 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 : OzInverter OzCntrl PCB rev12 Sept2017
  Page 2 of 13   |
|||||
| Author | Message | ||||
Madness Guru  Joined: 08/10/2011 Location: AustraliaPosts: 2498 |
|
||||
Clockmanfr Guru Joined: 23/10/2015 Location: FrancePosts: 429 |
Hi Madness, Thanks for your kind offer, I have sent you a PM with my details. Regards what your circuit is doing. I think it is great what you have achieved. With the OzInverter project I really try to keep things as simple as possible with as few things as reasonable to reduce things that need re-setting or repairing. The OzInverter book and PCB's are going out all over the place and most ordinary folk do need guidance and simple instructions. So please do use me as a simple person who needs there hand holding a bit. Yes, Mad the super whooper Internet speeds are great, and I will sit down and get beyond the Blinking stage with the Ard/Nano. But so fast, fast, fast, all my Banks anti fraud software picked up that my Bank activity was very different from the last 10 years, and promptly stopped all my accounts. In the end I had to turn up in person at a Branch in The UK, with identity with photo and address showing, and my Passport, to prove I was who I am. Thanks 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. |
||||
| Clockmanfr Guru Joined: 23/10/2015 Location: FrancePosts: 429 |
Hi brac321, Thanks for that Link. Its interesting to see that there are ever increasing cheap modules/units now days. How is your OzInverter toroid coming? My apologies for the toroid core manufacture in the UK that I list. They have stopped retailing just the 0.3mm thick silicone Iron wound cores, and gone over to retailing the toroid's fully wound. But like other toroid manufactures they do not get any where near the size and winding we require for a good power and longevity OzInverter. 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. |
||||
| Madness Guru Joined: 08/10/2011 Location: AustraliaPosts: 2498 |
Hi Clockman, I understand what you are trying to achieve, eventually, I hope to have a menu driven system to set charge voltage etc. But I am not at that point yet, I am aiming for something along the lines of the menus of the Midnite controller you already have. Please keep in mind I am not focused on basic simple fool proof for myself. An alternative might be to design it so that there are trim pots to set the absorb and float voltages plus another for setting the absorb time. Otherwise, people would need to change values in the code and upload it. This could be a step too far for some. There are only 10 types of people in the world: those who understand binary, and those who don't. |
||||
| Madness Guru Joined: 08/10/2011 Location: AustraliaPosts: 2498 |
Have you tried Alibaba, I regularly get emails from them with Chinese companies that manufacture Toroids to your specs. There are only 10 types of people in the world: those who understand binary, and those who don't. |
||||
| Rhume Newbie  Joined: 25/08/2017 Location: BelgiumPosts: 3 |
Dear Madness, i'm think you can found it here. i buy ir2110 there. https://www.bifelectronic.com |
||||
| Warpspeed Guru Joined: 09/08/2007 Location: AustraliaPosts: 4406 |
Toroids, like house bricks, and lead acid batteries are heavy buggers. They may be cheap bought for cash from the toroid factory located right across the road, but shipping them half way around the world can make a very big difference to the economics. Seek out a manufacturer of the silicon iron strip wound toroids in your own country. You may get lucky, or maybe not. In Australia, Tasmania, and New Zealand, the only manufacturer in this rather remote part of the world are AWM Cores in South Australia. The other option is to do a sharp deal on something you can buy secondhand, strip and then rewind. Cheers, Tony. |
||||
| brac321 Regular Member  Joined: 30/11/2016 Location: SloveniaPosts: 54 |
Clockman, I already got my toroid, a year ago, from Germany... It is an heavy 7.5 kW "monster". Only need to unwound all the secondaries and wound new low voltage on ... In the moment just waiting on some components to arrive, to finalize your boards. Will also do some mods on OzCooling pcb. it is voltage dependent... P.S. SMD fets are still a live, think they will stay there   OffGrid 24/7: 250w black mono solar panels 6.4kWp, 1x Midnite Classic 150, 1x MorningStar MPPT 60, battery bank 840 Ah @ 48v, modified LF-8K inverter, DC/DC, hot water harvesting, etc. |
||||
| Rhume Newbie Joined: 25/08/2017 Location: BelgiumPosts: 3 |
