 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 : Solar : Direct Solar Electronics.
Page 1 of 2   |
|||||
| Author | Message | ||||
| Davo99 Guru  Joined: 03/06/2019 Location: AustraliaPosts: 1578 |
Hello All, I'm interested in doing a solar setup for heating a small greenhouse and maybe an extra hot water heater. I have a bunch of used panels and have them coupled to a water heater element I have matched the ohms with to give some useful heat. Of course the thing falls outside the optimum very easily and quickly and become inefficient. Very easy to have 2Kw of panels doing 500W where a similar set of panels on the same roof is doing more than double that hooked to a GTI. Monitoring things, it's obvious I'd be better off connecting them to a GTI and run the element through a PWM to control the amps down and let it heat over time. I would like to make it more efficient and grid independent though. Watching some YT vids some people have built boosters / controllers to keep the panels at the best voltage and improve efficiency. One guy modifys pre made inverter boards but watch his vids as many times I have, it's way beyond me. I can see this would be a great advantage but firstly i'm a complete electronics idiot and secondly it's subsequently over my head. These guys that say it's simple are clearly talking in relative terms! :0) In the past I have cobbled together a number of pre made components of all different off the shelf things and got the outcome I wanted. May not be the most elegant, simple or efficient but they have always saved a fortune over the specialised items to do the job. I am wondering about Piecing together some pre made components to get the same result with this problem. What I have in mind is an MPPT controller ( a real rather than fake one) or several Combined to give the desired amps and a PWM controller. The idea is I hook up the MPPT and then put the PWM after it to control the load being a 2000 or 3600W element. My probably incorrect thought is the MPPT ( battery charger) souldn't care if it's charging a battery or powering a load as long as that load does not present itself as a dead short like a mains powered element would. By putting a PWM behind it, I can reduce the load so I can use mains elements like with the direct connection but reduce the load on the controller to what it can manage. I'm thinking (probably too simplistically) that the MPPT would keep the panels at the right voltage and the PWM would limit the current so I can adjust it to what the controller can handle or a bit below it for Chinese over rating propensity. I have my doubts about this because it seems too simple and straight forward so can anyone confirm that and/ or point me to anything that would work and I might be able to cobble together? I was also thinking of something in Arduino which I have little skill in but might be able to cobble together with a sketch and schematic. Thanks for all and any help. |
||||
| davef Guru Joined: 14/05/2006 Location: New ZealandPosts: 499 |
Could you post a link to the guy that modifies inverter boards? I had a crack at that and came adrift. Reading through this thread: 3KW solar gave me the basic idea that lead to: Simple solar diverter This currently handles 150-300Watts worth of PV panels in a caravan. I have almost completed a 1KW unit for solar diversion on my house system. Basically, you charge a BIG capacitor up to Vmp plus a bit and then discharge it a bit below Vmp, ie say +/- 1 volt at 20Vmp (36 cells panels for a 13V system). So, as well as the BIG capacitor you need some BIG N-channel FETs. SolarMike actually applies AC to the load, whereas I am feeding a DC rated immersion element ... well that is what they claim. What voltage do you get off your PV system? |
||||
| hotwater Senior Member  Joined: 29/08/2017 Location: United StatesPosts: 120 |
I run a NANO water heating system heating. Modifying an inverter board isn't any harder than building one based on arduino. Just turning a FET on and off. It is mandatory to have a capacitor bank to store charge current during the off cycles. Just what voltage do you have to plat with and element ohms? Think of this as a learning experience. Take a week to install each part if needed. |
||||
LadyN Guru Joined: 26/01/2019 Location: United StatesPosts: 408 |
