Sine wave power inverter

[T.Jackson]

hi

can you tell me about pic basic?

Manish

In a nut shell - PIC Basic is a true compiled, mature, robust, quick & easy no fuss solution for all your PIC-based embedded system needs.

In fact, so easy-to-use, if you can't write a program in it, programming
just isn't for you. It just couldn't get any simpler.

Trent Jackson

[MKA341]

where can i get pic basic software?

[keithdoxey]

where can i get pic basic software?

http://www.melabs.com/purchase.htm

[MKA341]

your design is very good, can you give me some idea about making a inverter by using pic basic?

Hi PICers,

I want to design a 24V DC to 50Hz/240AC power inverter. My choice of PIC is the 18F452 (HWPWM). But I am not sure whether the ADC conversion time would be short enough for dynamic correction.

Any thoughts

Regards

Sougata
The Andig Technologies

[sougata]

Hi,

Manish originally the inverter was built ono the PIC18F452 and had loads of features like true synchronized changeover using MOSFET based static switches to perform like a pseudo online UPS. Actually it was an adaptation of my motor control system using 18F4331. However the BOM cost was high so it was scaled dowm to mimic other designs already based on 16F72. Uses a pseudo software PWM. (Utilizes the inbuilt unit in compare mode to generate int and fire mosfets). Half-Bridge driving and a half cycle boost PFC for charging control. Vertical-Horizontal current based lookup table to get precalculated points (16F series do not support harware multiply). Most of the credit goes to Mr. Pramod Sinha of Jetmark Systems Delhi cause my 16F skill sucks. We could not share the code as it is a commercial project and we are legally binded. For clarifications and solving issues this forum is a great place and people (far far better than myself, Perhaps) would surely help you out. Best of luck. Where are you located in Delhi, I drop in there almost every month and stay in A6 Paschim Vihar.

[pramodsinha]

I have written a sine wave generating routine in pbp. It is tested.but i am too busy because summer is my business season give me few days to edit and post it to forum.
thanks everybody.

[mincing]

dear Sougata,

i am working on similar pic based inverter too.Could you please tell me which topology are you using for inveter bridge.

I am working on H-bridge.Have difficulty in Identifiying the brdge drivers.

It would be great if you could help me on transformer desin.

[mincing]

Dear Souhgata,

Thank you for giving so much of info.What was the ratings of UPS you have made.and what were the efficiencies.
Could you please help with Transforner design information.

Also I have seen a round core trnsformer some ............

also if you are using full bridge topology how do you drive the bridge..........

If you could help with LLC filter inductor design and transformer ,that would be really great.

Thanks in advance

[sougata]

Hi Mincing,

I was too busy in a project and really coudn't get the time to reply to the thread.

1. For the driver part I use TLP250 from Toshiba. Get the datasheet and everything should be easy.

2. For the bridge you would be using all N-Channel Mosfet perhaps cause they are readily available and have much lower Rds On compared to P-Channel. However switching the high side MOSFETs are a little tricky. You need to have a charge-pump circuit to switch them on. Try the IR2101 it makes your design much easy.

3. To maintain low distortion in the output use a flux reset method when changing cycle. You do this by turning of the high-side mosfet, then turning on the other low side mosfet (both low side mosfets on) and then turning off the one which is not required for the current cycle and finally starting the high side driver of the corresponding cycle.

4. Please note that keep your high-side MOSFETs turned on during the entire cycle and apply PWM to the low side MOSFET. Initially I tried steering the PWMs to all of the four MOSFETs to make up for the conduction switching loss. But it did not produce satisfactory results.

5. Now for the transformer design. Say you are operating on a 12 volts battery. Now when the battery is full charged it is upto 13.8 volts nominal and before the deep-discharge of at around 10.8 volts. You can reduce the voltage with PWM but cannot increase it with the current scheme (boosting can be done, theoritically using a boost PFC like technique but I failed to do so). So keep your transformer primary at around 7.5 to 8 volts. This accomodates conduction loss, battery low, and so on.

6. Now the problem is that the same transformer is being used for charging and you need to boost the voltage during charging. Well a PFC boost method can be applied. See the attached PDF. Note the high side rectifiers are your high side mosfet's body diode.

I am attaching a few more documents links which can be handy during the design process. Sorry for Rapidshare. The documents are more than the allowed 200kb and a multipart zip/rar would have been around 10 parts.

Here are the document links:

http://rapidshare.com/files/34936881...r_with_PFC.pdf

http://rapidshare.com/files/34936883...ter_Design.pdf

http://rapidshare.com/files/34936882...c-inverter.pdf



I am sorry I may not be able to provide active support on this thread due to time limitation but will always try too within my knowledge and scope. (I am a completely self-taught with little or no academic background, So I might not be accurate all the times and ready to learn from others.)

