Optimized Steam Engine Setup

[Patric20878]

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

On the low-tech version though, I can't really see the why one would want to focus on compactness with such starter tech. Early game isn't a time when you have the luxury to be min-maxing a temporary compact build, only to replace it shortly after, but eh. This design practically completely reverses the compactness back to that original steam power setup, as well as the stackability from being able to use fast inserters, and the only thing I see much different at all from the original is that it's rotated 90 degrees relative to the belt compared to the original.

If anything, since it is oriented for low tech bases, I feel it should just replace the original design on your tips thread if it's superior, rather than add it like it's a variant. It's an optimized version of the original with the same exact compactness but optimized for low-tech, so the original is by now totally made obsolete by this and my design.

Also, I believe the "Item Sorter Through Splitters" thing under Aru's pics in your thread is out of place, seeing my low power alert section is right under it.

[Aru]

One thing I can see about that layout is the furnaces at the end will be a lower priority when it comes to coal. The first line will get half while the third one down will get like 1/8th. Other than that it's a good layout, I'll be sure to add it to the TDT

That should be okay. By my calculation, if boilers need 780 kW of fuel, produce 390 kW in water heat, each row is 10 engines, engines use (and produce) 510 kW, coal is 8000 kJ each, and yellow belts are ~13.392 items/second, that should be 13.392/510/10*390/780*8000 ~= 10.504 rows of engines from the main yellow belt. The side belts would each only be using 1/10.504 of their maximum throughput. Splitters only split half to each belt when both belts are open. If one is blocked or absent, everything goes to the other belt, and yellow splitters don't impede the maximum throughput of a single yellow belt. Or even two yellow belts, really. You are right that the last ones will be lower priority, but it shouldn't be possible for the first rows to deprive other rows of any coal. Unless it's just not getting enough coal to begin with... but then, it doesn't matter which boilers get that coal, they all make the same energy from it.

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

On the low-tech version though, I can't really see the why one would want to focus on compactness with such starter tech. Early game isn't a time when you have the luxury to be min-maxing a temporary compact build, only to replace it shortly after, but eh. This design practically completely reverses the compactness back to that original steam power setup, as well as the stackability from being able to use fast inserters, and the only thing I see much different at all from the original is that it's rotated 90 degrees relative to the belt compared to the original.

If anything, since it is oriented for low tech bases, I feel it should just replace the original design on your tips thread if it's superior, rather than add it like it's a variant. It's an optimized version of the original with the same exact compactness but optimized for low-tech, so the original is by now totally made obsolete by this and my design.

Also, I believe the "Item Sorter Through Splitters" thing under Aru's pics in your thread is out of place, seeing my low power alert section is right under it.

You have inspired many! It is just like that first one yes, but it's one tile shorter, and uses fewer and cheaper poles (2 small per row instead of 3 medium), and only orange inserters, and it's designed to be tiled with no gaps between rows. As compact as possible. Well, sort of (one tile shorter)... if you count the poles at the end, it's not really any shorter. But poles don't block movement, and they can be moved farther out and still reach the engines, you could put buildings right on the end instead. So maybe 0.75 tiles shorter, hah. And I did theory for just how many rows you can tile, all stacked together tight like that. I really like how you offset one boiler at the end of the rows like that, with the pipe and single boiler-to-boiler inserter, I don't think I would have solved that puzzle on my own.

You don't want to focus on designing as-compact-as-possible things in the early game, so that's the advantage of designing them in advance like we are, we do the min-max ahead of time so it can easily be placed and tiled later. And it is fully stackable without fast inserters, up to 100 engines from just a single (fully compressed) yellow belt coming in from one side.

I took ideas from the first one and made a slightly more optimized version (by tech and resource) for personal use (the best I could make it), and I thought I'd show it. I didn't know if it would make it to the first post. They can say, 'made by patric and slightly improved by Aru' if they want, or just not use my name at all, that's fine, that's why I put it here instead of directly in the TDT thread, because it's derivative. The hardest part for the post was stitching the image together. Making the layout was for me.