Hi Davo99! I am building a similar system as you - using 6x 300W 35Vmpp@9A panels. Yes, if you can please share some links to those videos, I would like to check them out. PWM is usually duty cycle modulation to handle the transfer of energy - if all you want to do is heat water, the MPPT unit by itself followed by a MOSFET switch (basically on/off PWM) should be sufficient. Eventually, if you're using a consumer grade residential HWS w/ AC relay contacts, we will have to emulate AC so you don't burn out your contacts but that is a bit far away in the future. Let's get the basics right first. I will let other more experienced than I share their thoughts. I/We brainstormed a unidirectional switch - check it out as you might need it later! MPPT chargers are designed assuming they will be charging batteries, so they do care whether they are driving a battery or a purely resistive load. The good news is that a MPPT water heater cuircuit will be much simpler than a complicated charger and it will be optimized to use the cheap $10 (or less) AC water heating elements you can buy from any HW store! You're absolutely on the right track - if something is not simple and straight forward, then it needs to be. I am planning on doing exactly that and everyone here has been extremely supportive about it. Please do not underestimate your potential - I am/was just like you with no clue about power electronics a few months ago, but now I can not only dream about projects but think of how to go about it. Yes, I will need the geniuses here to teach me the details, but now, I am actually able to design the circuits (with a lot of help). Interesting - do you have links to SolarMike's project? Is the ac SQAURE or MSW? I am on the route to building some MSW inverters that run directly off HVDC (110V DC - 210V DC) from PV |
||||
| Davo99 Guru Joined: 03/06/2019 Location: AustraliaPosts: 1578 |
The inverter Mod vid is here:https://www.youtube.com/watch?v=rf6bjPaNAsE There are several by the same person, obviously knowledgeable and well practised with this stuff. Just check the channel name for others he has done. My initial plan is to use this to heat a drum of water in the little greenhouse as a heat storage for the night. Been working well so far when the water gets up to heat. I can go straight DC for this because I doubt in winter I'll ever get enough heat to need to control temp. If I do I'll just add more water and have greater volume at lower temp and still get more stored heat. I would like to use it for home hot water boosting and I was going to use a DC SSR for that and use the thermostat as a low power switch for the SSR but if there is a better way I can get my head around, so much better. Due to some complications with anxiety and other things I do have bit of a learning difficulty these days and looking at diagrams and things and getting it to sink in is not easy or straightforward. Is there a link to the unidirectional switch you mention? I have got friendly with the local scrap yard bloke and can pretty much go and get elements any time I want. I'm going to look for a HWS that I can experiment on and if OK, can Plumb before the main tank as a pre heater. For now I think that's the best way to go as I can't stuff the main one and suffer the fallout of the wife and Daughter. This would take the load off the main heater and save on power but always mean there was full hot water available. ATM the elements I have pulled are 3600W Being 15.xx ohm and 4800w @ 12.xx Ohm. I notice they vary a bit even with the same element rating probably due to " wear" from being used. They are consistent in the reading I get from one day to the next so it does not seem to be my measuring that is out. As for panels, I have a load of them I have bought used. Just put up 5 Kw of 250's and have another 18 I think of 190's so throwing enough power at the project isn't an issue. I do notice that a lot of these controller run low Voltage, 60 or so which I guess is fine if you can get the amps and heating out of the elements. I'd really like something that will do a KW minimum and 2 Kw would be better. takes a fair bit of power to heat water ago fiddling round with 3 or 500W won't suit my needs and I can get that with direct hookup anyway. I'm in Sydney and would be open to just buying something off someone willing to make me a working unit. I know people say you have to learn etc but for me, it's far easier to learn how a car works having one to look at and deconstruct as it were that it is to get someone to build one that's never seen one before and does not know the principals involved. I would also be interested to know if an off the shelf MPPT controller and a PWM or a simple current limiting circuit behind it would work? Thanks again for any and all help. |
||||
| LadyN Guru Joined: 26/01/2019 Location: United StatesPosts: 408 |