I may start up a project on high frequency double-conversion sine wave inverter that utilizes Ferrite Core Transformers as PsDayama suggested. However it would be based on the dsPIC 30F2020. I still don't know whether I would be able to take up that project. Cause I have very little time left for myself catering to clients.

Anyways its good to see that the thread is alive again and would refresh my knowledge base.

[mincing]

Thanks Sougatha!

However I am building SI with Full bridge and isolation transformer.

My Dc input to bridge is 48V DC.

I am using optically isolated gate drivers.I need a 4 output power supply with isolation.

I am looking for suggestions.

[sougata]

Hi,

I mentioned TLP250 for the gate drivers. They have an optically isolated input and a totem pole output. It works great with IGBTs too. While your DC input is 48 volts be carefull about choosing the MOSFETs . Generally MOSFETs with greater breakover voltage have higher RdsOn. While it may seem logical to use a 100volts MOSFETs for the design. Please note that hard turn off of the MOSFETs connected to inductive load causes ringing in the drain as well as gate. So surges may kill them. Snubbers are recommended and fully avalanche rated MOSFETs should be used. If you keep your PWM frequency lower that will result in lower switching losses however may produce audible noise.

Best of luck in your design.

[mincing]

I have selected Mosfets with very low Rds on about 5mOhmm and 100V, 100A devices.


I also have optically isolated drivers from Avago ,but how about the power supply to these devices on output side i.e.Totempole and ground.I understand that a four out put supply with 18V and current? is required.
Is there any other means or method to escape this!

If you could suggest that Power supply cky that will be great.

[mincing]

Hi I browsed through the forum.
Which JPG were you talking about!

Hello!
just scroll down and U will find the schematic in jpg format.
I don't know if Saugata will give the code.
Some part he has already explained.

[sougata]

Hi Mincing,

I would like to know which Avago(Agillent ?? ) drivers you are using and which MOSFETs. What is the intended configuration and if it is being directly connected to load or by a transformer.

Please give the required spec of your design in details.

[mincing]

i am using ACNW3130.seems to be of right choise.Welcome any comments.

I am using a full bridge configuration.with isolation transformer at the out put.Bridge would feed transformer.

[sougata]

Hi,

Please post at least a block diagram if you are not through with your schematic. Also it would be helpful to know exactly what are the specifications of your system or what you are trying to achieve. Otherwise it is hard to suggest a scheme.

[Charles Linquis]

Many years ago I made a 400W . 120V *PURE* sine wave inverter using a PIC. The input voltage was 48. I used a normal inverter (with a ferrite transformer) to give me a regulated 185VDC to feed the PIC-based H-bridge sine wave section.

I used high-side drivers and FETS on the "top" two legs of the bridge and ordinary (low side) drivers and FETs on the lower two.

The high-side drivers can't be "PWM-ed". So I alternately turned them on (one for each half-cycle) while using 78Khz PWM on the lower two. Simple CMOS gates prevented "punch-through" by not allowing the upper and lower FETs on the same side to conduct simultaneously. A small (47uH) inductor was in series with the load and a .1uF was across the load. A scope showed the output to be a perfect sine wave with no glitches whatsoever.

The PWM value was from a lookup table (the program was written in assembly).

The switching FETs needed only a small heat sink.

[sougata]

4. Please note that keep your high-side MOSFETs turned on during the entire cycle and apply PWM to the low side MOSFET. Initially I tried steering the PWMs to all of the four MOSFETs to make up for the conduction switching loss. But it did not produce satisfactory results.

Hi Charles,

Your system was essentially a double conversion system. Although it has two power stages but the sinewave regulation becomes real easy when you have your DC bus regulated. I had my original design of the DC boost using a SG3525 followed by a PIC driving MOSFETs to produce the sinewave. Now I am trying to utilize a dsPIC (I am in the process of learning it and still a Newbie ) to handle both the conversions as well as battery charging on mains availability. My knowledge in SMPS is very limited till date and I am working hard for those "gotchas". So wish me luck. (I need it for keeping the number of toasted MOSFET/Chopper low, obtained during trial phase)

[mincing]

Many years ago I made a 400W . 120V *PURE* sine wave inverter using a PIC. The input voltage was 48. I used a normal inverter (with a ferrite transformer) to give me a regulated 185VDC to feed the PIC-based H-bridge sine wave section.

I used high-side drivers and FETS on the "top" two legs of the bridge and ordinary (low side) drivers and FETs on the lower two.

The high-side drivers can't be "PWM-ed". So I alternately turned them on (one for each half-cycle) while using 78Khz PWM on the lower two. Simple CMOS gates prevented "punch-through" by not allowing the upper and lower FETs on the same side to conduct simultaneously. A small (47uH) inductor was in series with the load and a .1uF was across the load. A scope showed the output to be a perfect sine wave with no glitches whatsoever.