More info Toroidal core for all power electronics Toroidal Cores Producing assortment General description Dimension min. (mm) max. (mm) D outer diameter 27 800 d inner diameter 17 500 h core heigh 10 180 Sh chamfered edge* 2x45° 5x45° * by customer request Application possibilities of toroidal cores ▪ measurement transformers ▪ power and regulating transformers ▪ inductors ▪ transducers BENEFITS OF TOROIDAL CORES ▪ Close flux path ▪ Low magnetic leakage ▪ Low humming ▪ Compact dimensions ▪ Light weight The table of toroidal cores qualities Quality Typical values Bmax ( T ) ** Heff = 3A/m Guaranteed values Bmax ( T ) Heff = 30A/m Colour marking *** N - normal 0,050 min 1,30 black S - standart 0,075 min 1,40 red SU - super 0,110 min 1,60 green SU-EX - extra selection * 0,130 min 1,70 white * quality SU-EX – by customer’s request ** typical values – by customer’s request *** colour marking - by customer’s request The tolerance of toroidal cores dimensions Inner diameter Acceptable deviation Outer diameter Acceptable deviation to 50 ± 0,5 to 50 ± 0,75 50 - 100 ± 0,75 50 - 100 ± 1 100 - 150 ± 1 100 - 150 ± 1,2 150 - 300 ± 1,2 150 - 300 ± 1,5 300 - 400 ± 1,5 300 - 400 ± 1,75 400 - 500 ± 2 400 - 5000 ± 2 500 - 800 ± 2,5 500 - 800 ± 3 800 - 1000 ± 3 800 - 1000 ± 3,5 Core height Band width Acceptable deviation to 50 mm ± 0,2 from 50 mm ± 0,3 Toroidal cores are supplied in different types of construction and in different types of quality according to used base material. Types of toroidal cores - toroidal core Drawing Toroidal core is basically wounded grain oriented silicon strip containing approximately 3% of silicium to the circle or other form. - toroidal core with chamfered edges Photo Toroidal cores can be chamfered on the inside and outside edges and make possible to wound the coil without using protective capping strip. Another advantage of chamfered edges is to eliminate possibility of breakage the banding and also to eliminate possibility of breakage the winding. - toroidal core with protective coating Photo Due to customer wish the cores can be supplied with protective lamination. These cores can be also in combination with chamfered edges. This brings a big advantage because you can wound winding without using protective capping strip. Also you don’t have to band the cores before wound base threads. Another big advantage of protective coating is hardening of the core therefore the cores are more resistant against mechanical influence and better stability of magnetic properties. - toroidal core with air gaps Photo Toroidal cores with air gap can be produced with exactly defined one or more air gaps. These cores are usually used for chokes, combined instrument transformers and also for transformers for audio technology (e.g. amplifiers). - toroidal core of special shapes Photo Toroidal cores of special shapes are produced due to customer requirements and can be cone, elliptical, ovoid shapes and so on. These cores are usually used for illuminating and measurement engineering. |
||||
| Clockmanfr Guru Joined: 23/10/2015 Location: FrancePosts: 429 |
Brac321, I do like boxes like that.  For the OzInverter with a good general 6kW rating. I find that a silicone iron wound strip core about, 200mm outside diameter, 100mm inner diameter and overall core thickness of 120mm, (2 x 60mm thick cores, stuck together, as it seems that the winding machine most manufactures use is limited to max 60mm). Weight of the silicone core is a good general measure when selecting other assorted cores or used cores. Weight for the bare core for the 6kW OzInverter should be about 18kg to 22kg. At 230mm outside diameter, 100mm internal diameter and at 140mm thick the core weighs about 35kg. At that weight a benefit is less copper windings, but one heads into the masochistic area of weight lifting. And at 35kg core its easy to damage the copper windings as you lift and turn the toroid unless you take special precautions for resting the toroid. I use 5 layers of 1.1mm thick rubber matt. 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. |
||||
| Clockmanfr Guru Joined: 23/10/2015 Location: FrancePosts: 429 |
Hi Warpspeed, Yes locally sourced cores are always best for saving on postage and packing costs. I was always surprised, that the core manufacturers go on about how much magnetic degradation there was when using to much resin/varnish when coating the bare cores. Is it really that bad. ? 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. |
||||
| Clockmanfr Guru Joined: 23/10/2015 Location: FrancePosts: 429 |
Hi Rhume, Very Interesting link for us Europeans. https://www.bifelectronic.com https://www.bifelectronic.com How much do you think for a silicone iron core ... 200mm Outside diameter x 100mm inner diameter and 120mm thick. Postage and Packing for sending around Europe? 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. |
||||
| Warpspeed Guru Joined: 09/08/2007 Location: AustraliaPosts: 4406 |
There are a few basic things to bear in mind when sizing a suitable toroidal core. The first is to decide what is most important to you, efficiency at flat out full rated power, or inverter zero load idling power. Or maybe some reasonable compromise. Maximum power rating only has to do with safe temperature rise, and these large transformers heat up relatively slowly. So there is no real problem pulling really high power for very brief periods as will be typical of a domestic inverter. Temperature rise has more to do with the wire gauge, and the length of wire rather than the core size. Although obviously a larger core can contain heavier windings which is where the final power rating comes from. Design of a commercial transformer is very different to the design of a home brew inverter transformer, there are some very different requirements and goals. So what the commercial transformer guys do has very little relevance, in fact you can get into quite deep trouble copying similar design figures. A commercial transformer design has only one real goal, maximum profit for the transformer company. So the basic idea is to use the absolute minimum of copper and iron and