Here you go: https://www.thebackshed.com/forum/forum_posts.asp?TID=11367 Look at Solar Mike helping me out with an implementation of what I was talking about! Isn't this site and the people here amazing? Also of interest might be: http://www.thebackshed.com/forum/forum_posts.asp?TID=11391 I would not waste money behind a DC SSR if I were you but take the threads above forward. DON'T second guess yourself - just start building while being safe and post your progress! Those are exactly the ones I was starting off with but now I would recommend buying 120V 1.5kW elements to get the basic blocks up and running. We will then graduate to higher power. OK, that channel is absolutely brilliant and the gentleman who makes those videos is obviously talented, brilliant and knowledgeable. I have watched each and every video atleast 10 times and some of them easily double that. I now have a vague idea that the creator has a shunting type battery charger in parallel to his HWS and the HWS is used as the dump load: whether I'm right or not, maybe we will figure it out together. I just wish the creator had posted the design of his whole system as otherwise, looking at parts of it is confusing. However, I personally have certain disabilities that make those videos hard to follow - I have extremely poor vision and unless there are clear schematics, I have a hard time reversing photos to obtain schematics from them. Obviously, he designed the parts taking his whole system into consideration, but I believe, he assumes that either knowing his his whole system is unnecessary for the viewer or too much detail. Further, the creator of those videos seem to be in a state of resignation and unhappy over his target audience - probably because the people commenting on his videos ask him about too many details that he himself thinks are not important to know or trivial to figure out. Looking at those questions, I can see there are stumbling blocks for those searching for those videos in the first place. Infact, people without those stumbling blocks can design the system like his without his videos: so his videos end up frustrating the very target audience he ends up attracting unknowingly. The more concerning thing is that people without those stumbling blocks do design their own systems from scratch while the creator shows systems put together, modified from ready made systems - and those people attack him, like in his MPPT charger video and those people stop visting his channel as well. Unknown to the creator though, his circuits have been copied in many projects on hackaday and other forums - some posts link to his videos/posts directly but since youtube does not show the creator the incoming/referral links to his videos, he is not aware of that detail. I use his channel as a source of inspiration on what to do and what can be done but leave it at that without further pressing the matter. He's not unique in his viewpoint though - my physics and computer teachers are like him too: they are absolutely great people to help me dig deep into the subject matter but will not, at all, help in maters they deem as trivial or of little value or the responsibility of the reader. SORRY for the long post - I hope the creator can read this post in its entirely but I'm afraid he might misinterpret my thoughts above, so I will just post it here and get your (and other's) input to see if I am being reasonable or not. Why don't we do this - you take a video from his channel and post your questions here - you, I and anyone else interested here will then figure out the details and what the creator assumes to be the responsibility of the reader. |
||||
| Davo99 Guru Joined: 03/06/2019 Location: AustraliaPosts: 1578 |
Thanks for the links. I do try to find and study all I can myself. I have taught myself a LOT from the net and even set up profitable businesses from what I have seen and people I have spoken to on forums that have become good personal friends. I often run into the problem of not knowing what something is called in the first place in order to search for it. Example ATM is wanting like paving stones retaining wall blocks that are cut wedge shape so I can make a neat Circle to build up for a fire put. NFI what they are called and I must have looked at 50 things about fit pits using them but not one single instructional tells you what the heck they are called!! I search for pavers and retaining walls and see everything but what I am after. I have already bought a load of SSR's for other things so money is already spent on thise but open to other ways of cooking the chook. Between this and My last post looking for more info, I remembered the several old inverter boards I have up the back I have picked up. One already disassembled has 6 350V, 1000uf caps and a few smaller 350 uf, 350V caps so these should be a good start to the cap bank. I think I have another inverter up there the same I haven't pulled apart yet. Just trying to figure out how many I would need to be useful. The board also has some 44A, 600V Mosfets which may be handy for the switching and be smaller than the SSR's. I'll read up on the bit about switching them as that was what I was just thinking. I REALLY don't want to insult anyone or burn any bridges with the Gurus here or else where but I agree with what you said RE the boffins who know this stuff. The creator of those vids seems disappointed that no one follows his vids and builds the things. I'd love to but while they may be ABC simple to him, they sure as heck are not to me. One of my skills is in marketing, advertising and sales and from this perspective I can see