Optimized Steam Engine Setup

[BlakeMW]

Well that's bad. If placement order affects this, why doesn't replacing offshore pumps and pipes make them split more evenly?

That's due to directional biases, water prefers to flow more in some directions than others which can cause the flow from one offshore pump to block the flow from another by completely filling the pipe before the other offshore pump has the chance to add water. It happens because factorio does not have a realistic pressure model. I'll try and give a graphic example:

Water from pump 1 wants to go up.
Water from pump 2 also wants to go up.
Once pump3 gets its turn, its water can't get to the bottom boiler string, because the up-biased water from pump 1 and 2 has blocked it. Result: Pump3 is underutilized and the bottom boiler string is starved.
The directional biases here dominate over placement order bias.

That's not a perfect description of what happens, but I think something like that does happen. Remember: Not a pressure model. Don't expect it to act like water.

Btw something else is that small pumps can do some seriously weird things. I have setups which definitely show small pumps causing water to teleport ahead - actually skipping a bunch of pipe segments then backfilling. This behavior is highly directional (i.e. tends to happen only in certain directions) and it's difficult to explain, except as some weird hack to make small pumps work "better". I haven't come up with setups which definitely show small pumps causing problems, but I wouldn't be surprised if small pumps in some cases allow water to skip over boilers and not get heated. It may also be a similar phenomena to offshore pumps in empty systems, where the water seems to rapidly flow to the end of the system, too rapidly for the boilers to heat it properly. Small pumps are always fine after boilers though (although of dubious usefulness except as valves).

[Aru]

Oh, I meant how much stacking it in rows of 15 engines each affects stackability. Wasn't talking about perimeter. Would the main belt still only be able to support 10ish rows or would it support more?

Also, I want to confirm if I'm using your quadratic equation correctly. If I have 15 steam engines and 20 boilers, each row would require 2.16 burner inserters to prevent slowdown right?

And also, I think all burner inserters would have to be placed on the closest row to the coal source to prevent slowdown to the fullest. However, the fact that coal comes much slower shadows the slowdown from not having enough burner inserters, because it takes quite a while for all boilers to fill up with coal on a normal belt. Can you factor this into your math and recalculate? It would appear that it's possible for just one burner inserter per row to be enough to "prevent" slowdown, because any amount of slowdown is faster than the time it takes to fill boilers.

I don't remember the calculation for certain, but I think a yellow belt of coal can supply 13.392*8000/2/510 = ~105.0353 engines, or 13.392*8000/2 = ~53,568 kW.

Here's some more python (2.7):

Code: Select all

def quad(a,b,c):
    d = (b**2 - 4*a*c)**0.5 / (float(a)*2)
    e = -b / (float(a)*2)
    return e + d, e - d

def burners(engines, inserters):
    engines, inserters = float(engines), float(inserters)
    a = -390/510./engines*0.866
    b = -a*inserters + 390/8000.*2
    c = -510*engines/8000.*2
    return quad(a,b,c)

>>> burners(15, 20)
(2.1608155020954167, 20.047614059105506)

Looks right, without much re-analysis. But this calculation doesn't take much into account, it's simplistic.

For the last part, you get less delay the closer they are to the source. But if they're at the end of each boiler row (relative to direction of water flow), they don't get used in normal operation except under heavy load, which could be nice because they're less efficient. I'm not sure what you're asking though.

edit: 0.831 would be better than 0.866, according to https://wiki.factorio.com/index.php?tit ... throughput , so a better result is 2.2516

[Patric20878]

@Blake: Argh, I hope 0.13 fixes this. For now I'll just make the best possible design I can manage to and post pics. Maybe you can see if you can optimize logic or something without reducing power/area much once you see them. Too complicated to learn by explanation. Should be ready in a day or so.

@Aru: Know how there's a main belt, then they branch off into each row, and how I have the required burner inserters per 2 rows, placed on each branch? I'm saying that placing all burner inserters onto the closest row would be the best solution for reducing delay, since that would cut the time it takes for coal to reach further branches as well. But coal takes a long time to fill all boilers, so I'm questioning the relevance of your equation against the fact that even with no slowdown, the time it takes to fill boilers still slows it down. So the number of burner inserters should be calculated to the minimum that slowdown is lesser than boiler filling time.

[Aru]

Takes time to fill boilers? You're using the word slow to refer to different things, and I'm not sure what you're saying.