maximise the safe temperature rise under continuous full rated power. The same designer could not give a damn about magnetising current or transformer hum. They also do not care about any massive inrush current surge at initial turn on due to the flux doubling effect. Maximum power rating, minimum size, and least cost wins every time. The designers job depends entirely on how mean and how daring the penny pinching design can be made. To do this, the flux density in the core is run right up to saturation, which means fewer turns required per volt, which allows thicker wire in the available space, and in turn a higher current rating. Less winding resistance also produces better voltage regulation, and the final result is what you end up buying off the shelf. So what I want to discuss here are what might be more suitable design figures for an inverter transformer, which is a rather special transformer application. Flux density is the first major consideration. Commercial transformers using grain oriented silicon steel are normally run at about 1.5 Teslas (15,000 Gauss) which allows about a ten percent margin for the 230 volt mains actually being more like 253 volts. The core manufacturers AWM in South Australia suggest the absolute maximum design flux density of 1.7 Teslas, although that will produce an objectional core hum. Anything more than that and the core will definitely go into hard saturation. Note that transformer flux only has to do with the applied voltage, load current has no effect on flux density. Now there is a phenomena known as flux doubling. The flux in the core rises in either direction swinging through zero twice each mains cycle. If you suddenly interrupt the incoming power, there can be a high residual magnetic flux remaining in the core. (it becomes strongly magnetised). If you then switch it back on, you may be unlucky enough to be on the part of the mains cycle that tries to drive the magnetic flux in the same direction as the residual magnetism and of course it goes into violent magnetic saturation. In theory the flux might try to go to double, but that is just not possible if we have designed our transformer to use ALL of the available flux swing the core can support. These transformers can easily blow fuses or trip circuit breakers when switched on. No real problem UNLESS this transformer primary is being powered by some mosfets. It can be done, but it requires a very slow soft start, and its hairy and very dangerous. So we absolutely must lower the flux swing in our transformer for inverter operation. The lower the better, but the 1.5 Teslas the transformer manufacturers use is rather high for us. How much lower is entirely up to you, but for example if you go as low as 0.85 Teslas, the flux could in fact safely double to 1.7 Teslas, and there will be no possibility of any inrush surge at power up. That would be unreasonably low and just not practical. But its something to keep in mind. A second reason for lowering the design flux density in an inverter is that it vastly reduces the zero load magnetising current. You can test this yourself with a variac, by measuring the magnetising current with an amp meter in the primary. Try it at 250 volts, 230 volts, 210 volts and so on down. You will find that the no load primary current at 250 volts will be huge. Every reduction in voltage gives an even larger reduction in no load current. There is tremendous advantage in reducing the flux density even slightly, as it will have a disproportionately large reduction in no load loss which is where our inverter spends most of its life, especially at night under battery power. This is much more important for off grid than for a grid tie transformer. Grid tie transformers can be run at higher flux density and generally are to produce a smaller cheaper transformer. An off grid inverter transformer will definitely benefit from a lower flux density, as it will save a lot of night time battery amp hours (which are expensive). If you are stripping an old toroid and count the turns and work out the volts per turn for that core, realise that it may not be ideal for your new inverter. If you can fit a few more turns on, do so. An extra ten or twenty percent higher turns on both primary and secondary would be well worth doing if there is room. Much better to measure the cross sectional area of the toroid and come up with a square cm figure, then from that you can work out the turns per volt at 50Hz for whatever flux density you decide upon. Something around 1.2 to 1.3 Teslas might be about right. Less would be better still, but fitting in all those extra turns becomes difficult unless you use thinner wire. Thinner wire means more heat at flat out full power. But how often do you really need that ? For a domestic inverter, especially off grid, its a much better tradeoff to reduce idling power. It can still produce massive short term power which should be adequate for motor starting. Do we really need full rated power continuously ? Here is a handy calculator for working out transformer flux density: https://www.electricaltechnology.org/2014/02/maximum-flux-density-bmax-calculator.html A typical grid tie toroid might look something like this: Voltage 230v Frequency .00005 Mhz (50Hz) Turns 230 Area 30sq cm (60mm x 50mm core cross sectional area) Flux density 1.5 Teslas (15,000 Gauss) That is o/k, but a lower flux density for off grid would be better. Cheers, Tony. |