he is missing his market completely and being frustrated by it. It's not people don't appreciate his efforts, admire his skills and would love to build what he is showing, it's just over their heads and too hard to piece together. I also got the impression if you could follow what he was doing you'd have the knowledge and smarts to work it out for yourself and could probably just look at it and see what was going on. There are things I'm considered expert in and I have seen things from the other side. I have had difficulty explaining things to people starting out because I too have forgotten what it was like to be at that point and also find myself incapable of getting the info across without confusing the guts out of those I am trying to teach. I can well understand it's not easy being at the top of the tree and trying to tell others how to climb it when they don't speak your language. I have come across this many times with electronics and has been a disappointing thing in learning Arduino for me. May experts say " You have to learn this yourself" and only have half the detail like the sketch but not the components you need or the layout. It's like asking someone to build a car that has never seen one or really understands what it does and what it required to make it work. I'm not an idiot, I doubt a lot of these guys could keep up with me in repairing vehicles, rebuilding engines, setting up computer servers and networks, business principals of the Restaurant industry and other weird and useless things because simply their skill set is different to mine. We haven't all got the time to learn everything. The other thing which some seem to take exception to is that fact that people may not want to learn all about electronics and that is not the project. In this case, the project for me is not learning how to do this, it's using it as a part of my Greenhouse operation which I would like to go further with and building it into my other DIY self sufficiency projects. Just like a solar panel or a GTI, building them is not the project, they are just a part of a larger goal that relies on them. Just like a car manufacturer makes some components of a vehicle themselves, they buy in many others. The idea is not to manufacture those parts but rather the end goal which is assembling and selling Vehicles. I appreciate the time people take posting this info but at the end of the day I think they are just on such a high level with it they forget what it was like when they started. Probably a lot easier to learn with formal training but I couldn't get that at my age if I tried because I'd need an apprenticeship to get into tech to get it. The best we can do is find out what we can but as I have said with other things, It can be difficult to get to the top of a 20 meter ladder when the first 3 Meters of rungs are missing. The vids have been useful to me in that I DO understand the theory of operation and why it is important. I can see the great benefit in what he is saying and am convinced it is the only way to go for any decent efficiency. This is why I'd like to do one or have one built for me. Once I saw one I'[m sure it would be easier to copy. I have trouble with diagrams/ schematics right from the start because for one thing I can't lay the components out like they show and get confused with the connection points. anything over about half dozen components and I'm out of my shallow depth. This is why I'm wondering about the MPPT Controller. I'm thinking that's got the ability to regulate the panels. I'm wondering if I can use that to control a capacitor bank and where I could add that in or if I could use it like taking the load off the thermostat in a water heater and just using it as a switch to Drive and SSR or a Mosfet that does take the load. At least with this I can keep things clear in my mind about what I need to do. I'm also thinking that for the more knowledgeable this should be a cake walk in arduino. All one would need to do would be monitor the voltage coming from the panels and turn on a Switch ( relay/ Mosfet) when the voltage was at target value and switch it off when it fell below set parameter. Panels could charge capacitor bank which was switched to the element. I'm not even sure it needs MPPT, just PWM at the set points would seem to do the trick but maybe that is MPPT?  Looking at that I ran into a dead end looking for voltage dividers. I'd have liked to get that up to 150V+ But I couldn't find anything to do it. I'm not sure how much power I can get from something like 60V but maybe I could use multiple elements? Anyway, getting confused in my head now about what to do and which way to go so I'll go and do some more reading and see if I can get a strategy in mind. :0) |
||||
| LadyN Guru Joined: 26/01/2019 Location: United StatesPosts: 408 |