You or I could make a much more sophisticated calculation* that takes everything into account, but eeeh....

[Patric20878]

Eh, the slowdown you refer to with insufficient burner inserters, and the time it takes to fill all boilers with coal. Filling 90 on full consumption takes like half a game day, takes forever. What's the minimum number of burner inserters that has a faster recovery time than boiler filling time?

[Aru]

You have to be way more specific. What slow down? You mean, electrical devices operating more slowly because they're not getting enough power? You mean everything stopping because coal supply is cut off? "Filling 90 on full consumption", what is this? 90 boilers? And what is "full consumption"? Does that mean that power demand is matched to the maximum power output? And, I presume "boiler filling time" has something to do with the time it takes to come back to full power. So what does "faster recovery than boiler filling time" mean?

[Patric20878]

You have to be way more specific. What slow down? You mean, electrical devices operating more slowly because they're not getting enough power? You mean everything stopping because coal supply is cut off? "Filling 90 on full consumption", what is this? 90 boilers? And what is "full consumption"? Does that mean that power demand is matched to the maximum power output? And, I presume "boiler filling time" has something to do with the time it takes to come back to full power. So what does "faster recovery than boiler filling time" mean?

Slowdown refers to the slowdown your quadratic equation calculates burner inserters for. You should know what slowdown that refers to. 90 steam engines worth of boilers, which is 120 boilers in my setup. Full consumption = consumption is equal to or exceeds production. When all boilers are filled with coal, steam engines produce at full power.

Let's just say it takes half a game day to fill all boilers with coal. Your quadratic formula calculates minimum burner inserters needed to prevent slowdown on recovery from coal depletion. You made the formula, you know what slowdown this refers to. So instead of how many burner inserters gives 0% slowdown, I need the number that gives slowdown equal to the time it takes to fill all boilers with coal.

Because regardless of how less slowdown there is to recover to full power output from having enough burner inserters, the long time it takes to fill boilers with coal still causes major slowdown.

And hum. Think I decided between 20 and 21 boilers. It'll be 21, because 20 doesn't always start at 45.6 MW, and letting it heat up first won't get it to 45.9 either. Must be that funny mechanics you talked about, Blake. Testing this is exhausting, there are way too many different combinations and no way to shift builds by a tile or two in map editor. Rebuilding 90 steam engines over and over again is tedious as hell.

[Aru]

Here's what it calculates:

The number of burner inserters required to transfer enough fuel, to produce the ratio of power supplied to power demanded (demand is assumed to be at or less than the maximum power output) high enough, for the orange inserters to operate quickly enough that the engines can produce maximum power. Of course, it would only happen in two totally distinct steps like this if the belt went from no coal one moment, to completely filled with coal the next moment. So if you were to position this number of burner inserters around the coal input for example, there should be no major delay in automatically restarting from nothing. If you have 100 engines, and 14 burner inserters, then the biggest delay to reach full power will certainly be how long it takes for coal to reach all of the inserters.

If you had a lesser number of burner inserters, and the belts suddenly filled with coal, then the first wave of burner inserter transfers wouldn't be enough for the subsequent first wave of orange inserter transfers to bring the engines to full power. They would still come to full power, but it would be slower. With one burner inserter for say, 100 engines, it would be very, very slow.

I'm thinking of this more as the process of increasing power production from nothing to maximum, rather than as a slowing down.

So... the scenario you're imagining, is where all burners are reached by fresh coal before any of the other inserters, right? And you want to minimize the number of burner inserters, while still reaching near full power when coal reaches the very last inserter? Am I understanding it now? To do it that way, you would have less power during the restart process, but you'd reach full power at about the same time, right? The calculation was more about having enough burner inserters in a large steam array, that adding even more of them wouldn't make it start up much faster. So that belt travel delay is the major factor in restarting a large array, rather than a low burner inserter count.

To be honest, I think the easiest way to determine that, would be in-game trial and error. The inserter ratio would probably depend on the size of the array. To get a meaningful approximation, would need simulating the action of each inserter independently. But if I wanted to use burner inserters to maker the restart process smoother, I wouldn't skimp, I'd use enough that coal propagation through a large array is the only meaningful limiting factor at every moment.

And as for shifting buildings around, why not just use blueprints and construction bots in-game? It's funny that such a thing should be easier in-game than it is in an editor. All of my tests have been in-game.

