Big OzInverter, torroids question?

[ClockmanFrance]

Making a Big OzInverter

Hi,
I am putting together my No 2, Big OzInverter.

I have a couple of questions about the toroid.

Yes, I have read most of all ‘oztules’ 60,000 words of wisdom, he has gratefully posted on 3 forums. I thought that I had understood the toroid concepts. But sadly my head is still spinning a bit.!

My No 1, 6kW OzInverter, 48vdc to 230vac 50HZ, works fine and does everything I wanted. ........The cores were 2off 190mm outside diameter x 90mm centre hole and each core was 60mm high. These were stacked together, so the rectangle section through the core was 50mm x 120mm high. 4off /4 in hand, 1.8mm diameter at 118 turns Secondary, and 50mm/2, 14 turns, Primary.

My No 2, OzInverter, will be for the final installations here in Normandy, and will incorporate a better metering system. The cores will be 130mm outside diameter, 100mm diameter centre hole and each at 70mm high, stacked together that gives me a rectangle section of the core 65mm x 140mm high. That’s just over 30% more core mass.

Question, ….. Does this 30% more mass mean a proportional increase in the secondary windings, ie, should the 4off/4 in hand 1.8mm diameter, now be 158 turns. Or can I get away with less turns and save copper losses?

Or is the extra copper/turns, only proportional to the outside diameter increase and not the height, so the increase of turns would be about 20% in this case.

Question, ….. I am using 75mm/2 for the Primary, again is there a proportional rise in the number of turns to say 18 turns. ?

Thank you.

[oztules]

Lemme see...... I make that as about 9100mmsq cross section... so ... and here I can only go rule of thumb....9000/2800=3.25

So I figure about 3.25 volts per turn.
So the secondary will be about 73 turns, and the primary will be about 9 turns.

How that would be worked out mathematically I can only postulate, but the in the real world measurements I have done, it would seem to indicate that roughly 2800mmsq of cross section at 50hz will yield about 1turn/volt.... so 9000mmsq would be about 3.25 volts/turn.

If we wind for a few more volts than we want, our saturation is further away... so say 240 or 260v for a 220v system gives us leeway and lower magnetising current  ... from there about 1:8  primary : secondary.. and we have a workable start point.... at least that how I would tackle it.

Provide you have a crane close handy to hold the thing for you, it will be easy to wind.. big hole and plenty of room. It will be important to map out the wire sectors in advance.. or it will all end upon one side

Your core manufacturer will be able to give you exact figures I suspect.

.............oztules

[ClockmanFrance]

Many thanks oztules.

Crikey I was going 100% the wrong way, I was adding turns/copper when I should have been subtracting.
Good job oztules is guiding me along.

"If we wind for a few more volts than we want, our saturation is further away... so say 240 or 260v for a 220v system gives us leeway and lower magnetising current  ... from there about 1:8  primary : secondary.. and we have a workable start point.... at least that how I would tackle it."

Thanks,.... I will go for 80 turns on the secondary and 10 on the primary and see what I get when I test the secondary.

"Provide you have a crane close handy to hold the thing for you, it will be easy to wind.. big hole and plenty of room. It will be important to map out the wire sectors in advance.. or it will all end upon one side "

My last set of cores were 20kgs, this new set will be 35kgs, so I will do a soft pad that the toroid can sit on that is on the cut out bench, while I wind, and lift/rotate about 6 times each winding.

Yes, I will do some templates and see how the spacing of the secondary works out, I might even add another so I have 5off/5 in hand, off 80 turns at 1.8mm diameter, depends on how the wires lay.

I am looking forward to this build...........
 

[frackers]

Hmmm - this is getting interesting!!

Based on the 2800sqmm for 1V/turn and on the 1st layer being next to the core, and the cores I've made  (210 outside, 90 inside, 25.4 thick)  then:

 4 of my cores would require 111 turns for a total of 22.5m of wire per layer, 14 turn primary
 5 of the cores would require 89 turns for a total of 22.5m of wire per layer, 11 turn primary

I think this pushes me to the 5 core solution as it doesn't increase the copper loss, it improves the cooling ('cause its bigger!!) and makes more room in the centre for thicker primary due to less turns.

No excuses now - I have 1kg rolls of 3 thou and 5 thou mylar and only 1 more core to make (I'm making 10, 5 for me, 5 for a mate).

[frackers]

Based on the 2800sqmm for 1V/turn and on the 1st layer being next to the core, and the cores I've made  (210 outside, 90 inside, 25.4 thick)  then:

 4 of my cores would require 111 turns for a total of 22.5m of wire per layer, 14 turn primary
 5 of the cores would require 89 turns for a total of 22.5m of wire per layer, 11 turn primary

I think this pushes me to the 5 core solution as it doesn't increase the copper loss, it improves the cooling ('cause its bigger!!) and makes more room in the centre for thicker primary due to less turns.

Well, I've made a 5 core toroid, it has come in at 25kg so there is certainly some metal in it! Could someone check my calculations as I just came up with different answers to those above!

Core size 210mm outside(O), 90mm inside(I), 127mm high(H).

Single turn is  H*2 + (O-I) = 374mm

Cross section is ((O-I)/2) * H = 7620sqmm

so using 2800sqmm per volt and aiming for 240v secondary and 30v primary
Secondary Turns = 2800/7620*240 = 89 turns - maybe use 90?
Primary Turns = 2800/7620*30 = 11 turns - maybe use 12?

