EDIT : Disregard that, I'm too old school. Last fusion was 1.6.4. ;P
[GregTech-5][1.7.10-FORGE-1355+][Unofficial but approved Port][Stable] Even GT5 Experimental is slowly getting stable.
-
-
So, quick question about Fusion: Is it possible to run continually a helium plasma fusion?
I mean, if my math is right, It would need about 94 electrolyzers + 133 centrifugues. Does anyone (sane) do that?
Or you guys run it at intervals?Edit: Randall is great!
Unless the recipes have changed costs since July, I think your math is a little off. I calculated things on July 8th (relevant post), and presuming you intend to electrolyze water for the hydrogen (not the cheapest in terms of EU, but easiest to get an abundant supply of, especially if not using the CodeChickenCore finite water setting), what I came up with was a total of 118 LV electrolyzers (which can be put into 8 processing arrays), 157 LV centrifuges (10 arrays), and 32 MV centrifuges (2 arrays).
-
Oh, I see, thanks.
About my math, well, I was at work, so I have missed something.
So, if isn't to ask too much, how much Deuterium and Tririum I should have before starting the fusion, to have a positive EU gain, (using plasma generators mark I, and considering only the fusion start and running process)?
I calculated about 20.000l each, but, as Mauve said, my math is a little off... -
I hadn't calculated that part before, but now I come up with a figure of 26252 L of each to break even with the startup cost, so over that would give positive EU gain. 37975 L each if you want to recover the processing costs of making the deuterium and tritium as well (though that's still ignoring cable losses, the costs of making empty cells to fill with oxygen when electrolyzing the water, and any costs of pumping the water)
-
I hadn't calculated that part before, but now I come up with a figure of 26252 L of each to break even with the startup cost, so over that would give positive EU gain. 37975 L each if you want to recover the processing costs of making the deuterium and tritium as well (though that's still ignoring cable losses, the costs of making empty cells to fill with oxygen when electrolyzing the water, and any costs of pumping the water)
Wow, way more than I though. Thanks again! -
Hello.
I running non-stop process of helium plasma fusion, then plasma goes into large plasma generator with large tungstensteel turbine.
tungstensteel turbine has turbine efficency 140 and optimal plasma flow 30000 eu/t -- i don't know how exactly this numbers (should) work, but since i've updated greg from 5.08.26 to 5.08.27 version
large plasma generator consumes MUCH more plasma, and now it consume more plasma than fuion reactor produces, so i run out of plasma very fast. In previous version reactor was producing slightly more plasma than LPG was consuming.i measured how much energy i get from stack of helium plasma cells in LPG with tungstensteel turbine in different versions
v5.08.26 ~366,7mil ~42k EU/t
v5.08.27 ~234,7mil ~63k EU/t
So, i think something is broken in .27 version... -
I did not limit plasma flow and energy output. I use ZPM dynamo hatch (LuV is not enough) and quantum tank with plasma next to the input hatch + pump cover.
Never thought there can be overflow... as i always saw expected numbers 30000*140%=42k EU/t in v .26 and before (optimal=maximum)
I'll try to limit flow with fluid regulators.
30k/4096k*1000=7.32
so in that case i have to inject 7 or 8 mB/t into LPG to get optimal flow? -
Its also worth noting that if the LPG wants 8 mb/t (optimal), then it will consume up to 10 mb/t if that's available in the hatch(es). The large turbines can always eat up to 25% above their optimal flow, although it does so very inefficiently. Its always best to run at optimal rate.
-
I did not limit plasma flow and energy output. I use ZPM dynamo hatch (LuV is not enough) and quantum tank with plasma next to the input hatch + pump cover.
Never thought there can be overflow... as i always saw expected numbers 30000*140%=42k EU/t in v .26 and before (optimal=maximum)
I'll try to limit flow with fluid regulators.
30k/4096k*1000=7.32
so in that case i have to inject 7 or 8 mB/t into LPG to get optimal flow?
