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.
Grogster Admin Group Joined: 31/12/2012 Location: New ZealandPosts: 9308
Posted: 12:03am 21 Jun 2016
Copy link to clipboard
Print this post
@ Madness - DEFINTELY still want them!!!!!! Have just replied to your PM - was away all day........
@ BrianD - Yeah, that has been my experience too. The only way I have found to do them in NZ, is to have your own custom toroid wound by the factory, but as others here have been commenting - that's still pretty expensive. Madness MIGHT be able to help you too. Flick him a PM.Smoke makes things work. When the smoke gets out, it stops!
Madness Guru Joined: 08/10/2011 Location: AustraliaPosts: 2498
Posted: 12:48am 21 Jun 2016
Copy link to clipboard
Print this post
If there are any other people wanting transformers please let me know ASAP. It is a 3 hour return trip to get them, I will be doing another trip very soon but don't want to do any more after that.
PM Sent to BrianDThere are only 10 types of people in the world: those who understand binary, and those who don't.
@Mad sorry just to digress a little from the topic but i guess still the same.
@Oz please does this calculation also workout for the E and I transformers. can i adapt this method to get the volt per turns.
secondly how do you determine that this size of torroid can handle this much power?
thanks
oztules Guru Joined: 26/07/2007 Location: AustraliaPosts: 1686
Posted: 12:03am 22 Jun 2016
Copy link to clipboard
Print this post
tunde.
Different laminations ... different effects.
You need to find it out empirically.... ie if it has a ac winding you can use ( eg 240), then use it, wrap the ten turns around the center core, and measure and do the simple maths to get turns/volt.
From somewhere in my past 4 sq inches per volt @50hz rings a bell, but measure it.
Power in a transformer is a simple function of how much sag you can tolerate, and how much heat you can stand/or cooling is available.
A small transformer can do big job if it is cooled ( liquid even ), or if it's duty cycle is low... eg electric fence transformer may do 300000 watts, but pulses for only .000010 seconds each second. It weighs a kilo or so.... yet a 300000w transformer with 100% duty cycle would be huge.
So it is a compromise of all those things....
You need thick wire to keep copper losses down, and thick wire needs bigger winding window... and you can see where this is going... bigger is better.. and costs more in every way.
...............oztulesVillage idiot...or... just another hack out of his depth
Madness Guru Joined: 08/10/2011 Location: AustraliaPosts: 2498
Posted: 12:53am 22 Jun 2016
Copy link to clipboard
Print this post
Recently I saw a guy on Youtube with a transformer in a bucket of transmission fluid getting much higher power with little heating.
Has the big cold hit you yet OZ? We are supposed to get down to 6 on Sunday up here in QLD, but that is probably warm for you.
GaryThere are only 10 types of people in the world: those who understand binary, and those who don't.
jack fork stik Newbie Joined: 08/12/2012 Location: AustraliaPosts: 34
Posted: 05:24am 22 Jun 2016
Copy link to clipboard
Print this post
im thinking of filling an inspire 2.5 box with oil and using 1x 2.5kw tranny as Oztules has often mentioned it. my old single/3 phrase welder is oil cooled and still has a ton of power dispite being yrs older then me [now 57yo].i remember seeing that welder as a kid as my dad took odd jobs to the owner of it many times and i got at an auction for $30 complete with old leads i sold for scrap [got my money back].when using it with 8 gauge rods for an hr the 10 litres of auto trans oil i replaced the oil with doesnt get much above luke warm and that was in summer!
ltopower Regular Member Joined: 08/03/2019 Location: United KingdomPosts: 64
Posted: 11:39pm 10 Mar 2019
Copy link to clipboard
Print this post
I realise this is an old thread.... I have been reading up over the last several months and going through the hundreds, if not thousands of separate posts on this forum and others regarding inverter builds.
I have a question, because I have been trying to do all the math behind all of the rules of thumb and see how much can be worked out. I like numbers...
This one had me stumped for a bit "The 2800 was derived from simple measurement" and I have started to wonder if this was based on a transformer with an agressive design of 1.6 Tesla or a really good quality core at the time.
One formula to work out the turns per volt is given by :
=10000 /(4.44 * Hz * Tesla * cm2)
When using a 2800mm2 core the flux density is 1.6 Tesla.
Reducing the flux density to 1.5 Teslas (which is a standard commercial transformer flux density for toroids for one manufacturer) could reduce the idle losses by 15%....
The reason that this may be of interest is where the inverters are back fed and the resulting voltages climb with the charge as this would increase the losses and push closer to more saturation.
Does this sound wrong ?
Warpspeed Guru Joined: 09/08/2007 Location: AustraliaPosts: 4406
Posted: 12:11am 11 Mar 2019
Copy link to clipboard
Print this post
Commercial transformer designers take a whole different approach to transformer design. For them its a business, and the idea is to manufacture the smallest transformer with the least amount of steel and copper at the lowest cost.
Many transformer applications are for inside equipment, and will never be run at anything other than full maximum load.
So what I am getting at here, is that for them the most efficient transformer is the lowest COST solution. They do not give a damn about no load idling power, or even how hot it gets at flat out full load. Kva per dollar is the holy grail.
The generally accepted flux levels for commercially wound toroidal transformers is around 1.6T, and that allows a bit of allowance for a slightly high mains voltage.
