How NOT to make a Modified Square Wave Inverter
HTML-код
- Опубликовано: 22 сен 2018
- Fast Prototype Option ($2 for 10 PCBs): jlcpcb.com
Previous video: • What is the best Rever...
Afrotechmods video: • ⚡ Don't build this 12V...
Facebook: / greatscottlab
Twitter: / greatscottlab
Support me for more videos: www.patreon.com/GreatScott?ty=h
You can get more information (schematic, code,....) about my inverter circuit here: / 21571279
In this video I will show you how to create a decent low voltage inverter circuit which outputs a modified square wave. Along the way though, I will also connect this low voltage signal to a transformer in order to bring it up to mains voltage levels. This way I will be showing you how unsafe and dangerous such a DIY mains voltage inverter circuit can be. Let's get started!
Thanks to JLCPCB for sponsoring this video
Visit jlcpcb.com to get professional PCBs for low prices
Music:
2011 Lookalike by Bartlebeats
Killing Time, Kevin MacLeod
(incompetech.com) 
Наука
You can use this circuit to make a sine wave inverter. Instead of driving the fet's with the low frequency square waves , use a higher frequency(say 20khz) pwm signal. Apply a half cycle sine modulated PWM signal to each side of the bridge. One side at a time obviously :0)
A small LC filter on the output of the transformer will give you a nice clean sine wave at 220V .
I've used same setup to build a 2KVA sine wave inverter. works great. Mine works at 48VDC and uses
more parallel fets , but the basic circuit. Not sure if the Arduino can do a fast enough pwm , but an STM32
board will handle it no problem!!
FYI , the diodes in the circuit are not really needed. The fets have inverse parallel diodes in them anyway.
Unless you are using schottky's , they won't be doing anything.
Keep up the great videos!!
Cheers
Rob
Indeed completely true. Anyone looking for advice should follow Rob's message.
Basically a form of class-D amplifier by the looks, that'd work real nice C:
I must try this sometime...
Exactly my thoughts - pwm could solve ALL the mentioned issues (but certainly would've required a more fundamental approach). This is the part, when GreatScott is not that great... He shouldn't have even touched the subject without gaining enough knowledge. Now he just bulshitted a whole lot of believers.
U smashed my lack of knowledge and money for a brand new inverter right in. Thank you.
Can u plzz provide the circuit diagram..
-Can you make an inverter?
-No
-...
Actually you can if you use SPWM and a little bit of know how.
welp
Denied 😂
Was that a challenge?
He's referencing a comment great Scott made to someone that he shows on the video
I have a 300W Sine wave inverter and its great. We get sporadic power outages here and with that thing i can power my fridge, charge my phone, power a light all from a old ebike battery (that i connected together for a massive ~12V 40 Ah. It gets charged by a 50W solar panel and is completely off grid. I used it 2 times and each time it was just brilliant to be the only person who had electricity. That smug feeling is worth it.
I know the feeling - my recent inverter purchase gave a similar feeling during a planned outage; and only i had 240VAC!
On a side note, be careful with that fridge; i'm amazed it runs off the inverter at all; usually the compressor can have a high start-up current (imagine compressor piston is at top dead center - a lot of current will flow to even get the thing moving) which generally makes most inverters sad...
It may work most of the time, but it may also knock out your inverter one day... Just a thought!
300W can power a fridge?
@@KashifJavedForex nahh i think 1000w inverter is good to be on the safe side (500 is fine)
10:48 "I calculated an efficiency of 84%" proceeds to write 83
he said of around 84% so he rounded down
Sent by ElectroBoom.
I'm an old EE and hobbyist/troubleshooter. This is a great site. Wish this was around in the 70's and 80's!
Subbed.
I love how you tell us the disadvantages of the circuits you build
Perfect circuit to power up light bulbs.
Hello your student tathagat
Maybe...
LMAO
More like perfect circuit to turn things into light bulbs
AD9833 + Opamp switching mosfet to produce pure sin wave
- "Can you pleeeease make an inverter? It would be the best thing in the world! Thank you!!!"
- *NO.*
true chad
Savage
Excellent tutorial! I remember watching AfroTechMods video a while back and it's good to hear you shout out to it!
I would like to suggest that you include the shorting function, where you turn on both of the bottom fets during the dead-time of the inverter.
