 Notice. New forum software under development. It's going to miss a few functions and look a bit ugly for a while, but I'm working on it full time now as the old forum was too unstable. Couple days, all good. If you notice any issues, please contact me. |
Forum Index : Electronics : Summary / list of members' inverter projects?
Page 1 of 2   |
|||||
| Author | Message | ||||
| flyingfishfinger Senior Member  Joined: 12/09/2020 Location: United StatesPosts: 110 |
Hi all- I've been reading this forum with great interest for the past couple of weeks. I'm heading up to our lot next week & hope to lay a pipe & measure pressure for my F&P hydro generator, but in the meantime I'm doing some pre-planning for the (near) future. In particular, I will need to build / acquire a decent inverter in the next couple of months (initially I thought I could wait until next year, but some recent developments in terms of setting up infrastructure may have accelerated this timeline). So, being a new member here - I'm having trouble sorting out what's what and who designed it, so I'm hoping folks can help me get an overview of inverter projects. Having read a bit about each of the following projects, it's not clear to me where they originate, what each is good for, the relation between them and if there's "THE" thread for each, and whether there are others. - Madinverter - Warpinverter - Nanoverter -...? Any guidance to the right pages / topics would be appreciated. If it helps narrow down the choice: Right now I have a 12V system, but a 24V system is probably better & the switch is easy - there's zero infrastructure out there so far. A 2000-3000W system seems like a good target, the long-term plan is to put up a small one-room type cabin. This is my family's first off-grid project, so I've got a lot to learn. I'm an EE by training but I spend most of my time at the micro & nanoamp level, thus these high-power projects are (sort of) new to me. I used to be on a solar car team here in California though, and we did the WSC in Australia in 2011 which gives me some familiarity with PV-related stuff. Cheers, Rafael Edited 2021-05-15 06:58 by flyingfishfinger |
||||
| Warpspeed Guru  Joined: 09/08/2007 Location: AustraliaPosts: 4406 |
Hi Rafael, There is a very long and rich history associated with home brewing inverters, and over the years, many different Forum members have made large contributions to different aspects of the problem, of how to make a sturdy, reliable, efficient inverter at an affordable cost. All this involves a great many different people, far too many to mention individually, but here is a very brief and incomplete synopsis. The pioneer and real hero is Oztules (John Tulloch) who started building and perfecting the original Oz inverter. This combined a low cost commercial Chinese driver board with a very large home wound toroidal transformer. This has started an evolving trend, which has continued on over the years. Since the earliest Oz inverters, a great deal has been learned, and improvements gradually made. The main areas of change are in the design of the magnetics, particularly the use of a series choke, which has significantly lowered the idling power. The original Chinese EG driver board was not without its quirks and problems, and Madness (Gary) came up with a circuit that solved many of the problem of continuing mystery blow ups. The next very significant step forward was the development of using a basic Nano computer board instead of the usual Chinese driver boards. Open source software development for that has come a very long way with increased functionality and reliability, and is where we are right now. Its now fairly straightforward to wind your own toroid and choke, and use the Madness gate driver circuit along with a Nano driver board to make it all go. There are still some unsolved problems trying to drive very large numbers of paralleled mosfets, getting them all to switch simultaneously and to share the load. So PWM inverters up to about 4Kw to 5Kw are pretty easy to get going. Above that power level it becomes increasingly difficult to get the type of reliability we require. It has been done successfully many times, but there have also been a lot of very frustrating failed high power projects. In other words, physical layout and gate driver circuitry becomes far more critical as the power level increases with a larger number of parallel mosfets. Development is continuing, but the basic problem still remains, high frequency PWM does not scale up well at very high power levels. The Warpverter is an attempt to avoid all the problems of high frequency switching and the critical physical layout requirements by combining the outputs of four low frequency square wave inverters to generate a very low distortion sine wave. In fact its direct digital to analog conversion at a very high power level. The disadvantage of the Warpverter is the large number of parts required to build what is essentially four inverters instead of just one inverter. So for up to perhaps 4Kw to 5Kw the usual high frequency PWM inverter design will be a lot simpler and cheaper to build with fewer parts. Above that power level is where the Warpverter really shines. There is really no practical upper power limit, it scales upward in power really well. Its definitely not for everyone though. Performance of PWM versus a Warpverter are pretty much identical as regards idling power, flat out efficiency, surge power rating, and harmonic distortion. No advantages or disadvantages there in measured performance between choosing either type. Edited 2021-05-15 08:57 by Warpspeed Cheers, Tony. |
