 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 : hipower atmega168/328 mppt smps PVcharger
| Author | Message | ||||
| cukovanny Newbie  Joined: 09/06/2019 Location: YugoslaviaPosts: 7 |
hi to all. greetings from Serbia, exYu. i see that some/most of members here are great and inventive in power electronics and mcu designs and I learned a lot for my enthusiastic off grid projects from you. so, i wanted to share my own design of high power scalable PV mppt charge regulator. looking forward for comments and perhaps some improvements. :) The left atmega328 generates a push pull out of phase of 180 degrees with frequency, max duty and deathtime adjustable in code itself. when you set it,put in 1/2 of desired freq. actual duty depends on the input in the comparator to A0 like this: higher than 2V2-duty goes up. less than 2V2-duty goes down. I did not measure how fast the response is, but it's certainly very fast. possible much better than 10 microseconds. the right atmega328 measures what is seen on the schema itself and adjusts the output to D5 to ensure that the input voltage remains within the most favorable range - in order to extract the max power from the FN cells at that moment. the code itself is simple and takes the current voltage and power of photopanel into account . this technique is known in the capitalist world under the acronym mppt :) driver for MOSFET/IGBT is great, i mean really brutal and with proper scaling you can drive many kW with it. its a thing of scaling small toroid, transistors/flyweel diode and accu inductor. i stole this kind of isolated drive type topology from one huge welding apparatus of 30kW. wind small drive toroid transformer trifillar with 22turns and wind it on small toroid core that is R16 / N30 / K10000 material type. main inductor material is yellow-white accu toroid that you can find in secondary stage of any atx computer supply. dunno what kind of material is it, but i now its gapped material that is usable within 20-100khz switching frq. these yellow-white toroids are approx 30-40mm in diameter and as i said - im sourcing it from broken atx power supplies. for my 1kW PV sistem i wind stranded wire on 4 of these cores that are glued side by side. stranded wire is 36x0.20mm or 36x 0.25mm wire. it goes approx 18 turns on these four cores, and about two meters of fat stranded wire. if you have LCR meter , it should have 60-100uH inductance. also, ive done simple crowbar circuit for fatal exclusions/shorted transistors and so.... but it never stopped working or failed as im writing this :) best part is - that it is literary 20$ hipower solar regulator. most expensive part is 7$ I2C 16x2display and maybe high current ACS hall sensor as exotic solution. you can replace it with current shunt if needed. rest of components are free if you have access to old electronics dump - youre repurposing it. if you are interested of my design and/or have questions or have sugestions developing it further - please ask. my next stage would be to do real isolated topology with half or full bridge. ive made 3 peices of this regulator and it works now for two months on my INPUT of 75VDC and output od 14.4VDC. also one piece is working at input of 150VDC, and output of 28VDC. P.S. small mistake on pic i didnot notice: caps on 7812 and 7805 are 334(330nF) and 104(100nF), and electrolyte is 10uF.  2019-06-12_020817_left_atmega_pro.zip 2019-06-12_021910_right_atmega_pro.zip |
||||
LadyN Guru  Joined: 26/01/2019 Location: United StatesPosts: 408 |
Ago, WELCOME to this forum and THANK YOU for your contribution. 1. Where do you source your ACS hall sensors from? Ali? 2. How's your experience with them as far as reliability is concerned and calibration drift over time? |
||||
| cukovanny Newbie Joined: 09/06/2019 Location: YugoslaviaPosts: 7 |