I guess I should mention one more time, that my boilers and coal line being empty of coal at the same time, requiring this restart from nothing, has only ever happened to me once, and it was because I accidentally disconnected the mining drills on the coal field. That I knew to disconnect the engines from the rest of the base, and connect them to the drills, made the process far faster and less frustrating than it would have otherwise been. Having some burner inserters in the mix, probably wouldn't have helped unless I added burner drills on the coal, and gave them fuel. And then, it would have been slower that way.

[Patric20878]

There's too many unknown factors for me to do it from ingame. As blake said, even freaking placement order matters (ugh), and pipes don't even simulate actual water pressure, so I have no choice but to eliminate as many variables as possible, including testing from a pristine state everytime. Just a huge headache. I miss when things were more logical and compactness, power/area, and resource costs were the main things to worry about, instead of hoping the game makes water go in one direction instead of another, both equidistant.

And yeah, I'm mainly concerned with knowing if boiler fill time is ALWAYS the bottleneck, regardless of how many burner inserters there are (provided there's at least 1). The power jump thing is a secondary concern, because even if you power electric inserters as early as possible, much of the coal is "temporarily wasted" by going into boilers (at 5 max) instead of going to the next one. Having all boilers activate at once, even if just for 10 seconds, may well provide the same total amount of power that doing it gradually does.

[Aru]

With only one burner on a lot of engines, the coal will propogate a lot faster than the inserters can grab it, because the orange inserters are slow when given low power. There's not much that can be done to control the boiler input stack size without a more elaborate set up. And the set up would take more space, because there's so many boilers. If you could connect the boilers to logistics network like storage chests, and you used smart inserters instead of orange inserters, that would guarantee that the restart is super smooth, because you would be able to control the stack size and because fast/smart inserters are so much faster than orange (~2.7x as fast, at all power levels). You could control the input stack size with a constant combinator then... turn it down to 1, then back up to 5. But, like everything else that isn't a chest, you can't red/green wire boilers. If you could, I think the most important application is feeding smelters, especially ones making steel, because they keep going until the output slot is filled to a full stack. 100 steel in each furnace, is 500 iron plates of resources, it annoys me, especially for the furnaces in my science layout.

[Patric20878]

Maybe. But once coal fills up the first belt, it'll be at rest for the entire period between then and coal reaching the last boiler, which should be enough time I think.

[Shokubai]

I've been using a similar version of this for a while now. I do not weave belt inside though. I just have inserters take from one boiler to the next. I currently have one belt that feeds the first row of boilers and then 19 behind that which take from each other. For power I user a gap of 1 pipe at regular intervals to insert a pole of some nature.

2016-05-31 07_52_32-Factorio 0.12.33.png (302.84 KiB) Viewed 10761 times

It's more compact with standard poles and a 1 space pipe but I wanted to see what it was like this way.

[BlakeMW]

I figured out a 3/42/30 which is fully tileable, low tech, and uses strings of 21 boilers.


It works at 100% except it suffers the usual flaw that when started under full-load in certain orientations it fails to stabilize at 100%. It will stabilize once allowed to fill with water. My previous attempts at smallpump-less designs with 21 boilers didn't work, but this one does. Water should be delivered using at least 2 pipes, they can be pretty long.

Since i was examining it closely, I figured out more about the "cold start" problem. When the power plant is started under full load, either by introducing water into empty pipes, or by introducing coal to empty boilers it will fail to reach it's potential. If the water supply is poor (i.e. long pipes) it can be as low as 12.7MW (83%) and gets much closer to 100% with a better water supply (very short pipes or 3 pipes from the offshore pumps). If load is reduced to under 100% it will successfully stabilize and after that run properly as long as the pipes can deliver enough water.

The full load cold/empty start problem does not occur if the powerplant is generally orientated East to West or South to North (built with the offshore pumps on the south or east coast).

I'm not sure how seriously to take the "full load cold/empty start" problem, for a start it afflicts the 1/14/10 just as badly as any other design. In real games it's unlikely to happen with the exception of undersized backup steam, or perhaps after a death spiral in a purely steam powered factory. In the case of backup steam it can be avoided by orientating any steam setup East to West or South to North. It is much worse if the water supply is "poor" in terms of pipe capacity (i.e. the offshore pumps are trying to push water through a pipe system with barely enough capacity). My recommendation when testing power plant designs would be to allow them to stabilize under less than full load, before doing a full load test, as a cold/empty start under full load does not reveal flaws in a power plant design.

edit: fixed typo

[Qon]

I figured out a 3/42/20 which is fully tileable, low tech, and uses strings of 21 boilers.