I would love to be one of them and am happy to work with you to earn that privilege. I occasionally program firmware and little backend servers for businessmen like you who have other HW engineers working with them who are afraid of firmware/software but are gurus in hardware. So the key parameters for: i. Caps: Ripple current rating aka ESR ii. Mosfets: SOA and Rds_on. I have posts on that, look at my profile You have to know these (very easy to understand) and keep them in mind That's exactly my observation as well. Absolutely Right, plus in one of the creator's videos about MPPT chargers, a person who has been building an MPPT charger got in arguments with him and stopped commenting on the future videos, so I'm assuming that viewer was lost too. Can you state your objectives with that system? The risk here is that even if you buy something readymade, it will be custom. Either custom made for the author's own system or for you. There currently are two standard systems - legacy 12V systems that are sold retail and wholesale HVDC systems being installed by professionals that are meant to be used in a grid tie fashion the internals of which are designed not to be exposed by the user. We are in the middle of these - legacy 12V systems are too expensive because of the low volume and high currents and it's too much work trying to reverse engineer the wholesale HVDC systems. So everything we biuld for our use case is custom. Because of financial reasons I cannot have a custom system designed for me - so I have decided to invest my time and learn what's needed because I want to leave my family with something they can operate and maintain for decades to come. That's absolutely correct! It does need MPPT. Think of PWM as a dumb switch - it connects the PV panels directly to the load/elements when it's on. Now, PV panels are current sources but at a set point they provide the best current, so you need to discover what that set point is, otherwise the power that can be sourced from the array will be VERY VERY low. Rarely will you be able to match that set point to your load - so why even try! Instead, match your load to that set point and DC-DC Convertors (which a MPPT controller is) helps you do exactly that! I propose this: 1. Let me build some circuits for you - they will be untested and can kill you if you are not careful. I will try my best to tell you how it might injure you but you really have to take care of that yourself by hiring an electrician or someone who knows the danger of HVDC 2. You build them or get someone to do so and give me feedback 3. We do this until you get what you want As a first step, give me the datasheet of the caps and fets you have and of course, your dream system specs |
||||
| Davo99 Guru Joined: 03/06/2019 Location: AustraliaPosts: 1578 |
Thank you very much for your generous offer and support. Very kind and good of you. I like playing with high Voltage and am not worried by it. I have wired up all my own solar panels and been playing with IMAG generators for years. Currently working on hooking up a 12 Kw unit. I may know beans about electronics but I'm not thankfully as in the dark with electrics. As for electric projects, if it can't kill you it's probably not too useful or much fun. :0) Being cold and windy outside today I have been watching YT vids and trying to get ideas. Still thinking about the PWM driving a mosfet and cap banks. I notice there are a lot of little 12V Mppt boards out there but typically not much in the way of practical stuff. I'm still thinking if I could find a divider with a high enough voltage I could use a 12V board as a switch to Drive a higher voltage Mosfet. I have read about using plug packs but I'm not sure if that would only work with an AC source of power rather than a DC. Anyway..... Ultimately I would like something that would Generate 1KW Min, 2 KW would be nice. Obviously depends on the power available on the day but those capabilities on a good day would be ideal. I would like to use the standard water heater elements I have/ can get. I don't mind if I can cobble together a bunch of pre made parts like controllers and a few components to get the result I want. I have 2KW of panels I can dedicate to this conveniently and can add more, I think I have 4Kw of 190's up there now and was thinking of moving the 250's to another face of the roof and putting the 190's where they are now which is a west facing roof. Not ideal but they do OK especially with the brute force method. All my inverters are well over clocked with panels and as such still do OK even on crappy winter days. Only problem is while they still do good output, they don't get many hours of sun to do it but that's another story. The caps on the board Look like they are KMQ brand. They are 315V ( not 350 ) 1000uF Under the spec they have 25VM3M. Not sure if that is a spec or not. This is the spec sheet on the mosfets: https://www.datasheetq.com/datasheet-download/750893/1/Infineon/6R045 I think there are 4 of them, some of them have silicone or something obscuring them a bit. There are also k75t60 x2 I think and 4 thyristors, https://www.digchip.com/datasheets/parts/datasheet/244/CLA50E1200HB.php Also a couple of these Diodes although the ones I have are a 3 pin design with the middle pin being short and branching to pin 1. Maybe I got the number wrong, again it's hard to read even with a powerful magnifying glass. If any of these can be used, great, if not...there is always fleabay for components. Thanks Very much again for your help. It is most Humbling and appreciated. |