This will greatly reduce transformer noise, make the transformer's output waveform look totally textbook - and improve the inverter's overall efficiency when driving purely inductive loads.
Anyway, great video, as always :) .
"Invest into a proper one with a lot of safety features"
Yeah... without any kind of proper reverse polarity protection :D
#ElectroBOOM
All they need is basic FULL BRIDGE RECTIFIER!! Its that simple.
Meme aside, rectifier is one of the worst kind of reverse polarity protection because now you have twice the loss compared to a single diode protection.
Labs did you say diode..? I think you ment p-channel mosfet for reverse polarity protection
Nope, that's really what I meant. Full bridge rectifier has four diodes, two of which will conduct in the same time depending on input polarity. Because you have 2 diodes in the circuit instead of 1, now you have double the voltage drop and double the loss.
You can use a Zener and some resistors to limit the gate voltage.
Did I forget to thank you your work? I find particularly interesting, that you share also "how not to do", and your somewhat uncommon way to solder 😊
Your tutorials help me to distract from the all too perfect world of now.
As you mentioned, turning on the bottom FETs during dead time will solve your spike problem. i have built exactly this type of inverter using a discrete transistor high/low side driver and NMOS FETs only. The efficiency is very good and the dead time conduction in the bottom transistors does not seem to impact eff. significantly. It powers all my home's lights plus TV and most small loads and never runs hot and has been super reliable.
Hello I'm trying to build one. Any help with schematics?
Great Scott I'm watching your videos from last two years and learnt many things from you thanks for your videos n ,keep doing your job and again thank you so much
Great video as always! You mentioned the pure sin inverter. Can you make a tutorial on how to create a low voltage sinwave from 12V?
It is on my to do list
LV is straightforward. Just use SPWM and then filter it with an LC filter.
@Lassi Kinnunen true. Will work very well for low power. But use a subwoofer amp for 50hz.
Low THD too
Lassi Kinnunen neat
So 12V sinewave?
Interesting, I was planning to do a similar job using an old transformer and an IGBT module given to me. Probably the biggest challenge (as shown) being to tame the switching spikes.I'll start by examining a few commercial designs & those presented in electronics magazines (Electronics Australia etc) for some ideas that worked for them.
Your almost at a million subs. Good luck GreatScott!
Great video! I’ve made simple modified square inverter using PIC, N-channel MOSFETs, and transformer I got from audio amp. I reduced humming by connecting capacitor to secondary. This inverter outputs 120V but when I connect some loads to it, then voltage drops to about 60V. I’ll try to make a full bridge circuit.
So it is you who makes all the hand-drawn schematics in Wikipedia! Great Job!
Your common pcb soldering is amazing.Your soldering is better than a printed circuit board😍😍😍😍
It's actually quite easy with a good adjustable temperature soldering iron - which doesn't have to be too expensive!
wat
Thanks dear, only you're my real electronics teacher, i love your all videos truly.
Congratulations for 900k subscribers...👏👏👏👏
I always look forward to your videos thank you so much !!!!
Pure sine wave inverters do get tricky to build. Earlier this year I made a 10kw one using a small Atmel AVR micro, some IGBT blocks and a bloody great power transformer. Tracking modulation with load, while maintaining waveform integrity gets hard. I took a rising 1/4 of a sine wave cycle and sliced it into 120 areas. Each area was further sliced into 20 more slots that I could increase or decrease in duration. 1 slot = minimum drive, 19 slots=maximum drive. This quarter sine-wave was mirrored to synthesize the second quarter of the cycle . The combination was then flipped again for the second half of the cycle. A voltage feedback circuit drove one of 70 different drive intensity levels based upon output load conditions, to keep the 230V stable. This all happened on a small 8-bit micro with a 16 MHz crystal programmed in assembly language. But it was bloody hard work to get it right, but the output waveform was really clean. My recommendation: buy somebody else's inverter and watch tv instead.
i am only 10 years old and you teach me a lot Greatscott thank you for your effort have a nice day
Im like you.
Such a good video. Damn, you rock Scott!
always learn a lot from your videos
I am your biggest fan and coincidentally I got the same project for my masters thesis I was just doing some research on this and you uploaded this video :)
I've been in this business for quite some time, and since I can use microprocessors, I generate the digital equivalent of a sine wave, which gives me a perfectly shaped curve and frequency stability. Depending on the microprocessor I use, I do not even need condensers and coils to flatten out some of the bumps. A sample rate of 4 kHz over 50 Hz (80 steps) appears prefect at the output.