You can fine tune the fluid regulators more in the fluid per second mode. 7.32 mB/t is close to 146 mB/s: a much closer approximation for free. -
I can only say for sure that the large Turbine in .26 was broken.
366mio is too much. 64 cells * 4mio EU * 140% efficiency = 358mio.
Then there is loss due to spin up time and transfer loss. The large plasma gen also has a relatively huge internal rounding loss. Finally, did you limit the plasma flow somehow? Not that the overflow takes away most of your efficiency bonus. I'm currently not at home, so i can't see how much the turbine should produce.
You can fine tune the fluid regulators more in the fluid per second mode. 7.32 mB/t is close to 146 mB/s: a much closer approximation for free.
jesus, I never even thought of this. This is brilliant. -
Nope, if LPG can consume up to 10 mb/t then it will consume 10mb/t first 14 ticks, 6mb in 15th tick and 0 in remaining ticks. so it will not be optimal flow
-
Crap then.
-
If you guys don't mind, I still have some doubts about the Fusion math. Mauve, or anyone else, can help me here?
I'm going to calculate costs to produce Deuterium and Tritium, 1000l each:
Electrolyzed 30.000 water, 10 process (45000 EU each) = 20.000 Hydrogen, 450.000 EU loss
Centrifuged 20.000 Hydrogen, 5.000 process (320 EU each) = 5.000 Deuterium, 1.600.000 EU loss
Centrifuged 4.000 deuterium, 1.000 process (1280 EU each) = 1.000 Tritium, 1.280.000 EU lossSo, 1.000L of Deuterium and Tritium, costs 3.330.000 EU.
That runs about 8 process on the Fusion, 16 ticks each plus 4096 eu/t loss, 524.288 EU. So, each helium plasma cell costs 3,854,288 EU (not counting cable losses), and generate 4.096.000 Eu at 100% Efficiency. Profit of 241.712 EU.
Now, question time (assuming the above is correct):
1) Using the first plasma generator, mark I, with 50% efficiency, I can't make it yield more power than costs to produce; Am I missing something?
2) How the startup energy is used? If I have only the 40.000.000 required, and run 30 cells of helium plasma, it will costs 15,728,640 EU (30.000/125 = 240 process, 240 *16 ticks = 3840 ticks at 4096 eu each). That means I don't have to feed the reactor until the end of reaction, that energy will be used instead?
-
Quoted from "Drawfox"
Centrifuged 20.000 Hydrogen, 5.000 process (320 EU each) = 5.000 Deuterium, 1.600.000 EU loss
Centrifuged 4.000 deuterium, 1.000 process (1280 EU each) = 1.000 Tritium, 1.280.000 EU lossThese output 4 L per process, so the centrifuges only need to run 1250 times (400k EU - my old calculations had it as 80k EU, so maybe that recipe changed) and 250 times (320k EU) respectively to make that much deuterium and tritium.
-
These output 4 L per process, so the centrifuges only need to run 1250 times (400k EU - my old calculations had it as 80k EU, so maybe that recipe changed) and 250 times (320k EU) respectively to make that much deuterium and tritium.
Doh, I was using ratios, forgot that... -
I redid my calculations using 400k EU as the deuterium production cost, and now the EU profit at 50% plasma gen efficiency is 353712 EU per cell, so it would take a little over 113 cells worth of plasma to break even with the startup cost.
I'm a bit confused myself as to how the startup energy will be used, but my guess is that it takes that much to initiate fusion, then as long as it remains supplied with sufficient matter, it keeps the reaction going, using the "Total" (65536 EU for the D-T reaction) per operation.
-
Jep, it needs the startup energy to turn on. After that it uses the EU/t of the recipe. After the Fusion stops, the startup energy is needed again.
I haven't tested this behavior so I'll just ask:
If the internal EU storage dips below the activation amount in the middle of a process does the process stop? If not could this be used to regulate the production of EU to the amount of EU being fed in? (useful for starting up)