One layer of the secondary is = 374* 89 = 33.3M (again rounded up).

I have 12kg of 1.8mm wire so based on the density of copper being  8930 kg per m3, I have a total of 12000/(.9^2*pi*8.93) = 528m
Possible layers = 528 / 33.3 = 15.9 (i.e. will do 15 full ones). I was planning on 4 so I have enough wire!!

The last couple of weeks has been making sure the cores are flat, sealing them with exterior polyurethane varnish then stacking them with polyester material (raided from the wife sewing room - wadding that compresses down to 0.2mm) that allowed me to get some moisture into the joints made with Durabond Aquadhere polyurethane glue. I chose polyurethane as its good for up to 140C and if I made a mistake they will split down again unlike epoxy.

[oztules]

Yes I figure about 2.7v/turn
 
So 90 turns and 12 will work, or even  a few more to get the magnetising current down, as you have a heap of steel there to coerce into action.
I would probably go for 96:12.

...............oztules

[frackers]

Just started on the first layer of mylar tape but I think I should have something to cover the 'sharp' edges of the toroid - it looks like some sort of fibreglass strip in http://www.anotherpower.com/board/index.php/topic,1044.msg10166.html#msg10166 but since I've started from cores that used cardboard, what would be a suitable alternative?

Some sort of thin plastic right angle extrusion with a rounded outside edge would work for the outer faces but not quite sure what would work for the hole in the middle!

[oztules]

Glass matt is what they use... but I guess cotton will work just as well with some sort of binder ( polyester fiber glass , epoxy etc).
Anything just to make the wire not able to be bruised and exposed to the core... ie provide some solid radius to the edges.

...............oztules

[frackers]

I'm tempted to just run an angle grinder round it to take off the edge - maybe use a mounted stone for the inner edge. I'll try that on a spare core and see how it comes out.

[frackers]

I'm tempted to just run an angle grinder round it to take off the edge - maybe use a mounted stone for the inner edge. I'll try that on a spare core and see how it comes out.

Tried it on a spare core and not very good - the material is quite soft and I don't have a fine grinding wheel so rather too tatty round the edges!!

I did succeed with some fine irrigate pipe (what I call spaghetti pipe) split down the middle and held in place with tape until the first layer of mylar went on to hold it permanently. Looks pretty good!

Not having taken a toroid apart, I've no idea how many layers of mylar I should be using. I'm using 5 thou for this first core insulation layer and then I'll move over to 3 thou for the inter-winding layer separation. I'm doing a 50% overlap so I have a double thickness of mylar but no idea if that is all that is required

Can someone who has stripped a toroid give me a clue or 3? Progress so far can be seen here (weekends are not good for uploading pix!) http://gilks.ath.cx/gallery3/index.php/Building-a-PJ-inverter

Cheers

[ClockmanFrance]

I use this on the edges.

80 grit flap wheel.

It works well and leaves a smooth small radius, with no roughness, so the copper wire forms a neat non stressed corner.

From the book....

" I normally use 0.055mm, or 50 micron thick mylar tape its just right.

I half lap the Mylar tape on its self on the exterior of the toroid, and I do 2off complete wraps as a minimum, with the super thin breaking stuff I have salvaged I did 3off wraps, Waste not.
Don't forget the epoxy resin on each secondary winding before you cover it with Mylar tape. I like the thin viscosity stuff that I mix and lightly paint on.

 I have some thin Molinex sheet that I put under the whole toroid to collect any excess resin.

 The following evening turn the toroid over, peel off the Molinex sheet, Mylar tape on the hardened, but still soft, epoxy resin coated winding, and then next copper wire winding goes on.

I use a small length of mylar adhesive tape here its yellow colour, to hold the mylar wrapping tape in position.

I get some mylar from the USA...... http://www.ebay.co.uk/itm/181696004499?_trksid=p2060353.m1438.l2649&ssPageName=STRK%3AMEBIDX%3AIT
 ..... its 0.125mm thick, 120micron, but i found that it is just too thick.  Especially when its not flexible enough to allow the next winding of the secondary to fit down in the gap of the previous winding, especially on the internal centre hole windings. But it does come oval shape and goes down a 62mm dia hole okay without having to unwind. And it is ideal for the last winding before the Primary goes on."

Trust this helps 'frackers.'

Toroid first wrap with 50 micron, that's 2 layers thick.

[oztules]

Frackers, I like what you have done thus far.... 2 layers as Clockman says is what I have done also.

.................oztules

[frackers]

Frackers, I like what you have done thus far.

Thanks John

Shopping for paintable epoxy tomorrow so I can finish off the first secondary layer. More pix in the gallery on my website.

[frackers]

(yet) Another question - does the fan on the 48v units run at 48v or 12v? My mate Andy who has scrounged up nearly all the bits for this build managed a few rack shelves full of 48v 115mm Papst 5w fans with tacho outputs

Assuming the fan runs at input volts I can put one on the control board and put the other 2 I have (more available!!) on whatever Atmel/STM micro I use to monitor the system with. Perhaps PWM them to keep the noise down if full wack is not required!

[ClockmanFrance]

On the cheap 48v PowerJack 10000 inverter I purchased, the two 120mm fans were 24v in series with their own heat sensors. The PJ safety fan was 12v. Their is a socket on the control board that gives 12v and it runs my 'OzCooling' circuit with 2 off 12v 120mm fans.