[Patric20878]

Actually, your setup uses 5 small poles every 2 rows. Mine uses 5 medium poles every 3 rows. Once you stack my design away from the coal belt, you'll find you only need 5 poles every 3 rows, not the 10 it seems to show, so actually, you use cheaper poles, but not fewer. And yeah, poles don't block movement, but it still is the same width as the original, 53 tiles, whether it does or not. It's the same as redstone dust in Minecraft redstone circuitry, which occupies a block but has no collision model, but is always counted as a full block, measurement wise. Just like those, poles still count the full tile in width, even if not the full tile in mass (aka occupied volume), and mass vs volume is an important distinction to make. And yeah, been a while so I don't even remember if I did the theory on how many you could stack like that, but if you change the row of normal inserters to fast inserters, the chain would only be limited by the max rate of the long inserters in between. And you are correct in your reply to Skellitor that the priority and 1/2, 1/4, 1/8 thing wouldn't matter. Those belts act as buffers; once they fill up, coal proceeds through the main belt as usual.

Also, I think I may have forgotten to mention something deeper about my design my original post, on its compactness. While in my picture I show the full 30 steam engine setup with either 2 water tanks or more, in actual practice, I'd actually only set them up in sets of 29, replacing the middle last steam engine with the last power pole. This is because compactness and efficiency is only really only meaningful if you express it in terms of production over volume, otherwise a super compact design that produces almost nothing is just useless. So if disregarding the fancy water boilers and whatnot and going straight for compactness, just the rectangular area containing the 29 steam engines would be 50 tiles wide, 9 tiles long, which would be ~15.52 tiles per steam engine. Yours and the original would be 30 steam engines at 53 tiles wide, 9 tiles long, which would be 15.9 tiles per steam engine, where a steam engine is 15 tiles. Unfortunately, that translates to using 9*3=27 tiles instead of 15 just to fit the 30th engine, and not in a compact cuboid way either. The above is what I really consider to be my design's main compacting advantage, being literally 3 blocks narrower and opening up that much more space when not using the water tank feature.

And actually, looking closer at the way your steam engines are stacked again...yours actually seem like a better way to do it, along a coal belt instead of away, for when you need to stack en masse. The boiler to boiler thing is indeed a cool trick, but I think your way has it beat. Steam engine rows can't just be stacked in 2 dimensions since it must draw water from a lake, so my original idea with it being able to stack away from the coal belt is actually rather pointless now that I look at it. Because stacking engines on your design doesn't increase load on previous inserters, it is only limited by coal belt throughput, not inserter speed, so funnily enough, from this new perspective, your design's advantage isn't just low-tech, it's actually stackability! And that is the real advantage of your design. In fact, if I upgrade my design to incorporate your belt/inserter setup (with major credit to you of course), your design's low-tech would be almost no practical advantage at all, for the reasons below:

The only difference in tech between yours and mine would be the medium electric pole. My design wouldn't even have any use for fast or long inserters anymore. The question is, in what stage of the game would you EVER require massive amounts of steam engines, but still NOT have such basic tech like medium poles? This is almost like someone building a blue science factory before a green science one, it's simply the wrong logical order, even if you can. Thus, from a practical standpoint, your design would offer no tech/resource advantage, but instead, offer increased stackability. In this way, building this new upgraded design would not only be just as low tech and easy to build, it'd also be as compact as mine, saving the player from needing to tear down the setup and rearrange it again, a beautiful solution. Originally, I considered your setup to be optimized for resource/tech but not compactness, so thought it'd be a better idea to present our designs as separate optimized designs with different focus, but now I see that our designs can be combined to be better than both of our originals, so good job on your design for having a such an important feature I overlooked. I'm not sure if your backup power thing is the same as mine and CBA to look, so won't be including that into my upgrade.

[Aru]

I guess, I like small poles because they're so cost effective, and cost effective is important since steam is the first source of electricity you get. I use small poles almost exclusively in all of my premade designs, because they're so cheap, and they craft so quickly, and these designs are mostly meant for the main base where materials are within reach. (Likewise orange inserters are a lot cheaper than blue/red inserters.) The main advantages of steam, are cost, space and tech efficiency, so those are exactly the advantages I wanted to press, to make steam engines as worthwhile as possible. A lot of the advantage of solar is it's simplicity for not requiring routing of water and coal, so having simple designs in advance help to narrow that difference and make steam easier / more viable. I don't see how it's like making blue packs before red, since medium poles require research, while steam power and small poles do not (and incidentally research actually requires steam power).

