 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 : Few quesions about inverters
| Author | Message | ||||
| gigabyte091 Newbie  Joined: 15/10/2018 Location: CroatiaPosts: 17 |
First of all, hi to everyone on forum  Im trying to build small sinewave inverter based on EGS002 board, nothing fancy or high power for the first time, about 200 to 500W.I have several 12V and 24V to 230V transformers, EI style core from old APC UPS. Im plan to power old turntable with induction motor (about 20 or 30W), wireless phone with old style iron core transformer (10VA), two 6W LED bulbs and maybe an LED LCD TV (30 or 40W max). Battery voltage is not critical as i have about 50 pcs 7Ah 12V CSB UPS batteries, but i prefer 24V system. How to calculate required primary voltage for 24V battery ? I read all about tunning secondary to 75Hz with capacitor, but i dont know how to calculate choke inductance. Also for power switches i will be using RFP40N10 MOSFETS, 8 pcs, 2 in parallel and for testing i will use integrated IR2110 drivers but i plan to make my own driver boards with TLP250 drivers. |
||||
| Warpspeed Guru  Joined: 09/08/2007 Location: AustraliaPosts: 4406 |
Welcome to the Forum gigabyte. [quote]How to calculate required primary voltage for 24V battery ? [/quote] First thing you need to decide is the type of battery and the minimum battery voltage the inverter will ever see. It may be a "nominal" 24 volts but lets, assume it can get down to a 20 volt minimum under extreme discharge. Now your inverter needs to generate a sine wave, and the peak of a sine wave is a lot higher than the stated rms voltage. If everything was perfect, the largest sine wave you can make out of 20 volts of dc has a peak voltage of 20 volts. So the rms voltage on the primary will be 0.707 of that or 14.14 volts. But there will also be some conduction losses all through the inverter when its working hard, and some voltage drops at flat out full power that need to be allowed for as well. You might realistically need a 12 volt or a 13 volt primary winding for a 24volt rated inverter. That sounds pretty extreme, but in practice that will be what is required. Cheers, Tony. |
||||
Madness Guru Joined: 08/10/2011 Location: AustraliaPosts: 2498 |
As per this post from Oztules you will need a 18:1 ratio with the secondary for a 13 volt primary. This is assuming you are building a EG8010 based Inverter. If the is any chance that you will be building anything of more substantial power I would recommend using 48V from the outset otherwise it is much more difficult to convert later on. 4 X HY4008 MOSFETs would easily give you the power you require. There are only 10 types of people in the world: those who understand binary, and those who don't. |
||||
Clockmanfr Guru Joined: 23/10/2015 Location: FrancePosts: 427 |
I am thrashing 2nd edition at present. here is a bit about the toroid. 2018-10-16_170018_OzInverterNo2BookToroid.pdf Everything is possible, just give me time. 3 HughP's 3.7m Wind T's (14 years). 5kW PV on 3 Trackers, (10 yrs). 21kW PV AC coupled SH GTI's. OzInverter created Grid. 1300ah 48v. |
||||
| gigabyte091 Newbie Joined: 15/10/2018 Location: CroatiaPosts: 17 |
Thank you all for your answers. The reason for 24V system is that i have ready 24V solar charge controler and 50W solar panel and i think that for 200W (120W in average) 24V is high enough voltage. Ofcourse if i ever build something bigger i will go for 48 or higher voltages. MOSFETs ill be using are from old apc ups, it is 100V 40A 0.04ohm rdson. This is for the first or test inverter so if i burn any of them i wont be sorry haha. How do you calculate required choke ? is this dertemined by experimentation ? |
||||
| Warpspeed Guru Joined: 09/08/2007 Location: AustraliaPosts: 4406 |
[quote]How do you calculate required choke ? [/quote] That is a perfectly valid question, unfortunately there is no precise single correct answer. Its not a simple thing to explain either... The function of the choke is to reduce the peak currents seen by the mosfet switches when they switch the voltage on and off very fast into any capacitance across the load that the mosfets see. That load capacitance has to be charged and discharged twice per cycle to the battery voltage at 23Khz, and that can represent a lot of circulating power. The transformer secondary, will have capacitance between turns, and between the winding and the core, and possibly extra added physical capacitance required to filter out any residual PWM and tune the transformer. That charge and discharge current will be magnified by the square of the turns ratio and reflected back into the primary. Its no joke ! It can considerably increase the idling power as most of us already know. There are other problems as well with a rapidly pulsing current through the transformer at 23Khz. Skin effect in the wire which unnecessary heats up the wire, especially at high inverter power levels, and eddy current and hysteresis loss in the transformer core which also adds to increased idling power. The series choke will also reduce the peak currents through the mosfets at high power which can greatly reduce the stresses on the mosfets and add to robustness and reliability. Its