||||
mackoffgrid Guru Joined: 13/03/2017 Location: AustraliaPosts: 460 |
Gday Dave, Are your panels hooked up to GTIs? Dave, I've forgotten 90% of what I learnt about electronics and have picked up a few things along the way. As I'm getting older, I've lost all my bravado in all respects and have picked up a healthy respect for high voltage. So what is a IMAG generator? I'm not sure that I've picked up on everything you've said and not wanting to stand on Natasha's toes, but he goes it. (its also windy and cold outside so I don't want to go down to the shed). You could just string 240 -ish volts together and PWM / Bit Bang (using your favorite micro) that lot into your heater elements. This is the easiest. You may not feel comfortable with high voltage swimming around you roof etc so you may prefer to string your panels into ~48 volt strings. You could then use one of the inverter designs off this board with some custom software to achieve MPPT etc. You also don't need a sine wave as you probably appreciate. You could use the OzInverter with the nano controller. I'm building a WarpVerter so my mind goes to using that topology, but with only one transformer - so just a square wave inverter - real simple. Either way, you'll be able to cook many kWatts of water. Cheers Andrew |
||||
| Davo99 Guru Joined: 03/06/2019 Location: AustraliaPosts: 1578 |
Hi Andrew, I have a bunch of panels hooked up to GTI's atm. I have 4 arrays each with Multiple strings and on an inverter each. Some of my strings run 350ish volts and the others are around 270 depending on the position I have to lay them out. One array has an old transformer GTI which I like and that has 5 panel strings. I have played with just using a bunch of panels and hooking things up and learned a lot with it. I was running a 500W light off a bunch of panels that had an open circuit voltage of about 400V. I learned at least with resistive loads, as long as the power from the panels does not exceed the rated wattage of the load, they just pull down and the load will work fine even if at reduced output. That can be handy for some things but less so others. An IMAG is an Induction Motor As Generator. A 3 phase motor spun over synchronous Voltage to act as a generator. I have been playing with a 5Kw unit and some smaller ones for years but recently acquired a 12 Kw motor that weighs in at a healthy 170 Kg. There are many different way you can set these up. You can do them as 3 phase or single, you can have them as stand alone generators using caps or grid tie them to backfeed. You can back feed one or 2 phases and drive a load with the others and so on. Very entertaining, interesting fun and practical. I know what you mean about forgetting 90% of what you know. In building this big IMAG I have to refer back to Vids and pics I took of what I did to remember setups that I knew by heart when I was playing with it regularly. High voltage doesn't worry me in the slightest. I went though a very dark time in my life a few years back and put my being on the blue marble to the test. I fired the IMAG up, had stupid voltage and amps coming out of it with bare connections everywhere and paid no attention to them what so ever as I fiddled about with the thing. I was surprised it didn't flash over or a connection arc onto me. After firing it up once or twice a day for about a week and nothing happening, I figured I couldn't be that lucky unless the universe was telling me something so I packed it up for the time being and went back to not tempting fate about as hard as I could without pushing it to only one conclusion. I have done a LOT of stupid and dangerous things all my life and NEVER got more than a scratch if that. The numerous times I have been to hospitals and medical centres however was from everyday things that no one would ever say " Be careful" with. Had I hurt myself doing stupid things I knew damn well I shouldn't do, I'd say serve myself right but when you are doing an everyday thing like walking up the drive way to get the letters and slip on wet leaves and mud that was normally not there and end up having your head sewn up so your brains don't run out all over your shoes, It's bloody annoying actually.  