3/42/30


That's a nice design. I'll use it when I build my powerplant. Thanks. It's also smaller than my design somehow.
I'll use blue for the vertical belt and medium poles instead though since I dislike small poles and I'm going to need several parallell belts for coal input anyways.

[Aru]

I figured out a 3/42/30 which is fully tileable, low tech, and uses strings of 21 boilers.


It works at 100% except it suffers the usual flaw that when started under full-load in certain orientations it fails to stabilize at 100%. It will stabilize once allowed to fill with water. My previous attempts at smallpump-less designs with 21 boilers didn't work, but this one does. Water should be delivered using at least 2 pipes, they can be pretty long.

Since i was examining it closely, I figured out more about the "cold start" problem. When the power plant is started under full load, either by introducing water into empty pipes, or by introducing coal to empty boilers it will fail to reach it's potential. If the water supply is poor (i.e. long pipes) it can be as low as 12.7MW (83%) and gets much closer to 100% with a better water supply (very short pipes or 3 pipes from the offshore pumps). If load is reduced to under 100% it will successfully stabilize and after that run properly as long as the pipes can deliver enough water.

The full load cold/empty start problem does not occur if the powerplant is generally orientated East to West or South to North (built with the offshore pumps on the south or east coast).

I'm not sure how seriously to take the "full load cold/empty start" problem, for a start it afflicts the 1/14/10 just as badly as any other design. In real games it's unlikely to happen with the exception of undersized backup steam, or perhaps after a death spiral in a purely steam powered factory. In the case of backup steam it can be avoided by orientating any steam setup East to West or South to North. It is much worse if the water supply is "poor" in terms of pipe capacity (i.e. the offshore pumps are trying to push water through a pipe system with barely enough capacity). My recommendation when testing power plant designs would be to allow them to stabilize under less than full load, before doing a full load test, as a cold/empty start under full load does not reveal flaws in a power plant design.

edit: fixed typo

That is very nice. And, I don't worry about the cold start. The only time it affected me, I was too new to the game to be interested in browsing the forums (and finding layouts like this) anyway. Most people I think, that want to experience the game some before exploring forums and finding these designs, will already be at the point that it won't happen to them, or they'll know how to deal with it better if it does (disconnect the base, manually feed fuel, connect to drills on coal patch, have a chest with red slots on each resource line for collecting and manual grabbing, including coal). I don't remember exactly the parameters of the calculation, so I'm not sure if it carries over, but the power ratio for operating orange inserters fast enough to fully fuel all engines was about 11%. (I think the relevant parameter is the 14:10 ratio.) So, my layout is still exactly the one I developed before being spoiled by any source of information except in-game testing, the same design that you see earlier in this thread, with all orange inserters. I said in this thread, and the TDT thread, that it was inspired by existing steam engine layouts, but I lied cause I was trying to be polite. I see now that it was totally unnecessary politeness. Patric saw my layout within a day of it being posted, and changed out his design in the first post within 5 days (Patric: "Dang, I see my thread made history to still have posts on it almost 15 months later lol. Haven't been here for over a year".... ). He deleted the original images (at least 3 of them, from the image host) (I wonder if he'll delete this one too, that someone put on the wiki), personally I would have just moved them to the bottom of the post and put them in a spoiler, at the most, since it inspired so much conversation, and the thread doesn't make much sense without seeing them. He had a few intermediate replacements, each one a little better. The most coincidental insight, was the offset of the pipe on every other engine row. Or rather, every third row here, every other row on mine, and I used an inserter on the opposite side instead of a red on the same side. I think this has at least been a productive and worthwhile thread, despite the distasteful contention. The back-and-forth, and analysis, is pretty similar to the most productive cooperative engineering groups I've ever been in. (edit: And like I suggested earlier in the thread, the designs gradually kind of converged, they look far more similar now.)

Patric's 'almost 15 month' old (I re-uploaded because it has since been deleted from wiki.factorio.com/images/Compact_power_plant_2.jpg):
Mine from page 2:

And, the death spiral, isn't really there... I had a misconception about the interpretation of that calculation from earlier, because I was focusing more on looking at it from the positive side. (Because, Patric mentioned using all burners instead of orange, so I calculated how many you really need to cold start the engines at full speed.) The further the power supply / demand ratio drops under 11%, the slower it is to come back to 11%. At 0 it doesn't happen, at 11% the rest of the climb depends only on fuel availability. The point of that silly quadratic equation earlier, is to calculate how many burner inserters you need to push the ratio to 11% (including the contribution of the slowed-down orange ones), assuming power demand is matched to (or less than) peak engine output.