And my backup power thing as it involves the steam layout, is only the pumps, the other stuff is separate. The purpose is different but the pumps are the same, because yours alerts of low coal and shuts off water supply, while mine turns *on* water supply when solar / accumulators are insufficient. (To prioritize using accumulators over steam to save coal and pollution, and in event of large laser defense, because steam is such high power production per area and per cost.) That is also not original, but the pumps are used in most such systems (both backups and others too, like your early warning) so I included them. The backup works without combinators but they help keep it from turning back off prematurely. I could argue that a system that shuts engines off when coal is low (that uses radar and fluid handling), is not needed once you reach that stage of progress. It's a stronger argument than saying small poles in a steam power design is silly, but not one I would make. There is nothing one can design that is more worthy of small poles, because there is nothing else that you do earlier in the game that involves electricity.

And yes, the primary upgrade from yours, was showing that you only need cheaper no-research items for full steam effectiveness and high degree of compactness (no: blue or red inserters, red belts, medium poles, underground belts, gap between rows). And that you can power a lot of engines (100) with one yellow belt, and you can snake that yellow belt to every row of engines without any gaps between rows for more belts, and therefore don't need higher inserters to keep up with coal transfer. The existing designs already had compactness. It seemed odd, that such a popular post should rely on researched tech for a 'default' kind of steam power design. I just made mine as um, simple and elegantly generic as I could, and also cheap. This seems like one of the most likely threads for new players to find. Steam power is such a standard thing that neither one of us probably deserves credit. A hundred people have probably arrived independently at different aspects of these designs, such as pump:engine ratio, boiler:engine ratio, how to make compact, how to use cheap low tech. But this reference is worthy of optimizing because it's such a common (necessary, in fact) aspect of the game, especially for new players. We do "have the luxury to be min-maxing", so I am. For me that's most of the joy of the game anyway.

[Patric20878]

The main advantage of steam power is that it's the only source of power besides solar power. It extends well into mid-game, up until you mass produce solar panels to the point that they actually beat out the steam engines. So there is plenty of time to upgrade. What I said is having a massive amount of steam engines is never something you need in early game, only in mid game, and in midgame, you should've had ample steel production set up long ago, along with red/green science. So the analogy here is red science is to maybe a row or little amount of steam engines, green science is to medium electric poles, and a massive amount of steam engines is to blue science. If you have a massive number of steam engines before you can even use them anytime soon, priorities are obviously wrong, and if you can use them, your base should've been advanced enough to have medium poles researched long ago. And as for setting up my design early game, there's no reason I can't just leave the 2nd row of my 3-row set blank, using small poles to connect to them, until I have medium poles to finalize the compactness given by my 3-row set design. Hence, there's never a time when I feel the need to set it up suboptimally, then tear it down and re-set it up optimally, when I could just partially set it up optimally, then complete the empty rows also optimally later, saving the time of tearing down all the suboptimal bits. And eh, sounds like your backup system is allowing steam power production to backup solar power production, where my backup system is about allowing water tank storage to backup steam power production. Then k, as that isn't directly related to steam power specifically, I'll leave that to you, outside my setup.

As for not needing researched items to make a steam engine setup, I think anyone knows that. But when you don't use researched items, you lose in compactness. It's no more odd that my setup uses medium poles to increase compactness than the fact that yours doesn't use burner inserters to pull coal, something which I'm leaning on doing again. Like said, there's little reason to build in a suboptimal layout just because tech isn't researched yet. Simply building the steam engine rows that don't require medium poles is simpler than building all 3 rows suboptimally, then taking extra time to replace them later. As for new players, either I'll add in this part about completing 2 of every 3 rows until it's researched, or it should be obvious that optimizing anything in every engineering game ever made was never a topic newbies should be focusing on. Just getting it to work is what the wiki is for

I'll be designing the upgrade to my design and see what I end up with. Been a year and it's scary how many things I don't remember, heh. But not requiring power poles on the inserters at all should make for a great improvement. Would you clarify also, does 100 engines per yellow belt mean 10 engine rows? If it is, that's pretty decent, so I'll just put yellow belts instead of red ones by default. Also, I had no idea this thread was popular until now, wow 15k views

(If only the TC factory design community was this active...)