less likely to go bang ... So the answer is, the more choke inductance there is, the more the above problems are reduced. But those blessings do not come for free. The choke is going to add additional series resistance which can produce significant heating and power loss flat out. If the choke saturates it effectively stops working, so although a wimpy little ferrite choke can noticeably reduce idling power it does nothing to reduce the damaging current spikes through the mosfets at higher power levels. So how much inductance ? A practical approach to this would be to select a steel core that looks to be big enough judged by size and experience, and what the commercial inverter designers might use at a similar power level. Bigger is ALWAYS better here, and the commercial guys are cheap skates. Next step is to select a wire size that is not going to excessively overheat at full flat out inverter power. Something like four amps per mm squared might be about right as a first guess. Once you have your choke core and wire selected, fit as many turns on as you can possibly squeeze onto that core. The more copper you can possibly fit onto the core the better the final result is going to be for a given size of core. The very last step is to adjust the air gap. That is going to be a compromise where you trade off some inductance for a higher saturation threshold. A higher power lower voltage inverter will require a larger gap. A low power high voltage inverter will need more inductance and a smaller gap. Two ways to adjust the air gap. Best is to test the choke on a test fixture where the saturation current and inductance can be measured. Set the gap to get an acceptably high saturation figure of perhaps 1.5 times (minimum) the expected full load inverter dc current. The other way is to use a Hall current sensor and measure the ac ripple current through the choke with an oscilloscope on the actual inverter. Work the inverter up in stages up to full flat out maximum power, and adjust the air gap for minimum peak to peak ac ripple current. A definite minimum will be found. That will adjust the choke for the maximum incremental inductance of which it is capable under those specific inverter operating conditions, given the core size and turns you have fitted onto it. Whatever the actual inductance value turns out to be, it is the optimum for what you have. If the ripple current is still unacceptably high, you need to start again with a larger steel core with more turns and more copper. If it overheats, you need fewer turns of fatter wire, but then that will reduce the achievable maximum inductance. Its all a compromise, and practical testing and development is probably a better approach to all of this than just trying to work out a theoretical requirement with pencil and paper. Start out by hunting down a nice big steel core and work from there. Cheers, Tony. |
||||
| gigabyte091 Newbie Joined: 15/10/2018 Location: CroatiaPosts: 17 |
Thank you, your and answers of other forum menbers are really helpfull. I have a lot of diferent inductor cores, ferrite or steel cores, making air gap in ferrite cores is not a problem, steady hand and dremel For transformer tunning capacitor i used formula for resonant frequency, measured inductance on transformer secondary and for frequency i used 75 Hz. |
||||
| Warpspeed Guru Joined: 09/08/2007 Location: AustraliaPosts: 4406 |
Use a laminated steel core, preferably silicon steel U cores if you can find some. Second choice would be salvaged ordinary soft iron E and I transformer laminations. The same identical ac ripple current at 23Khz flows through both the choke and the transformer primary as they are both in series. If the ripple current can be kept very low in the transformer, it will also be low in the choke, and both can have a steel core. Steel core will be much cheaper, and will allow several times the flux density of ferrite before it reaches saturation. Its a far more suitable material for what we are trying to do. Ferrite is for high frequency transformers where you have a high frequency square wave on the primary, and expect to see an identical square wave on the secondary, as in a switching power supply that has minimum core loss. We are trying to ELIMINATE that 23Khz component with a relatively high series inductance. If the choke is working properly the magnetic core will see only a 50Hz sine wave current, with absolutely minimal 23Khz ripple current superimposed. Yes a ferrite choke will reduce the idling current, but it will just saturate and stop working even with a small inverter load. Use steel and your mosfets will be working under far less stress if the choke can always work below saturation. Cheers, Tony. |
||||
| gigabyte091 Newbie Joined: 15/10/2018 Location: CroatiaPosts: 17 |
So I find this choke from old LG Inverter AC, it's rated @ 28A and 680 uH and have about 5 or 6 mm air gap on the middle column. Also i included picture of choke, mosfet setup and transformer im planning to use. Now i have to buy TO220 mounting kit and made small busbar for H bridge. Im planning to use 2 4700 uF 50V elkos on the input.  |
||||
| Warpspeed Guru Joined: 09/08/2007 Location: AustraliaPosts: 4406 |