||||
| flyingfishfinger Senior Member Joined: 12/09/2020 Location: United StatesPosts: 110 |
Hi, Thanks for the summary. This makes a lot more sense now. This sounds like it's the "state of the art". Madness's gate driver = power board, as I've seen the term used? Can you point me to where I might find the "current" sources for these boards? In my perusing thus far I see lots of people building them & reporting on their experiences, but not specifically where the latest designs & firmware are. Sometimes some seem to pop up in attachments to posts, but those are really hard to find and easy to miss... I'm definitely in the DIY & cheap versus expensive & off-the-shelf mood. My work has all the usual debugging tools save for high current power supplies, but I should be able to set up one of these projects given the layout, parts list & firmware. EDIT: I don't think I'll need the Warpverter power levels, but I'd love to read up on it as well. Cheers, R Edited 2021-05-15 08:56 by flyingfishfinger |
||||
| Warpspeed Guru Joined: 09/08/2007 Location: AustraliaPosts: 4406 |
I have sort of gone my own way with the Warpverter, and have not really followed the latest in PWM developments. But agree, that from what you have said, PWM would be the better choice for you. Just sit tight, someone else can better bring you up to speed with the latest... Cheers, Tony. |
||||
renewableMark Guru Joined: 09/12/2017 Location: AustraliaPosts: 1678 |
Madness no longer sells his power boards and he wont give out the file so you are screwd there. BUT Poida has recently made his own version, and he inteds to make it available for free to anyone to download and get boards made. I think he was perfecting it so you might need to wait a little bit till he publishes that. The nanoverter is actually just a control card, you hook this upto a power board. That is also free to download and have boards made. If you live in the States perhaps get 5 or 10 sets done and sell them to other locals. BTW these boards will run fine on 24v but not on 12v. Edited 2021-05-15 13:34 by renewableMark Cheers Caveman Mark Off grid eastern Melb |
||||
| flyingfishfinger Senior Member Joined: 12/09/2020 Location: United StatesPosts: 110 |
That's fine, I'll use a 24V system. I spent a good while reading Poida's MPPT thread, as one of those will probably also be needed (the little Victron BlueSolar's can't handle the full power of the F&P fully loaded) ... but if his MPPTs can charge 12V batteries, shouldn't the inverters also be able to run on 12V? I look forward to hearing from Poida, hope he chimes in. Cheers, R |
||||
| nickskethisniks Guru Joined: 17/10/2017 Location: BelgiumPosts: 458 |
With a few modifications it will work on 12V, it's the power supply, specially to the mosfetdrivers that's the problem. I might forgotten threads, but these are the most important: Poida's Various aspects of home brew inverters http://www.thebackshed.com/forum/ViewTopic.php?TID=9409 Buildinglink ozz inverter Renewablemark: https://www.thebackshed.com/forum/ViewTopic.php?FID=4&TID=10179 Nanoverter build renewablemark: https://www.thebackshed.com/forum/ViewTopic.php?FID=4&TID=11041 24V nanoverter Renewablemark https://www.thebackshed.com/forum/ViewTopic.php?FID=4&TID=11564 Inverter software topic, nanoverter by wiseguy https://www.thebackshed.com/forum/ViewTopic.php?FID=4&TID=11101 Madness boards: https://www.thebackshed.com/forum/ViewTopic.php?FID=4&TID=10348 Edited 2021-05-15 19:25 by nickskethisniks |
||||
InPhase Senior Member Joined: 15/12/2020 Location: United StatesPosts: 178 |
I'm very new to this too and have managed to get a working prototype with the info I've cleaned from this forum. Use the Nano control route simply because it is sustainable. And the sine wave is better. The 8010 approach relies on there being 8010 chips, a proprietary design likely owned by the PRC. With the Nano, theoretically you could use just about any microcontroller to replace it, wired correctly of course. So there's no worry of spare parts. The Warpverter is suited to >5 kW power. Plus it appears to be quite particular to the build quality. The standard H-bridge topology has some wiggle room for sloppiness, but the Warpverter expects a tip-top hardware environment and accurate component selection. Tony, that isn't an indictment of your design, BTW. It is a compliment. There is nothing particularly special about the Madness board, as far as I can dig into it online. I don't own one. He hasn't released a schematic but from what I can tell the biggest feature about it is a charge pump for the high side FETs and push-pull totem poles for each quarter of the bridge made from complementary NPN/PNP power transistors. Each quarter has an RC snubber. My first attempt at a board used totem poles to drive the FETs but a couple of blow ups led me to optocouplers to isolate the power side from the control side. I used TLP350 FET drive optos to switch the totem poles to switch the FETs. I found that I got optimal switching straight from the TLP350 without the intermediate totem pole. I use isolated DC supplies for the bridge, two for the high side and the low side shares one. The biggest thing you will face being in the States is finding toroids for transformers. We don't have the excess scrap toroid market like they seem to in the Southern hemisphere. There are several on eBay right now, one may work for you. |
||||
| noneyabussiness Guru Joined: 31/07/2017 Location: AustraliaPosts: 512 |