and thank you my Lady for warm welcome and reply. i always wanted to build non-clamping solar controller, simple and powerfall and with "bom" list that you can buy at local hardware store. about ACS hall sensor, i have 2yrs expirience with 50Amp bidirectional sensors that came from Aliexpress. theyre little tricky to calibrate in the beggining, but very steady from then on. you need fine LF filter on output when using them in smps designs. also, if they dont work offtheshelf - thats it. chinese bussines :) just pick another. one more thing - you need math correcting it as amps goes up. on both sides. so, theyre not so precise, or it was my misuse or lack of knowledge.... but i wont be so hard on little chinese people, 95% of these sensors was ok. 10pcs for 18$ with shipment to my country. how do they counterfeit it ?!?!? as example, i will attach pic (in few hours) of one of my 1kw smps in oil bath (needed for silent run) that works in average temp of 60-80C for 1.5years now. original german transformer oil and no need for calibration of variable resistors nor other components.... no noticable shift in ACS sensor as i write this, no matter thermic and maybe chemicall distresses involved... cheers |
||||
| azhaque Senior Member  Joined: 21/02/2017 Location: PakistanPosts: 117 |
Hi Cukovanny, How does your design stop the reverse flow of current (from the battery to solar panel) during low light conditions? Is it by sensing the direction of the current by the ACS758? Regards azhaque |
||||
| cukovanny Newbie Joined: 09/06/2019 Location: YugoslaviaPosts: 7 |
no azhaque, i didnot had that problem. maybe because of factory diode inside small junction box on the back of panel itself that wont let current go backwards. i know that reverse flow is not present because i have series connected big old russian ampmeter on the output of controller which goes +75A and -75A. overnight it stands slightly shifted to a minus at about -0.3A due to regulator self consumption and fans in it ... but one big power diode (RURG3060 or something) on sole +100VDC input is solving eventual problems , does it ? |
||||
| cukovanny Newbie Joined: 09/06/2019 Location: YugoslaviaPosts: 7 |
promised oiled :) pic to LadyN:  and its display in that same moment (6h30 morning):  |
||||
| BenandAmber Guru Joined: 16/02/2019 Location: United StatesPosts: 961 |
Does the oil thing work very well I've thought about oil off and on for quite a while be warned i am good parrot but Dumber than a box of rocks |
||||
| cukovanny Newbie Joined: 09/06/2019 Location: YugoslaviaPosts: 7 |
one thing for sure: youll have more inert system in means of temperature... hard to get hot but also hard to cool down. one of these days ill find some small aquarium pump and let that oil circulate thru small aircon cooler from fiat punto or such a small vehicle. i have that little cooler somewere... in that case, you can double the power if you cool your ferite core and wire on it good enough... |
||||
| cukovanny Newbie Joined: 09/06/2019 Location: YugoslaviaPosts: 7 |
addendum about my solar smps mppt controller, and I'm not going far from the BenandAmber question: In this design, the MOSFET or igbt transistors are very tight driven, and the isolated gate driver itself holds the zero plateau strong and firmly and right in time when needed - when the other transistor starts to conduct. The result is that this pair transistors do not even warm up even with high input voltages and higher power. the only thing that is heating up in this circuit is the flyback diode. accu ferite core with wire winded on it has its working temp on most pc atx psu - around 80C - 120C forced cooling. no problem about it. im using double enameled industry standard copper wire that can withstand more than 180C without any problems. dispite all this, im not letting it cross 75-78C boundary. yes, my thermistor is stuck into windings on the ferite core :) in theory, if you could hold down the temperature of ferite and wire coil under 0C -you could double or triple the power. yes you can - with oil bath and peltier glued to this bath... :) cheers |
||||
| cukovanny Newbie Joined: 09/06/2019 Location: YugoslaviaPosts: 7 |
you know, BenandAmber, i've done it. last evening i connected small pump and small radiator that was built for automatic transmission oil cooler. results ? very, very good. it cooled down working solar regulator from 72C to 45C in 5 minutes. i tested it to see whats the max uncontrolled temp if something goes wrong and pump stops circulating oil in the bath. it went up to vicinity of 80C. only problem that i see right now is homogenisation of temperature over whole electronics (not good for electrolyte cond and maybe variable resistors), and possible repairs .... things goes sticky very fast :) as a reference, all process was going on when about 800W of power was flowing from PV to battery. |
||||
 Print this page |
