I've gived in my precedent message.
Anybody could program a PIC16C84 MCU ?
The guy say that the code for the inverter was written for an OLD version of MPASM
Thank you
Bye
Gaetan
For the i2K diy 2kw PWM true sine inverter from the link I've gived in my precedent message.
Anybody could program a PIC16C84 MCU ?
The guy say that the code for the inverter was written for an OLD version of MPASM
Thank you
Bye
Gaetan
gaetan8888 said:
Hi gaetan8888,
can u tell me the temperature of the two transformer at full 1000W load. Is it very very hot as the low side carry huge current (90amp!).....?
😕 😕
Hello roomi
Yes the transfo get hot but not too hot and there is a fan to cool it.
Btw, have you look at the diy 2kw PWM true sine inverter I've found on the Net; http://www.technology.niagarac.on.ca/people/mcsele/i2k.htm
it would need less filtering to clean the noises.
Do you know about PIC16C84 MCU programing, I dont have any experiences or knowledges or hardwares for PIC.
Thank
Bye
Gaetan
Yes the transfo get hot but not too hot and there is a fan to cool it.
Btw, have you look at the diy 2kw PWM true sine inverter I've found on the Net; http://www.technology.niagarac.on.ca/people/mcsele/i2k.htm
it would need less filtering to clean the noises.
Do you know about PIC16C84 MCU programing, I dont have any experiences or knowledges or hardwares for PIC.
Thank
Bye
Gaetan
gaetan8888 said:
actually, i have built a dc-dc converter (12vdc-350vdc) for making sine wave inverter. during load test at 350v dc the primary current (push-pull) was measured 33A and the transformer was very hot (abt 72degree celcius). I am worried abt what will happen when it takes full load current of 60 amp..!!!
my trnasformer is (3T+3T)/124T and i am using push-pull topology at 43Khz..... each primary winding consists of 7x21swg cu wire and secondary is of 1x21swg cu wire......
I don't have much pic experience. but u can find tons of inf on net for pic16f84.... but i do have some bascom avr experience........
Are you using a single conductor primary or a multiple conductor primary? Consider the AC skin effect of using a single large diameter conductor versus using mulitple smaller conductors to reduce copper losses. Litz wire is an option, but I perfer to using multiple conductors laid flat to avoid uneven winding with round bundles of conductors with a large diameter. Interleaving primary and secondary layers also helps a bit. A well laid out transformer will reduce inductance leakages which means small snubbers.
Also check for possible core saturation or near core saturation. Try adjusting the switching speed higher with an static output load to see if your running too near the point of core saturation. Plot out the current vs frequency using a static load to determine the optimum frequency for a given load. Consider that no core material has a perfectly square B-H loop.
One option if you use a microcontroller as a controller is to alter the switching speed as the load increases to avoid running to saturation issues. To make the design simpler, you can use several discrete frequencies instead of trying to develop a complex function that completely dynamically adjusts the frequency. ie for 0-250 Watts 43Khz, 251-500 48 Khz, 500-750 56 Khz, and so on). Using a Current sensor and an ADC input, you can adjust the switching speed to the appropriate frequency. You want to use low switching speeds at low loads to reduce switching losses, and higher switching speeds at high loads to reduce copper losses caused by near core saturation.
About the mcsele/i2k.htm diy 2kw PWM true sine inverter I've found on the Net.
I am not sure anymore it is a true sine wave generator:
From the i2k web page:
I am french speaking so sometime I have a bit difficulty to fully understand a english text.
By reading slowly his text, it is not very clear if it is a real sine output but it is seem to be a kind of modify sine, unfortunately the guy do not show the ouput signal on a oscilloscope.
If it is not a real sine, I will not build it.
Thank
Bye
Gaetan
I am not sure anymore it is a true sine wave generator:
From the i2k web page:
I am french speaking so sometime I have a bit difficulty to fully understand a english text.
By reading slowly his text, it is not very clear if it is a real sine output but it is seem to be a kind of modify sine, unfortunately the guy do not show the ouput signal on a oscilloscope.
If it is not a real sine, I will not build it.
Thank
Bye
Gaetan
TechGuy said:
Hi TechGuy,
a lot of thnks for valuable technique u described.........
i am using 7 number of wire of 21 swg together to make 3T + 3T as primary and a single 21 swg wire to make 124T as secondary (i am using push-pull topology).
i am confused whether the turn and size of wire selection of my transformer is right or not. i don't know the step by step method for building 45 khz switching transfrmr for inverter design. please help me in this regards............
gaetan8888 said:
by looking at the ASM and read the web link carefully my UNDERSTANDING about the I2K is, its a PWM sine wave , its close to a so called true sine wave but the harmonic content is a bit high then true sine wave inverter found in marketplace, I2K is designed for motor stuff and i guess motor dont mind some harmonics, there is lot of room for improvement like use a faster PIC with hardware PWM and there is alternative solution as well dont need to hack ASM and PIC stuff,
http://www.s4wsbox.com/?q=node/31
roomi said:
i am confused whether the turn and size of wire selection of my transformer is right or not. i don't know the step by step method for building 45 khz switching transfrmr for inverter design. please help me in this regards............
what type of core you are using, have you measure the cross section of the core,
areza said:
Hello
Reading this text, I was not sure it is a true sine wave generator:
From the i2k web page:
And for the inverter from the link, http://www.s4wsbox.com/?q=node/31 , it need 10x 12 volts batteries in serial, since it do not use an output transformer.
Thank
Gaetan
areza said:
I don't understand the type of core (i don't know what is ETD, ETD49, etc...Please help me in this regrads, I am newer in power elec.). The core area is 1.74 sq c.m.
That seems too small of a core for 45 khz @ 33A. Its probably saturating. If the core is a ferrite, you can probably drive it at a much higher frequency to avoid saturation. But the small core size will limit its maximum output (a guess about 750 Watts, but I could be wrong).
The issue I see is that its using fixed output inductors (200uh*2) which will limit its output power to about 250 Watts. The advantage of using an output transformer is that it has variable inductance based upon the output load. There are some other issues too. This looks more like a Modified squarewave inverter using a pair of output inductors and a cap to smooth out the square edges. It you put any load on it, the output waveform will become greatly distorted. It looks like the oscillation is running at 120 hz, and is using a 555 Timer to integrate the triangle wave into a squarewave that averages out to sinewave output.
Its not really true PWM sine wave controller. In a true-sine or pure-sine the switching frequency is much higher than 60 hz. A PWM sinewave output should look like this:
- Status
- Not open for further replies.