That would be perfect for about a 500 watt 24v inverter, which I am guessing it probably came from in the first place. It looks like you have hit the jackpot finding that choke. Cheers, Tony. |
||||
| gigabyte091 Newbie Joined: 15/10/2018 Location: CroatiaPosts: 17 |
It actually came from old air conditioner, i think i have another one, but 7A @ 4 mH, a slightly bigger core, also with a air gap which could be easily rewound if necessary. I first plan to test inverter protections without input caps on lab PSU which is current limited so in case of cross conduction i dont blow the h bridge but i already ordered some TLP250 so i can make your version of anti cross conduction driver. |
||||
| Warpspeed Guru Joined: 09/08/2007 Location: AustraliaPosts: 4406 |
Seven amps is not a lot, but if you are planning say a 24v 100 watt mini inverter it might be about right for that. That first one sounds like your best bet. Cheers, Tony. |
||||
| gigabyte091 Newbie Joined: 15/10/2018 Location: CroatiaPosts: 17 |
I just realized that i take a picture of wrong transformer, here is the picture of 12V transformer. It came from 2 kVA APC UPS, it had two of this transformers wired in series.  |
||||
| nickskethisniks Guru Joined: 17/10/2017 Location: BelgiumPosts: 443 |
My experience with those transformers of apc is that they have a verry high idle current, I never tested it with an inductor in series with the H-bridge. It's not standard in the apc design, something I'm curious about  . |
||||
| gigabyte091 Newbie Joined: 15/10/2018 Location: CroatiaPosts: 17 |
I have these transformer about a year now, and i was always looking where i can use them, and inverter is perfect. I think UPSes were SmartUPS series or something like that, im not sure. I still have control board somewhere. They were also pure sinewave inverters. |
||||
| gigabyte091 Newbie Joined: 15/10/2018 Location: CroatiaPosts: 17 |
So while im waiting for parts to arrive, i started to think about aux power supply for supplying power to fan, both high side drivers (i want to avoid bootstrap if possible), low side drivers and power for EGS002 board. After thinking for some time, i realized i want something simple, reliable, and cheap ofc. So i created small zvs driver used to drive flyback transformers to create arcs. its just a first prototype made from part from junk box This is the output waveform  and some pictures of prototype, homemade transformer, core is the same as the output chokes in ATX power supplys but 3 times larger.  and converter itself  Only problem, to low operating frequency and high idle current draw, 400 mA at 12V and transformer is warm, but everything else is cool. transformer is 15+15 turns center tap and series inductor is 32 turns on output inductor core from atx supply. capacitor is 0.68 uF wima mp3-x2 capacitor, and series inductor is about 100 uH I will try to find smaller capacitor so i can increase operating frequency. This is basically royer oscilator circuit and operating frequency is resonant frequency of LC circuit. I think i dont have enough turns in transformer for this frequency |
||||
| Warpspeed Guru Joined: 09/08/2007 Location: AustraliaPosts: 4406 |
If you want small and cheap, take a look at the postage stamp sized isolated power supplies on e-bay. These are intended to work from 80v to 265v mains input, but will actually start up and run at perhaps 30v to 35v dc input, but of course any heavy load on the output will just overload it with such a low input voltage. But for something like an isolated mosfet gate driver supply that only needs a very few mA to work, these little supplies can work very well. No way will one be able to power a 12v fan though, with 48v input. There are some larger Chinese non isolated buck regulators for doing that. But for very low power, and full isolation, these small supplies are pretty efficient, low cost, and small. I run sixteen of these to power the isolated gate drivers to four bridge inverters from +100v dc supply. They will start up and run at 35 volts and all sixteen draw about five watts, or roughly 312mW per inverter which is not too bad. https://www.ebay.com.au/itm/AC-DC-Switch-Power-Supply-Module-Original-Regulated-Power-Supply-Module-Board-AU/32340086573 8?hash=item4b4c3197ca:m:miK9yiItgC8mXGtetavHy4Q Cheers, Tony. |
||||
| gigabyte091 Newbie Joined: 15/10/2018 Location: CroatiaPosts: 17 |
Okay, its one solution, but im more of a diy guy so first i would like to try to make my own converter. Maybe small flyback with multiple secondary, based on uc3842 as this royer circuit does not have ocp... make it universal input, about 60 to 20V so you can use it on 24 or 48v battery |
||||
| smithjack232 Newbie Joined: 12/07/2024 Location: United KingdomPosts: 3 |
It sounds like you're asking me to create an issue related to a 4007 diode Could you please provide more details about what kind of issue you're encountering or what specific information you need regarding this diode? This will help me assist you more effectively. |
||||
Grogster Admin Group  Joined: 31/12/2012 Location: New ZealandPosts: 9257 |
This thread is six years old..... Smoke makes things work. When the smoke gets out, it stops! |
||||
 Print this page |