[Aru]

Would you clarify also, does 100 engines per yellow belt mean 10 engine rows? If it is, that's pretty decent, so I'll just put yellow belts instead of red ones by default.

Yep. I did a test, just to make sure the calculation was correct, screenshots in the TDT thread (100 engines at full throttle is a fun sight), and it's verified. If we keep this up our preferences might end up identical. You can use burner inserters, in fact I should reconsider the same. Maybe I will finally have a use for them.

[Patric20878]

K, I'll be redesigning the inserter/belt part of my design then. You make a good point with new players potentially not realizing they don't have all-or-nothing copy my design, so I'll be sure to take a pic of what I said about just building 2 of every 3 rows until medium poles are researched. And yeah, I do believe once I incorporate your design's strengths, our designs should converge into one, making for a pretty awesome upgrade to both of ours.

And heh, I see you share my tendency to make a post, then make 200 more edits on it thanks to proofreading and forgetting to say stuff

[Aru]

And heh, I see you share my tendency to make a post, then make 200 more edits on it thanks to proofreading and forgetting to say stuff

I do. I just try to make them fast enough that it doesn't reveal all my edits at the bottom, hah.

[Patric20878]

Heh, same.

From a glance, looks like I'll be able to compact the tube from your design easily enough using underground belts. Inserters will be changed to burner inserters. Hopefully I finish this soon though, because this wasn't what I had in mind for a break to take a rest from TC factory engineering

And darn, they removed the ability for inserters to pickup items from a belt diagonal from it since last time I played. No wonder you had that belt on the end. This will probably force a fugly u-turn on the splitter to be able to compact it then. How much does that drop throughput compared to just a turn again?

[Aru]

And darn, they removed the ability for inserters to pickup items from a belt diagonal from it since last time I played. No wonder you had that belt on the end. This will probably force a fugly u-turn on the splitter to be able to compact it then. How much does that drop throughput compared to just a turn again?

It doesn't. Ending belts one early was ugly, so they fixed it, and turns don't reduce throughput at all any more. Side loads on red belts still do.

[Patric20878]

and turns don't reduce throughput at all any more

Holy shit, really? No need to belt logic and splitter merge anymore? Awesome!

[Aru]

and turns don't reduce throughput at all any more

Holy shit, really? No need to belt logic and splitter merge anymore? Awesome!

Well there's still complicated belt balancers (like these; but those are old too), and splitter merging, but it's more direct without the turn limitation. You still need either splitter merge, or a drop from a faster belt, to fully compress.

[Patric20878]

Well yeah, on stuff other than turns. But turns no longer having reduced throughput is just wonderful. Oho

Welp, attempted optimization on your design that puts boiler rows in pairs. Result wasn't pretty, and made it impossible to compact the rows to fit within the steam engine row's total height + 1 (the row for poles). So, going to do a hybrid setup, combining the boiler trick with your branching design. Easily saves space and looks very promising!

[Aru]

You should still be able to do 100 engines on yellow belt even with burner inserters.
Update: These numbers are wrong, because inserters use much less than the listed power during continuous operation.

Orange:
uses 0.49% of power produced, 24.93 kW per 10 engines
24.932kW/(510kW*10)~=0.00489 / power produced; (5100kW/8000kJ*2)/(.831/s) * (13kW - 0.4kW) + 14*0.4kW ~= 24.932kW

Burners:
uses 3.89% of the coal consumed by the engines (electrical production is unaffected), 396.373 kW per 10 engines
1/(.579/s)*180kW/8000kJ ~= 0.0389 / coal consumed; .0389*2*510kW*10 ~= 396.373 kW

Items per second for 100 engines with all burner inserters, (1/.579*180kW / 8000 kJ + 1) * 12.75/s = 13.245 < 13.392

Edited to include the 50% efficiency of boilers (I forgot because boiler consumption isn't directly used).