Would you give me more details about microprocessor based inverter ? please .i am very interested about it .My email id nasimcisco@gmail.com .
Pretty darn good work, dude! Fantastic! 😃
But for the next one: good and expensive inverters, DIY or buy? 😂
Good presentation.. I am doing now a high freqency inverter, this helped me lot..
Thinking about using this circuit to drive a flyback transformer for my new high voltage supply. It might not be great for mains AC but it's still a great little circuit :) thanks scott
One of the great videos offered by you
Glad you liked it
I wish one day i could get entire knowledge on electronics just like you sir... as i am an electronics engineering student i have some hope on my wish...
its not that hard to learn but there is a lot to know, its rewarding career, there is heaps of electrical engineers and hobbyists on the Internet sharing their projects.
use google image search and you will find heaps of schematics for all sort of things very handy for beginners to learn.
Going to hit 1 Million sub...great going😜👍
Great vid as always!
Awesome video thx for sharing your knowledge
Man, I fully understand people's frustration when it comes to building an inverter! When I first started Learning Power Electronics in college, I thought I could do that, and I thought, Geez, it's so easy! But time went by and it was only when I was writing my undergraduate thesis (I was trying to build a switched mode power supply ) that I learned things like: "dead time", , "Switching Losses "," Harmonics "," Types of Electrical Loads ".... Unfortunaly this is not as simple as it sounds! There are many things that can instantly blow up your switches, blow up your controller. To do this, you will need to study a lot of things before, of course, this if you want to do things right without risking your life. GreatScott, again It was a great video, and thank you for alert, and protect people from this risks.
I know the pain! Most people don't understand the effect a too big a capacitor can have on filtering dc after a diode bridge: current as a dirac pulse, harmoics ftw. And thats simple electronic engeeniring. Now imagin trying to explain active filtering on powerlines and wave fronts.
Hey great Scott, you should build a diy air conditioner. It would be a fun project and be helpful in the summer. Thanks!
I put it on my to do list
Scott, the day you build a real air conditioner, not a swamp cooler or a cool plate, is the day you become a billionaire! Any different more affordable option than the expensive compressor/freon system we have now would totally change the world!
There are alternatives to gas phase change. But they have their own problems. Solid state is horribly inefficient. Reverse Stirling cycle is efficient, but physically very large.
Yes, peltier. And you can run a Sterling engine in reverse for cooling. It's a good way to get things really, really, really cold - you can condense oxygen and nitrogen from the air with one of those, but they can't actually move much heat. Then there is absorption refrigeration. They all have their niches, but when you want to cool a room or a building, gas phase change is the best we have right now.
The term "Freon" should really be avoided. It's a brand name, but the chemical it identifies has been changed several times over the years. The early ones are all long-ago banned by international agreement as CFCs, so the manufacturer switched to new chemicals but kept the same name. You need to use the right one for each AC unit, as the boiling point and optimal pressure differ.
Additional requirement: it must run of 24VDC for the DIY powerwall / off-grid enthusiasts 😁👍
Thanks, Scott. I wanted to make my own low-frequency inverter.
Love the video Great scott.
Congrats on 900.000k Subs ^^
Thanks
Such a precise circuit.
I love you Great Scott and appreciate all the videos it's incredibly inspiring. But it's a crime not to go over SPWM and RC/LC filters in this video. You gave us the H bridge beginner course when you know we can handle intermediate or advanced.
As always nice video!
Hi Scott!
Nice video and schematic, as usual! :)
Can you make a video on BMS circuits?
I'm particularly interested in active load balancing (charge shuttling or inductive)...
Keep up the good work!
That frankenstein AC wave lol. Good video.
Good video and information!
Thank you for this, you may have solved a problem I have with an inverter I picked up cheaply, I was trying to figure out what the programmable chip did, now I think I know, the chip is not functioning, and I have no means to program another, but if I can trace it's pin outs I might be able to replace it with an Arduino which I can program.
One of the problems with square waves is that they bring a ton of additional harmonic frequencies. Since harmonics are usually higher frequency, is it possible to round out the output using low-pass filters, or are the capacitor and inductor values to filter a cheap inverter too extreme?