The PWM value was from a lookup table (the program was written in assembly).

The switching FETs needed only a small heat sink.

Can I get to see the schematic.Idea is quite interesting.In your case how big would be the heat sinks on High side FETs.

Also I would like to know how to post block diagram/schematic in to the forum,so that I can put my design to give a clear understanding.

[Navneet Vinaik]

Hi Charles,

Now I am trying to utilize a dsPIC (I am in the process of learning it and still a Newbie ) to handle both the conversions as well as battery charging on mains availability. My knowledge in SMPS is very limited till date and I am working hard for those "gotchas". So wish me luck. (I need it for keeping the number of toasted MOSFET/Chopper low, obtained during trial phase)

Dear Sougata,

I have a short experience in this domain but have learnt quite a few things. I have developed a UPS on 16F676, square wave inverters on 16F72 and sinewave inverter again on 16F72 based on H bridge..... all in the past one year. Now, I am keen to work on ferrite core...so if you can guide me some thing..please help....I want to grow in this field, presently am working on SMPS designing(low power, upto 200W only)....the fire is there in me, everyone please guide....

[mincing]

Navneet Vinaik,

Dont worry!This field is more of a stabilized now.You will get every kind of information you need.

let me know what exactly you need.If not able to answer directly,I can give you links and sourcs.

[Rishi]

Hello friends I am power electronics enggineer having experience of 7 years plus.
presently i am working on double conversion inverters.

[mat janssen]

Hallo Rishi. Can you post the schema and the code for this? All of us will learn about this subject!

[Navneet Vinaik]

Dear Mincing,

Thanks for your reply, and sorry for the delay.

I want to understand the technicalities involved in designing systems on ferrites. To be precise, what would be the kind of drive systems, snubbers, clamps and most important transformer design in case of ferrite based systems.
I would be grateful if you could provide me some details etc.

Regards

Navneet Vinaik

[Rishi]

Hallo Rishi. Can you post the schema and the code for this? All of us will learn about this subject!

hello mat

Yes i will send u details but it will take some time.

i have design 12v/24v dc to 350v dc converter with analog ic sg3525 and already
achieve effeciency more then 90%.

[sougata]

Hi everybody,

I have been greatly helped by this site http://schmidt-walter.fbe.fh-darmsta..._e/smps_e.html . It is the english version of the original german site by the author. Great work, great javascript based smps calculator.

I hope things become easier right now.

Regular bread and butter (hard to earn in my city ) jobs leave me almost no time to work on this project. But seeing this thread still kicking does give me some inspiration.

Thanks to everybody who contributed to this thread.

[BrianT]

Take a look at Don Lancaster's site at www.tinaja.com. Search for MAGIC SINEWAVES and you will find information on how to build very low harmonic sinewaves.

HTH

Brian

[mincing]

Hi friends,

I have difficulty in seeing the bridge driven through with sinePWM.

I have built everything but the bridge is getting shorted when I try to run the inverter.

Any Ideas how to achieve dead time control in Software.

Sougata are you active?

[sougata]

Hi,

I am hardly getting time to get into the forum and I know I am missing something or the other.

Mincing,

Try to use a resistive load at the output of the bridge and inspect the i/o signals with a scope. Sometimes improver gate drive and/or switch off spike can kill MOSFETs. Please post the schematic and code snippets I will try to help if it is within my scope/knowledge. I am not as active now but you can count on every other member on this forum. Far far better than me.

If you find that you cannot reveal the code/schematic due to commercial problems you can pm me. Again I would try, no commitments.

[mincing]

Thanks for the reply!

I understood that the current drawn(peak current ) is actually higher than what is seen on front panel meter of the Power supply I used.

Do you have any calc. for peak currents that flow in to the bridge?

Actual current or Dc current is only about 1.5 to 2Amps but peak is over 5Amps. at about 15 to 20 V on no load with a transformer connected.



Also what happened to your Double conversion Inverter. I have some documents ,I like to forward to you on the subject.
I am not sure if I can post them here, give me your mail ID I 'll forward the same.

Any one who wants to build HF inverter.

Thanks again!

[mincing]

Dear Mincing,

Thanks for your reply, and sorry for the delay.

I want to understand the technicalities involved in designing systems on ferrites. To be precise, what would be the kind of drive systems, snubbers, clamps and most important transformer design in case of ferrite based systems.
I would be grateful if you could provide me some details etc.

Regards

Navneet Vinaik

could you Pl. give me your mail ID ,I will forward some documents!