Inverter Board With a few minor tweaks, its a turn key solution of sorts.... still need temp control ( it has it built in, but simple) just add transformer and choke and all other usual supporting stuff... VERY robust, I've comfortably ( admittedly fans were working) pulled 9kw through mine at 48v, for around half an hour before microwave and aircon kicked off... I think it works !! |
||||
| Warpspeed Guru Joined: 09/08/2007 Location: AustraliaPosts: 4406 |
The problem with critical layout only has to do with attempting to drive multiple mosfets in parallel, exactly the same as with a PWM inverter. If you wish to build a very high power 48v inverter, well over 100 amps will probably need to be switched, regardless of what type of inverter it is, and the parallel mosfet problems all return. The solution for really high power is to run off a higher dc voltage, and keep the current as low as possible. That will require fewer switching devices per leg, and the efficiency will be higher too. High voltage mosfets are miserable devices with very high Rdson compared to lower voltage mosfets. IGBTs are far better at higher voltages. All the grid tie inverters run at high dc voltage and use IGBTs for that reason. IGBTs are also a lot more rugged and fault tolerant than mosfets, but are inefficient due to higher forward voltage drop if run from lower dc voltages. So basically, multiple mosfets are best at 48v but for high power you need a lot of them in parallel, which becomes a problem. At much higher dc voltages (100v or more) IGBTs are much better, but the really big ones are quite slow, and running a lot of small IGBTs in parallel we are back to all the issues of running multiple devices in parallel. The advantage of the Warpverter is the very slow switching rate, we can use big very slow IGBTs with ratings like 500 amps per device, and we only need to drive one device per leg which is very easy to do without critical layout. The whole situation is very different switching at 20Khz instead of 50Hz, its x400 times faster. Its like comparing a 1,600 Mph fighter jet to 4mph walking speed. The problems are far less  Cheers, Tony. |
||||
poida Guru Joined: 02/02/2017 Location: AustraliaPosts: 1418 |
And that is where I come in... I have compatible PCB design files for similar boards to the power board mentioned above. I made 5 boards with a couple of errors and I tested one with 4 x 3 FETS to start up the air compressor. This pulls about 11kW DC during the motor start period of about 1/4 second. It passed with flying colours. I've since fixed the errors and now the files are good to go. Are you interested in this? Approximate cost per board was about $25 including freight. You must specify 2 oz copper. Bill of materials is almost self evident - just look at the silk screen. wronger than a phone book full of wrong phone numbers |
||||
Haxby Guru Joined: 07/07/2008 Location: AustraliaPosts: 423 |
Edited/deleted post: I mistakenly thought Poida was taking group orders. Poida, can you point to the latest nano inverter code? Edited 2021-05-16 19:29 by Haxby |
||||
| poida Guru Joined: 02/02/2017 Location: AustraliaPosts: 1418 |
Haxby have a look here http://www.thebackshed.com/forum/ViewTopic.php?PID=168915#168915#168915 I know, nano2 code options are crazy. pick one. I will help you get it running. wronger than a phone book full of wrong phone numbers |
||||
| Warpspeed Guru Joined: 09/08/2007 Location: AustraliaPosts: 4406 |
And there Rafael, I think your original question has now been answered. Poida (Peter) is da man. Cheers, Tony. |
||||
| flyingfishfinger Senior Member Joined: 12/09/2020 Location: United StatesPosts: 110 |
Hi, Thanks everyone for the overview & options. Yes, I think I'll take a look at Poida's board. Can it be configured to output 110V? Our bush block is in the US after all, so that's what will be needed. What state is it in? I see the testing that was done with 4x3 FETs - does adding the full 4x6 FETs increase the power handling capability? Cheers, R |
||||
| InPhase Senior Member Joined: 15/12/2020 Location: United StatesPosts: 178 |
The output voltage will depend on the transformer you use and some trimming with a pot on the Nano board. |
||||
| BenandAmber Guru Joined: 16/02/2019 Location: United StatesPosts: 961 |
Worpspeed said it all above!!! I have been using hy5608 I have melted the solder off the legs without blowing the mosfets The higher amperage mosfets means less mosfets It would be nice if a pro would check them out I could be missing something be warned i am good parrot but Dumber than a box of rocks |
||||
| flyingfishfinger Senior Member Joined: 12/09/2020 Location: United StatesPosts: 110 |
I've never wound a transformer before, so I'm unsure what to look for that will work with Poida's PCB. Open to guidance & suggestions, though! Perhaps I should start a separate thread for a "Poidaverter" build? Cheers, R |
||||
| InPhase Senior Member Joined: 15/12/2020 Location: United StatesPosts: 178 |
You're looking for a transformer that can supply the power you desire at the input and output voltages you choose. Those three parameters are immutable constants once you select them. The rest of the inverter is designed around them. So if you want a 24 VDC system that outputs 120 VAC at 3000 watts, you're looking for a transformer of approximately 3 kVA with the a 120 VAC primary and 16 VAC secondary (Used backwards). |
||||
| flyingfishfinger Senior Member Joined: 12/09/2020 Location: United StatesPosts: 110 |
From what I gather, this is not something I would get OTS - I should wind this on a suitable toroid myself right? Will start a thread for my build shortly. R |
||||
| Page 1 of 2 |
|||||
 Print this page |