It’s my understanding that the good inverters use VERY high freq PWM with duty cycles corresponding to the sine function. The higher frequency of the PWM allows them to be cleanly filtered out. You might be able to achieve this kind of inverter with this same circuit but a new program.
Yes that is correct. I made a circuit like this with the PWM signal you described. PWM signal can be created by an arduino and you use an LC filter at the end to filter out the PWM and make it a very clean sine wave
Not only that but they also input a lot of harmonics into the grid. Each country has its own laws but most countries limit home many harmonics you can put into the system. Also because the inverter may not be synced with the grid it can cause havoc with the power factor. An easy way to mitigate this is to use a boost converter and have it synced with the grid. And get it to shape the current to the grid voltage sin wave.
i think they indeed modulate the duty cycle - but the reason the frequency is very high is more often because they want to use ferrite material in their magnetics which have less magnetic loss and can be made smaller but which requires high frequency. I think that high frequency also helps reduce the values of components needed - and very often i think alot of harmonic noise simply carries along and stuff like chokes are needed to help reduce this. For that though i think scott would need to go to a (smaller) ferrite core as well.....but lower frequencies and that kind of modulation might work.
Ah yes you are right, wanting to go small you need to use high frequencies to stop the the ferrites from saturating. Which would cause havoc to be honest.
Basement ScienceE I don't understand why having big components is a problem, Scott isn't selling a product, he can shug in many caps and inductors as he pleases.
Usually the mosfet switching part of the invertor switches allready high voltage DC which was been stepped up and rectified by a switching power supply. The same voltage could also be turned in to an ac signal with some PWM magic and a LC filter to average out the high frequency PWM in to a nice sinewave. Using the arduino's analog inputs and a few optocouplers a simple feedback loop can be established to monitor the quality of the output.
I have made a pure sine wave inverter by making a step up power supply and using a class d amplifier with a 60 hz frequency generator... works perfectly...
I love your channel man! You should design an arbitrary waveform generator! Forget about the voltage conversion and explain the digital to analog conversation process.
Nice, thanks for sharing 😀👍
Hey love your videos . I would absolutely love to see you modify the h bridge to ha dle more power .. make a charge pump as part of your circuit .. or simply add more h bridge driver circuitry to double up on the h bridge to handle more power . Keep up the good work and great content as always
Very informative Video. Hey, I was wondering if we could also use H-Bridges like L293d or l298 instead of using 4 mosfets?
Awesome, well done!
You are the one who knows this business best.
I have been waiting for this video.
Nice video
Thanks, I love inverter or power electronics was hoping you would make a video on this area!
100% agree with your full bridge topology.
But could you make another few videos and improve on the design.
Part 2 - Adding output current and voltage feedback.
Part 3 - Adding a low pass LC filter to make a pure sine inverter.
Part 4 - Add input current / voltage feedback for use as MPPT tracking input (solar inverter).
This would be my dream series of videos and you have the know how!
Thumbs up if anyone else would enjoy.
Part 3 would include adding SPWM.
Great videos.... Love from india...❤❤❤❤
Useful video
sehr schön erklärt.
Smart guy. You have a beautiful brain.
Love your video
I'm very new to electronics but I'm kinda interested in getting into power electronics. My idea might be completely wrong and maybe won't work, but what I'm thinking is you would have a high frequency PWM like a normal DC buck converter would have. But instead of aiming for a set voltage, you are using a timer to calculate what the voltage should be based on the frequency you are going for, and then reduce/increase the PWM based on that value. You then reset timer for each sine pulse. You of course need a H bridge to get the negative voltages, and a SMPS transformer to get proper isolation. In a typical setup your battery negative will be your ground, so you'd want a centre tapped transformer and make the centre the ground. You then aim for 170 volts peak on the negative and positive side and the output would be similar to a 120/240v system. Essentially this would give you a crude pure sine inverter, at least I'm thinking. I could be way off here. It sounds like it would work but I have not tried it or seen anyone try it so maybe there's a good reason for that haha. Am I on the right path or completely off? I've read up on different ways of doing it too but they felt more complex. Suppose there's a good reason though.
FYI I use PSIM to simulate power electronics it's great!
Btw Great video.
There is also the issue with ripple current, which is hard on the capacitors in a lot of modern switchmode power supplies.
finally ! an inverter video ! wo wo :D
Nice Try! Keep uploading such a High Quality content videos where you can learn a lot of things!