[Navneet Vinaik]

Dear Mincing,

Thanks a lot for your helping hand...
My email ID is [email protected]; [email protected]

I am pretty confident that I can build a good quality HF inverter, presently busy in getting all the tools ready...I have an old oscilloscope which has served us for the past 45 years but for this technology I am arranging resources to get a DSO among others tools.
Please send me whatever help that you can, I would be grateful to you as it will help be do my homework before I get into the battlefield!!!

[mincing]

Hi sougata,

Hope everything is fine !

I have difficulty in reading the feed back and adjusting the PWM.
Presently I am reading a DC proportional to the output voltage.PWM is continuously adjusted and checked with the feed back.
However the moment ckt enter the closed loop out put becomes unstable.It keeps changing continuously.

Could you please throw some light on this./

Thanks,

[sougata]

Hi,

I think I have not come across any schematics, code snippets. I don't even know which PIC at what Fosc you are using. May be I missed something as I am not regularly visiting this forum. Details...Please.

[mincing]

Dera Sougata,

I have attached a simple block diagram of Inverter I am working on.

Hope this would give an Idea of my circuit.

Kindly suggest what should be feed back routine.
We are using C with MPLab.

Thanks and waiting for your reply

[sougata]

Hi,

I am not an expert on C (uNsUrE of my expertise of PBP aswell ) so may be cannot help much with the code (which has not been posted ) but here are some tips.

According to your schematic:

1. Feedback is from one cycle only (not recommended)
2. Driving through inverters means both MOSFETs doing PWM (wasting energy)
3. Use Flux reset when changing phase
4. Be sure that the computed PWM is not eating up cycles and resulting in uneven spread of the modulating PWM

Please report what happens when no feedback is used. Use a lower duty cycle lookup table to ensure that your MOSFETs are not fried due to full drive. Or use a lower input supply to the Bridge.

[Navneet Vinaik]

Hi,

I am not an expert on C (uNsUrE of my expertise of PBP aswell ) so may be cannot help much with the code (which has not been posted ) but here are some tips.

According to your schematic:

3. Use Flux reset when changing phase
.

Dear Sougata,

What is flux reset when changing phase, can you please give me some more information about it.

Regards,

Navneet

[mincing]

I was just trying to understand this concept of flux imbalance control.

However I could not make out how we do it practically.Do we have to implement it in software or hardware.

I observed lot of noise while the current wave drops tozero and below.NO load voltage waveform is a correct sinewave.

Kindly tell me how to implement the flux balance control.

regards,

Hi,

I am not an expert on C (uNsUrE of my expertise of PBP aswell ) so may be cannot help much with the code (which has not been posted ) but here are some tips.

According to your schematic:

1. Feedback is from one cycle only (not recommended)
2. Driving through inverters means both MOSFETs doing PWM (wasting energy)
3. Use Flux reset when changing phase
4. Be sure that the computed PWM is not eating up cycles and resulting in uneven spread of the modulating PWM

Please report what happens when no feedback is used. Use a lower duty cycle lookup table to ensure that your MOSFETs are not fried due to full drive. Or use a lower input supply to the Bridge.

[mincing]

When feed back is not used the out put is pure sinewave.
."

However I could not make out how we do it practically.Do we have to implement it in software or hardware.

I observed lot of noise while the current wave drops tozero and below.NO load voltage waveform is a correct sinewave.

I would also like to understand " I was just trying to understand this concept of flux imbalance control

Kindly tell me how to implement the flux balance control.

regards,

Hi,

I am not an expert on C (uNsUrE of my expertise of PBP aswell ) so may be cannot help much with the code (which has not been posted ) but here are some tips.

According to your schematic:

1. Feedback is from one cycle only (not recommended)
2. Driving through inverters means both MOSFETs doing PWM (wasting energy)
3. Use Flux reset when changing phase
4. Be sure that the computed PWM is not eating up cycles and resulting in uneven spread of the modulating PWM

Please report what happens when no feedback is used. Use a lower duty cycle lookup table to ensure that your MOSFETs are not fried due to full drive. Or use a lower input supply to the Bridge.

[sougata]

Hi,

I am not qualified enough to explain. So I would suggest you to go through these links.

http://techtrain.microchip.com/webse...MPS_090106.pdf

http://www.intersil.com/data/an/AN1246.pdf

In my software I do this while changing phase.

1. Switch off the High Side MOSFET
2. Wait for the dead time.
3. Switch on both low side (flux reset)
4. Switch off both low side
5. Switch on the other High Side
6. Start PWM on the corresponding low side.

If you are not getting sinewave in your output while you apply feedback then its your correction algo that is not working. I cannot release our code here without permission from pramod but there is certainly an example of a Sinewave Inverter using C which uses a delta lookup scheme to make computations easier. Here is the link

http://rapidshare.com/files/43362188/C_Sinewave.rar

Hope this helps.