[Patric20878]

Eh, no need to use math to prove this one. As long as the inserter pulls faster than boilers burn them, it's fast enough And if you chain boilers together, the first inserter's required rate is just coal usage rate * # of boilers in the chain.

And already got a beautifully compact design. This time I've decided to just go ahead and use 10-9-10 for the 3 rows. So many advantages over 10-10-10, such as having higher efficiency and only needing 4 medium poles instead of 5, heh.

[Aru]

Eh, no need to use math to prove this one. As long as the inserter pulls faster than boilers burn them, it's fast enough

It's cause the burner inserters use so much energy, it was more to see how many engines you can put on a yellow belt with them, I was just making sure yellow belt throughput isn't a factor for the previous 100 engine assertion, and it's not.

[Patric20878]

Actually in this setup, it doesn't use the 188 KW power worth of coal it states. You have to remember that it's pulling at much lower rates than its max. You should be able to find how much it's actually using by the expression, (burner inserter max rate) / (boiler coal consumption rate) * 188 KW. Which means it really doesn't use all that much energy here.

And seeing how it isn't actually anything to do with electricity, the only question is if coal flow to a branch exceeds the total consumption rates of all the boilers in that branch, and total coal flow to the system exceeds the total consumption of all boilers in the system. Or were you calculating that indirectly by representing coal input/consumption in KW? Was expecting item rates instead of energy consumption, if so. The unlabelled calculation you posted is rather confusing to read and I would like to know what each number means there and how you got them.

[Aru]

I edited units into the formulas, so they make sense. Also I forgot to include the half efficiency of boilers in both calculations (because boiler power isn't used) but it's there now.

It just occurred to me, that a useful calculation can be made about the downward spiral with orange inserters fueling boilers which sometimes sneaks up on people, and how many burner inserters you need to prevent it.

[Patric20878]

Oh, k.

Well seeing that I'm not using a single orange inserter anywhere in this new design, inserter speed will be reliant on coal flow instead of coal flow -> electricity production. Still been too long for me to understand where you're getting some those numbers from, but like said, as long as coal in exceeds coal out on both main and branch belt levels, splitters will automatically ensure the all boilers get powered eventually, and that's the only calculation I'd need to know. And man am I loving the burner inserters, even if you cut off coal supply an hour, it will still automatically start back up once you connect it back on, without any need to provide an additional boost of power to get the inserters started back up. Much better than electric inserter's circular dependency to power itself.

[Aru]

As for that downward spiral I mentioned with (14) electric inserters into (10) boilers, the threshold is ~10.96% (510*10/8000*2/.831/14). That is, if power supply drops under 10.96% of power demand, it can drop the rest of the way to 0 further because the inserters don't get enough power.

With one burner inserter and 13 electric, it's 6.44%. With two burners, it's 1.17%. Three burners, it's impossible.
You only need 2 to make this impossible.
(Technically any number will make this impossible such as 1 in 100, but you need 2 out of 14, or more precisely 1.36 out of 14, to avoid any delay in recovering from 0 power and any accelerated power drop.)

So really, the problem is not the decrease from this threshold to 0%, because it's a very small drop (11%), not far from 0% to begin with. The only real problem, then, is getting a huge engine array running again after it has run out of fuel, which might happen if your coal field unexpectedly empties. The solution in this situation, would be to first sever the engines from the rest of the grid. You can wire them to the mining drills on a new coal patch, and manually add some fuel, and re-attach the engines when the coal arrives. (This has only happened to me once, and it was early enough in the game that I didn't have to sever to restart the engines. I accidentally removed a pole that powered the coal miners, then later the engines ran out.)

So, while that 10.96% power threshold is already pretty low, it can be reduced to 0% by swapping 3 2 orange inserters for burner inserters on each row of engines. Because burner inserters use so much energy, make them the last three two so they are only used if needed. And because 11% is already low, there's not much point doing this with just one or two burners, it should be three two. But even if the inserters are adequate to fully fuel the engines, they need adequate fuel, and burner inserters won't help keep the drills mining the coal powered. They could hurt, by taking a cut of the coal supply (3.7% with all burners).