"Cold start"... that's so much better of a phrase, I wish I'd been using that from the beginning. ...... On second read, I think you might be talking about a totally different cold start issue, oops. About fluid mechanics, not inserter speed under low power.

Anyway, I think using these blueprints, along with blueprint string mod, is starting to spoil the game for me. While I love making the designs, once I have them, the game is less fun. I am wired like a programmer I guess... I want to make it as good as I possibly can, one time, and then re use it over and over. Or, maybe I'm just playing it too much. In part due to the use of these made-in-advance highly optimized designs, I have solar panels coming in before I need more than the 50 engines pictured above. But steam engines, are just so much cheaper per power than panels are. But as the game progresses, the attention required for the logistics of the engines' need for fuel and water, makes them take more time, attention, and management than just laying down panels and accumulators. I think, maybe the research for solar panels should be harder to get.

[Patric20878]

@Aru: Funny thing is, the only reason why I still update this thread is because I somewhat see it as a responsibility. I stopped playing Factorio over a year ago and have no plans to do so again. But the whole reason I ever made this thread was to contribute to making the best possible steam engine design, so here I still am doing so instead of playing Touhou or Tekkit Classic instead, heh. If you ever loved some software or game or whatever but found the dev abandoned it long ago, that's the kind of thing I want to avoid doing. I don't think I can entrust anyone else to be dedicated enough to continue improving on a design even long after they quit the game. Also, I didn't like the discussion we had either, but unfortunately like you've found as well, being nice and polite typically doesn't agree with being productive, and well, priorities man. Nothing personal, all professional, mkay?

And eh, I never really saw a need to keep the images once I make better versions of my designs. There's a reason I keep the version number accurate and mention what was in the previous designs. Don't like old versions cluttering my imgur account, but if anyone wanted to see old versions of my setup, they could always ask. I keep archives of all my previous versions of everything all backed up neatly on 3 different sources. No one's asked yet though. And no, I won't delete it from the wiki lol, it's hosted on the wiki, and I only delete the images from my imgur when I think it's obsolete. It's to be organized.

Meaning, if I can get the update I'm working on now to not have any major disadvantage at all compared to current design (v1.2), the current design pics will be replaced too. But it prob ain't happening because fluid mechanics are weird as heck. Until it is (if it is), the new version will be posted on the bottom of the original post. Think of it like stable and experimental release, lol.

[Aru]

That's a nice design. I'll use it when I build my powerplant. Thanks. It's also smaller than my design somehow.
I'll use blue for the vertical belt and medium poles instead though since I dislike small poles and I'm going to need several parallell belts for coal input anyways.

Thought I'd mention, a yellow belt can support 105 steam engines at full speed. I even tested it after the calculation, there's a picture somewhere with 10 rows of 10 at full load. Yellow belt items per second is 13.392, I have it memorized now, I've used it so much. That, and boilers are 50% efficiency, engines are 510 kW, and coal is 8 MJ each. And boilers produce 390 kW of heat. So, for 30 engines, you only need 510*30/390= ~39.231, 40 boilers, not 42.
Belt transport system
Transport belt experiments

[BlakeMW]

The back-and-forth, and analysis, is pretty similar to the most productive cooperative engineering groups I've ever been in.

And, the death spiral, isn't really there...

This seems to be true. I usually go all-electric inserters to save fuel and use fasts when pulling from belts (yellows for direct feed between boilers), often if you have any power at all the system can bootstrap itself because all the inserters move in tandem - sure it may be snail pace, but they all pick up a coal and slowly make their turn. Then suddenly a whole lot of boilers light up simultaneously, bringing the system to a healthy power level. I've tested this with designs with like 160 steam engines, connecting them up to a full load, then placing a single solar panel to bootstrap the system, which is like 0.1% power.

0.13 is going to change things, I imagine the rapid inserter will be extremely power-friendly and able to operate at a very low power level (that is I expect the belt will do most the work of loading a bunch of coal into the hand), in general designs with a lot of direct feed between boilers will probably go with fast and rapid inserters and burners will fall out of favor since the lack of stack bonus will cause them to burn even more fuel.