Who ever said, " Be careful walking down the perfectly concreted flat driveway when you go to get the mail"? I was running the heater element off 2 strings of 4 250W panels and it was working very well till the cloudy weather started coming over. Hooking in a volt and and amp meter I quickly realised that I'd be far better off coupling the panels to another GTI and just running the element though a PWM controller I wired up in a box with a power meter and a power point so I can control and see how much power I am running in something. The other thing I could do would be just parallel the panels with the 4 way connectors I have and run a couple of good PWM controllers into a couple of batteries and run my 2Kw inverter off that and limit the current as I have in the past with this with the PWM box. Thing is I have done that and it's all a bit clunky. I'd like to make it a lot more compact as I can see others have done and not have to throw $500 at the exercise. I'd also like to be more grid independent. I have shiploads of solar but little independence with it. Just seems really hard to utilise the power without a whole load of contraptions once one gets away from the grid. I have seen/ read of the Oz inverter but it is far and away beyond my abilities. I am surprised there are not more Chinese made boards for High voltage/ power solar. Cobbling together a bunch of little boards to get the result I want would be fine and I'm most at home with things like that but there is nothing I can get away with as easily here. I'm still trying to map out in my head exactly what is required for this. I understand where I need to go, I just have no clue of the vehicle and road to take to make it happen. But, If I can get my lame brain around it, I would like to. I spent a lot of time looking at different things today and learned a lot but probably confused myself more. I'm amazed there isn't more on this. The guy I linked to in the vids seems to have the best handle on this but there are too many missing rungs in the ladder for me to get my head round. Other things people have done seem far more Tiddly winks and not going to give the power output I want. I have plenty of panels to throw at this and i'd like to get some semblance of efficiency out of them. If I can't work something out I can always just hook the panels back to a GTI and run the element off the mains but I'd like to at least try to do something more independent. I have set up a lot of things with the mindset of being able to run them in the middle of a paddock and that's what I would like to achieve with this. |
||||
Tinker Guru Joined: 07/11/2007 Location: AustraliaPosts: 1904 |
Davo 99, Lady N, Please do us all a favor and use the wonderful hotlink link feature of this forum. It can be found in the 'post reply' header, the 4th icon from the left (looks like a globe with a chain link). This would make reading your long posts easier instead of having to use cutting and pasting text to see what that is all about. Thanks. Klaus |
||||
| Solar Mike Guru Joined: 08/02/2015 Location: New ZealandPosts: 1138 |
The easiest solution is to do as Hotwater has proposed, perhaps he hasn't explained things clearly enough, so I will give it a go. 1: arrange your PV panels into a series array that present a mp voltage somewhat higher than your HWC element is rated. 2: Feed this into a large capacitor, have a voltage detector that has a lower trip at less than the elements rated voltage and another at the element voltage. The voltage detectors can be your arduino or TL431 etc. 3:When the cap charges to the upper trip point the control circuit discharges the capacitor across the HWC element via a mosfet switch. The voltage on the cap will drop over a few milliseconds, depending on how much sun is shining and how big it is. When it drops below the lower set point the mosfet is turned off. If there is a lot of sun it will charge up very quickly, if not much sun then the charge period will take longer.... the cycle repeats. 4: The outcome is the element is subjected to a series of full power current pulses, their rate being determined by how many KWs of panels you have and the available sun light. There is always an off time (unless you have too much PV that would run the element continuously.. so increase the upper voltage trip point). 5: Not mppt but effectively a current booster as the capacitor can discharge at a higher current than your PV array can output. The capacitor may be several in parallel (1000uf or so) to reduce their stress and lower the esr. Cheers Mike |
||||
| Davo99 Guru Joined: 03/06/2019 Location: AustraliaPosts: 1578 |
Thanks Mike. Clarifies what I had worked out but wasn't confirmed. I would like to use an array of 8 panels which would give me 270V and 2Kw. You say to use a voltage Divider. I spent a number of hours today trying to find something to use as a voltage divider that will do that voltage but couldn't find anything. I see the TL431 has an input max of 5V. the most I could find with arduino dividers was a handling capacity of 25V so the question would be how do I use that on the 270V the 8 panels would generate? This is a stumbling block for me straight off. I take it the switching is done by the sketch in the arduino. I looked at that a lot today too and found nothing. Loads of battery chargers which I think I could adapt but nothing to handle Kw of power I'm after. Would the 6x 1000Uf caps I have be enough? Is more Better if I have another 6 on another inverter? Thanks again for your help. Much appreciated. |