||||
| Madness Guru Joined: 08/10/2011 Location: AustraliaPosts: 2498 |
According to that calculator for the Toroid, I made following Oztules method I get. Maximum Flux Density(Bmax): 0.7849 Tesla There are only 10 types of people in the world: those who understand binary, and those who don't. |
||||
| Warpspeed Guru Joined: 09/08/2007 Location: AustraliaPosts: 4406 |
That is outstanding. It sounds like you doubled up on the core cross section while keeping the original turns count. The no load idling power should be spectacularly low, and not the slightest chance of blowing up mosfets from any turn on surge. Should be dead silent as well. Cheers, Tony. |
||||
| Madness Guru Joined: 08/10/2011 Location: AustraliaPosts: 2498 |
Just realised I suffer from CRAFT, I went back and checked the number of turns, correct figure is Maximum Flux Density(Bmax): 0.8810 Tesla. It is close enough to silent with 230V AC connected to the secondary, however, it does buzz working as an inverter. There are only 10 types of people in the world: those who understand binary, and those who don't. |
||||
| Warpspeed Guru Joined: 09/08/2007 Location: AustraliaPosts: 4406 |
That is still exceptionally low. I am in the process of building up a PWM inverter myself at the moment. I don't really need it, but it will be an interesting learning exercise. My battery is nominally 100v consisting of thirty 50Ah Winstons. Plan to use two completely original Inspire toroids. The Inspire grid tie uses a 200 volt dc bus, but what I plan to do is use two of these toroids just as they are. Powered off only 100v dc I can only get 115v across each secondary, but with two toroids I can get 230v with both secondary windings in series. Its a very lazy way to do it, but I just could not be bothered with the whole toroid rewinding drama. It does have a couple of advantages though. Flux drops from 1.6T to 0.8T because all the voltages are now halved. No load losses should be very low, and I can mount both toroids flat side by side in my metalwork which reduces the height. Still working on a very elaborate BMS which is really what this is all about. The PWM inverter is just a minor side project and takes second place to the BMS and battery charger. Cheers, Tony. |
||||
| Madness Guru Joined: 08/10/2011 Location: AustraliaPosts: 2498 |
I agree winding (and unwinding) is a PITA. The smaller inverter I have built which has a single Aero-Sharp 3KW toroid draws around 25W at idle. I unwound the outer windings added 22 turns for the primary, only took 4 or 5 hours. It will handle 4.5KW for 10 - 15 minutes, combined with GTI it would do everything I want with a little thought about what turn on. I don't know how many turns are on the secondary but I am more than happy with it's performance. There are only 10 types of people in the world: those who understand binary, and those who don't. |
||||
| Warpspeed Guru Joined: 09/08/2007 Location: AustraliaPosts: 4406 |
Doing it all on a shoe string means salvaging and using surplus junk, but the PITA aspect may still make the cost savings very worthwhile for home construction. As long as its approached as fun and a hobby, that is fine. It can really appeal to the masochist. I was doing all this transformer stuff myself for a day job for years, and it just looks to me like a lot of really hard work. While Oz and Clockman have made the circuit and circuit board aspect of all this fairly straightforward, the magnetics part still leaves a bit of room for further ideas and experimentation. If I was going to build myself a multi kilowatt PWM inverter, I think I would pony up the dollars and buy a pair of brand new split C cores. These are exactly the same tape wound grain oriented silicon steel material as the good toroids. They are just machine wound on a rectangular mandrel instead of wound on a round mandrel, then sawed in half, and the faces precision ground to a mirror flat finish. You get supplied a fiberglass coil winding bobbin to suit, which is dead easy to wind without all the agony and grief of winding a toroid. Flat rectangular copper bar can be wound straight onto the bobbin for the primary. Its fast easy and much more space efficient than round welding cable. The magic of the tape wound toroid has more to do with the special material than the shape. Split C cores are exactly the same stuff, but are much easier to work with. I have no idea how much these C core halves cost new, but they are available in all sizes off the shelf up to a 290Kg pair of core halves. Cheers, Tony. |
||||
| Madness Guru Joined: 08/10/2011 Location: AustraliaPosts: 2498 |
Although it is a PITA, the cost and the performance of the end result is worth it. Would I want to do it all day every day? no way, but it is worth the effort to do a couple. I certainly would not go out and spend hundreds of dollars on a core plus the wire when unwinding what you call junk works fine. Is it perfect, no it is not in my opinion but for me a far better option than spending around $10,000 on an inverter that I would always be worried if it failed and the cost plus inconvenience. There are only 10 types of people in the world: those who understand binary, and those who don't. |
||||
| Page 2 of 13 |
|||||
 Print this page |