"Cold start"... that's so much better of a phrase, I wish I'd been using that from the beginning. ...... On second read, I think you might be talking about a totally different cold start issue, oops. About fluid mechanics, not inserter speed under low power.

I like "black start" for starting with no electricity as it's the correct term IRL, and I'm tending towards "cold start" for starting the system full of cold water (or no water).

Anyway, I think using these blueprints, along with blueprint string mod, is starting to spoil the game for me. While I love making the designs, once I have them, the game is less fun.

I do not use other people's blueprints for that reason, though I draw inspiration from them. I do use my own blueprint strings for refining setups (because they require so much design) and large steam power plants. Most blueprints I make on the fly.

I'm very fond of solar/steam as a compromise between steam and solar/accu, because you cut the non-blueprintable parts of steam, that is water and fuel supply, by 70%. You need no accus which is nice for saving oil and require 30% fewer solar panels because you don't need to charge accus. Death spiral and black start problems totally go away because your factory is fully powered during the day. You have 30% of your power generation as baseline (always available) meaning you only get brownouts if you over-extend, which is great for lasers.

I also don't have a ploppable solar blueprint w/integrated roboport. Nope. I just have like 10 roboports in the power armor and lay down the solar fields personally. I've used solar/accu blueprints before, but they're really really boring.

[Aru]

I also only use my own blueprints. For now, that's... oil processing, research, solar/accumulator array, and smelter lines. (Oh, and a turret blueprint, I think it's 89 items, for 9 personal roboports. It's for combat use, not defense.) Next up, is probably some kind of inclusive blueprint for production of intermediate items for personal use. The oil processing has cracking stuff, but the input pipes aren't connected, so I set up the storage and control scheme without a blueprint. I have a constant combinator where I set the number of tanks for each fluid, and the lower and upper storage thresholds as percentages for cracking. So, if I set it to 20 and 99, it'll crack A into B if: A is 99% or higher, in order to keep the refineries going, or B is under 20%.

Solar
Research

(I added a note about boilers in the post above)

[Qon]

Anyway, I think using these blueprints, along with blueprint string mod, is starting to spoil the game for me. While I love making the designs, once I have them, the game is less fun. I am wired like a programmer I guess... I want to make it as good as I possibly can, one time, and then re use it over and over.

I think it's nice to solve one part. There's more things to solve anyways so I don't have to solve the same thing over and over again. I think that would be boring.

Also there's challenges and challenge mods if you want to make all your designs and patterns useless. Right now one of my playthroughs is a burner only factory, no electric grid. It's not actually as tedious as that mihgt sound, it's still automatable. I have burner leech mod which means burner inserters can pull coal from burner units and from belts if necessary, and burner assembly machines. For the things that there's no burner equivalent machine to I pipe hot water instead of wire electricity and place steam engines next to the electric machine connected with poles without wires. I gets a bit tricky sometimes because you almost always need half the belt you pull from to have coal on (so you can't have multiple input ingredients) it and you don't have any fancy long inserters to reach several belts.

Thought I'd mention, a yellow belt can support 105 steam engines at full speed. I even tested it after the calculation, there's a picture somewhere with 10 rows of 10 at full load. Yellow belt items per second is 13.392, I have it memorized now, I've used it so much. That, and boilers are 50% efficiency, engines are 510 kW, and coal is 8 MJ each. And boilers produce 390 kW of heat. So, for 30 engines, you only need 510*30/390= ~39.231, 40 boilers, not 42.

Then a blue belt would support ~300 steam engines at 150MW. Sounds like I need like 20 blue belts of coal then. And that is before I start to my explosive expansion that will occur as soon as my 1 minute rocket factory is blueprintable for rocket spam on a massive scale. But thanks for the boiler number.

I do not use other people's blueprints for that reason, though I draw inspiration from them. I do use my own blueprint strings for refining setups (because they require so much design) and large steam power plants. Most blueprints I make on the fly.

I like doing my own, but for things like your steam setup it's so simple that it would be silly to not do it when I know it without looking at the picture. And I took part in it's design. I'm not going to avoid building a perfect design because someone else has done it also. Of course I will allow myself to make the best factory I can make. If I aquire knowledge from others then I should use it and maybe build on top of it. And It will have to be modified slighty if I want lots of GW out of it so it won't be enough to make and exact copy and plop and forget anyway.