And many budding inverter enthusiasts have reverse engineered commercial designs and just assumed 1.6T is the correct design figure, and argue that, as just about all transformers work at that level it must be correct !
Inverters are very different beasts, and there is a lot to be gained by running our toroids at far lower flux densities. For a start, our inverters run continuously, and for a domestic application usually at a small fraction of the full rated power.
Battery capacity is a very expensive commodity, so anything we can do to reduce idling power will be a giant step forward. If we are recycling old or free material, and size and weight are no deterrent, we can gain hugely by lowering the flux density to around 1.0T
Try it yourself. Get a commercial toroid with a 230v primary and measure the idling power at ten volt steps between about 240v and down to 150v or so. That might roughly correspond to 1.6T at 240v and 1.0T at 150v. Record the idling Va at each voltage, and you will be astounded how lowering the flux density reduces idling power.
Do it any way you want, but if you can fit the extra turn on, without reducing the wire size greatly, it will make a very nice inverter transformer.Cheers, Tony.
ltopower Regular Member Joined: 08/03/2019 Location: United KingdomPosts: 64
Posted: 07:23pm 11 Mar 2019
Copy link to clipboard
Print this post
BTW, warpspeed, hat tip, really appreciate all of the past posts, spent many hours reading through lots and lots and lots and... !
The 1.6T now makes full sense. I was starting to go round in circles with a spreadsheet around 6am, birds tweeting outside trying to work out all the variables for a transformer.... down the rabbit hole...
I had been thinking about the flux density and the SQRT(2) peak of the waveform and the genral acceptance of some degree of saturation at the peak. For a commercial at 1.6T, which would be saturated at the peak of the waveform, would we need to reduce all the way down to 1T to obtain the majority of the worth while reduction in losses at the expense of windings ?
The experiment of stepped voltage reduction and measurement is sounding like a curiosity must.... will have to borrow my dads variac on the next trip..
I have had a look at a number of different steels and the watts per kg of core losses at 1.5T and they seem to vary a bit between 0.73W/kg and 1.0W/kg for electrical steels (Tata Steel - Unisil). Even larger variation for Unisil-H. They provide figures for 1.7T but not below 1.5T.
I have a new transformer (commercial buy.. wind my own on the bucket list) that I'm hopefully picking up soon, 6kVA rated with core losses around 20-25W. All in just under 31kg. That is, as far as I can say, also under 1.5T, so wondering what mass or flux could get get lower..
The reason for the curiosity is that I have a 25kVA 3 phase 255kg 415/240V EI Style transformer (laminations and not a nice toroid) sitting, waiting, for a time to buzz again..
The turns ratio is 1.73 as a 415/240
Using the 415 in series I can then get 5.19 with the 240V in parallel, but unsure as to the full impact of the much lower flux, which may be 18% of the original design level ? Winding resistance on the 415V series side will likely be be horrendous (35A rated) ?
If I then removed just over 35% of the 240V windings I should then get around 8:1
The 240V windings are rated around 60A, so 180A in parallel at 29V ?
The flux density would be very very low in comparison, but still struggling to work through the full implications... say the losses are down to 0.1W/kg at 0.4T I still end up with 25W of losses and only around 5kVA wire rating ? Or would the transformer be de-rated further due to reduced flux linkage ????
I'm still rowking my way back out of the rabbit hole but I'm lost at the moment...
Warpspeed Guru Joined: 09/08/2007 Location: AustraliaPosts: 4406
Posted: 10:14pm 12 Mar 2019
Copy link to clipboard
Print this post
We don't really need to worry about the voltage peaks, its really the average that matters. Its all about how many 'volts x time' before core saturation sets in. Is why inductors saturate proportionally earlier at lower frequencies, the time part is just as important as the applied voltage.
Magnetizing current is an ever more steeply rising curve, until it goes totally vertical at full magnetic saturation. How steep it actually is, depends where you are on the curve and the quality (permeability) of the steel.
You may find that reducing the flux level by 10% can lower the magnetizing current by 30% a very worthwhile gain. If further down the slope, lowering the flux by 10% might only gain you 15% reduction.
Its perfectly possible at 1.0T to build a 4.5Kva transformer that has a magnetizing power of only 20Va.
Its also true that although the magnetizing watts per Kg of steel reduces with reducing flux density, going to a larger and larger core to do it can eventually result in no more net gains. But the whole thing would then be impossibly large and that is never going to be a problem for us.
There is a further big advantage of very low flux density. If we go low enough flux doubling ceases to be the destructive beast it is a higher flux density. The turn on current surge of commercial toroids can easily blow fuses and pop circuit breakers. We certainly don't want to pop our mosfets. Soft start is supposed to be a solution to that, but its still pretty dangerous to rely completely on some soft start mechanism to prevent your inverter from blowing up.
Reducing the flux density is a far better and safer solution. If there is no massive turn on current surge its one less thing that can possibly go wrong with your inverter.
Its really just a case of stepping up one or two sizes in core size, to provide more space for the extra turns on both primary and secondary. That is just something the commercial transformer guys would never do because of cost.
But we can make a very efficient, low idling current, cool running inverter transformer, although its going to be huge compared to "a bought one".
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.
Print this post
