No.... if we lost 4 volts@120 amps, we would light up Auckland
There is no exact science here... just back of the cornflake packet stuff.
Firstly, we need to recognize that the voltage will not be 48v... but anywhere from 46-60v... so the range is quite large. We need to wind for the worst situation where we are low on volts, and are running hard... so we under wind from where you thought, and it seems that around 8:1 for a 48v system is about right to cater for all the losses and make sure we are not going to drive the current to high trying to get 240v rms out from 46v at full load.... the 240v is actually 320v peak as you suggest, so the rms primary is the batt voltage, and the peaks we need at full power of the cycle is 320, so as you say 320/46 is nearer to 7:1... now some head space... and we get to 8:1... and for 24v we get 16:1 etc etc..... yep... gosh I'm scientific about this....
The cores i have are around 3600sqmm from memory for around 1.14v/turn... so 3000 will be near to 1v ( bit less perhaps, but easy to remember)... it is all rule of thumb stuff The figure is higher for EI units.... but as a very rough start line, it should be close enough. You won't find any definitive area/turns figures for the most part, as it is all a bit rubbery, and relies on knowing the steels parameters... which we don't.... but this chinese stuff is in that ball park. Your stuff may be better still, as it is likely old, and likely from a power station or similar govt enterprise in the day.
So to get that information, you can wind a core with x turns... say 12, and hit it with 12vac 50hz, and see if the current is very low... then increase V until you see a knee in the graph, and backwards again. You should find a point where current increases faster than the voltage increase, and this will be near the saturation area, where the inductance is not able to get enough reactance to cancel the EMF easily. At saturation the core behaves like air.....near 3 orders of magnitude worse than the steel... We want the back emf to cancel the forward voltage without using more current than we need.
Remember, with out the reactance due to the inductance and frequency.. the 12 turns will just be a near zero ohm resistor.... and burn up.... as would an air core at this frequency. does that make sense? It is only that the core increases the inductance, that enough reactance will be achieved at 50hz to limit the current..... thats why impedance is measured in ohms. It impedes the flow as a resistor would, except the resistance to flow is dynamic in nature... it changes with the frequency... resistors generally don't ( high frequency mixed with wire wounds do though).
So yes, I'm guessing that at if you can get the 7400mmsq, the 114:14 will work fine, probably better than mine as you will have a few more sqmm... it may be worth a turn or two extra without the turn... lower iron loss perhaps... sheer guess work, but I think we are very close to the truth.... only experiment will tell... I expect it will work fine
To get the thick primary in there... 14 turns will be difficult, maybe more iron, and less turns will allow you get to a stage where the turns are possible with the smaller hole you have.... note my photo. 14 turns of wire took up the whole hole with a single layer.... more would have made it messy, and hard to get a bolt through to hold the thing down. I reckon 20 would have been very very difficult for even this one.... 57 :7 would be easier to wind the primary... a dog for the secondary, and increase the iron loss..... which way to go?
Edit.... the other thing with stacking... the further we get away from a square cross section, the more wire we use for the same result.... ie a square has the shortest perimeter of all the quadrangle shapes for the maximum cross section. As we elongate, we enclose less area for the same perimeter length.... ie longer perimeter to enclose the same area....... just something else to confuse the issue. They do make circular cross section torroids for this reason, but unless your spending someone else's money, you won't afford them.
...................oztules