Perfect! Now I'm *not* going to make such a Modified Square Wave Inverter! That's exactly what I did *not* want to make all the time.
Gracias, muy buen video
You have nearly 1milion subs!
I have several MSW inverters with dead H bridges and I don't even bother fixing them. I just wire the DC out from the inverter directly to the plug. Most electronics these days have switching power supplies and they don't care. And it is a little more efficient.
Hi i like your vid and i hope you much more successful in the future
great job
Hi sir, I'm working as electronics professor. Your effort for each project is great and awesome 👍 I have two doubts always preset in my mind when thinking about sinewave inverters 1: Why RC phase shift oscillator or any other sinewave oscillator is not commonly used as oscillator source in pure sinewave inverter? Is any problem with that??
2: MOSFET have Giga ohms resistance in its gate , is it necessary to add 10 Ohm resistor in its gate ?
Kindly reply 🤗🤗
Very informative. Btw any idea why the sound vanished after connected the load at output of transformer.
Exciting video, but I have 3 questions about:
1.Can a pure sine wave inverter be used in an electric car?
2.Can the mosfets of an inverter circuit be oriented in a "X" pattern?
3.Would your DIY inverter schematic work in a electric car? (Adding safety features like all the commercial ones have of course)
how can you fix the voltage shape problem ?? any suggestions ?
I was waiting this project hope you continue modifying it
With a little more complicated programming, an LC filter and a circuit closely resembling this one you can create a full sine wave converter. I made one in uni
I know. I will do such a modification in a future video.
1:43, Dang what a savage rejection!!
Am I the only one noticing a deep inhale breath before each sentence? I've noticed that in this video and now I can't unheard it.
100k more for 1M! congrats in advance
The pull up resistors for the P-channel MOSFETs have to be between *gate and source* and not *gate and drain* , am I right? 🤔
Add a snubber network, then you can suppress voltage surge while switching transition.
Cheers from Indonesia
Those circuit designers for pure sine wave inverter are brilliant
Thank you!
I electrocuted myself with a home made inverter during my apprenticeship. Used 555 as the timer, and forgot all about what I was doing and after powering on for the first time with the step transformer in place, proceeded to touch the HV side. It was one of those deciding moments that re-affirmed my decision not to be an electrician and stick to instrumentation!
Great way to show how an inverter works without giving people a reason to make one and then probably electrocute themselves
That's nice 👍👍👍👍👍
I did 3-phase pure sine inverter on my master thesis. It costs me and my friend over 1000 euros to design, build and run. And thousands of work hours... Kind regards!
using a -1+1:1 transformer and some caps you can turn the modified square into a more sinusoidal waveform
also using a different driver that is
What about adding capacitor to perform low pass filter and make its resonate at 50 Hz?
Nice video as before, can you send the link for obtaining the output stepup transformer
Fun fact: apart of producing the pure sine wave is not an easy task, most of the electrica appliances would perfecty operate on DC current. What are they:
- Ohmic resistor, like heating elements, incandescent light bulbs, etc.
- Machines with universal motors like drills, grinders, they operates much more smoothly on DC.
- All appliances with switching power supply, from mobile chargers to modern computer's power supply.
- Machines with BLDC motors like central heating water pump. They also have switching power supply.
The last two appliances' power supply's first task is make DC from AC. Thus they will work on DC just as well as on AC.
What will not work:
- Induction motors.
- AC transformers.
These lists are far from complete but most of the electrical appliances around the house will fall to the first group.
One major disadvantage of the DC current is the stubborn electrical arc in the switches what makes serious damage in it.
Nice video :)
But man, I think we need a fast diode for the reliefs... like the MBR360 or something. The 1N40xx series is way too generic/low freq. for that...
Very cool. As the negative DC potential remains the same all the time I guess there's no need to apply any PWM to the low side mosfets but only the high side ones (in unipolar inverter)? This way a return path for currents always exist, also doing what you did in your workaround. It sounds reasonable?
I looked closer at your circuit and noticed that the input voltage at the mosfets cannot be higher than 12 V (it says 60 V) because the high side driver is powered by 12 V.
Instead I used bootstrap drivers IR2104 as in the Tesla Coil project and it worked great. Now I can apply up to 600 V at the Drain.