||||
| LadyN Guru Joined: 26/01/2019 Location: United StatesPosts: 408 |
Andrew you're too kind. No one on this board can stand on my toes: I can only learn from each of you so never let this possibility cloud your mind again! :D Davo, you already know how voltage dividers work but might not have internalized it yet. I'm happy to work the details through with you but here are my thoughts on setting the TL431 2.5V Vref from 270V. You will need to power the TL431 itself as well, and that too can be done via another voltage divider but let's walk through the TL431 2.5V Vref from 270V first: Say you have a few 1/4W resistors, so your R * (I ** 2) losses cannot exceed 1/4W or you will burn out those resistors. Keeping this in mind, let us see what ratio of 270V is 2.5V: 2.5/270 = 0.009 Of course, you will have to put in some margin, which is up to you because a 270V will vary depending on temp. and insolation but we leave that out for now to help simplify this post. So now we have two constraints: 1. We need a 0.009 ratio for a voltage divider 2. We need to ensure the current through that voltage divider does not make us exceed 1/4W This is a constraint satisfaction problem: a computer can solve this very well. You will need to give it the set of resistors you have and it will tell you what combination matches. A sample solution is: 1k, 1M resistor divider The 1k, 1M resistors tied together allow: 270/1001000 = 0.00027A = 0.27mA through it. The TL431 datasheet says the Vref source should be able to source atleast a few uA, so we are good there. Also, 0.27mA dissipates less than our 1/4W limit through the 1k, 1M resistors so we are good there too! The 1M resistor dissipates the most at ((0.27 / 1000) ** 2) * (1000000) = 0.073W << 0.25W margin So, you stick the 1k, 1M resistors in series, with the 1M resistor on the high side (to 270V) and 1k resistor on the low side (to ground) with the middle pin going to the TL431 2.5V Vref Others reading this, please let me know if and where I made mistakes. |
||||
| mackoffgrid Guru Joined: 13/03/2017 Location: AustraliaPosts: 460 |
Natasha, just a quick comment, you could be justified to use two 470k resistors in series just to increase your break down voltage. |
||||
| LadyN Guru Joined: 26/01/2019 Location: United StatesPosts: 408 |
Andrew, thank you for reading all of that post! Are you recommending replacing the 1M with 2x 470k so as to increase the current and hence effectively increase the voltage across the 1k which provides the Vref? |
||||
| mackoffgrid Guru Joined: 13/03/2017 Location: AustraliaPosts: 460 |
Natasha, I'm guilty of just skimming the post  My suggestion of two 470k is just precautionary, its more about the break down voltage - its not a major thing but its starting to get to the voltage where you're better off having 135v across a resistor x 2, rather than 270v across the one resistor. This can also help with creepage. |
||||
| Davo99 Guru Joined: 03/06/2019 Location: AustraliaPosts: 1578 |
Couple of questions..... If the dissipation is too much for a 1/4w, why not just use a half or a 1w Resistor? When you say: " with the middle pin going to the TL431 2.5V Vref" what Middle pin? aren't we just wiring resistors to the TL431 that has the pins on it? Where does this other middle pin come from? Sorry for the Dumb questions, I'm hopless at this. I can't even work out one component when it's handed to me on a Platter.  |
||||
| Tinker Guru Joined: 07/11/2007 Location: AustraliaPosts: 1904 |
Davo, its not so much the dissipation at the resistor(s) but their physical size. 1/4W resistors are a bit too small to handle high DC voltages safely, you might get flash over. If it was me I would use 1W resistors. 101 voltage divider: You can try this yourself in its simplest form. Take two resistors, anything you have at hand as long as one of them is bigger than 1000 Ohm or so to limit the current. Connect these two resistors in series. Connect the free ends to a battery, 12v is fine. Now, measure with your multi meter between one battery terminal and the junction of the two resistors. It will read less than the battery voltage. You have created a *voltage divider* which allows you to get a smaller voltage from a bigger one. There are limits with regard to power tapped of the divider but by selecting the right resistor values you can create just about any voltage. If you replace the two resistors with a potentiometer you can get a smooth adjustable output voltage. Klaus |
||||
| Page 1 of 2 |
|||||
 Print this page |
