95%+ per cycle throughput electrolyzer with only 2 pumps, no doors

[Saturnus]

4 minutes ago, xenoborg said:

For the love of god man it produces a net gain in power through a full hydrogen generator being fed from the thing non stop. so stop with this wasted power when it makes more power than it consumes.

Look, I'm not trying to bash you. All I'm saying is that you are wasting power, not a lot but some. As far as I know pretty much all electrolyzer set ups are power positive so what matters (to me at least) is, how much is the positive part?

Btw, do check out how I used a bridge for better merging on the oxygen line above.

[FIXBUGFIXBUGFIX]

There is no merge in this setup. And each oxygen package outputted by up pump is bigger than 500g

[xenoborg]

Yes all electrolyzers are power positive in that they run themselves as and when needed with a trickle of power per day going somewhere else, but how much power are you getting from it per cycle?

if you need to be running 2 electrolyzers and three pumps to produce enough hydrogen to supply a hydrogen generator with 60kg a cycle, if it even produces that much due to the fact its being overpressured as theres a layer of hydrogen at the top of the electrolyzer, causing the thing to not be producing 888g/s 112g/s and instead 566g/s 83g/s 333g/s 60g/s or whatever the numbers were, then you are producing less power per cycle for your colony than you otherwise could using an electrolyzer.

At that point it doesn't matter if you're wasting a bit here or a bit there, if at the end of a cycle you're producing a net gain of 114KJ from 480KJ costing 366KJ if its more power than you would gain from another "net gain" electrolyzer setup that can only give you <114KJ due to not being on as long / producing less.

[Saturnus]

15 minutes ago, R9MX4 said:

There is no merge in this setup. And each oxygen package outputted by up pump is bigger than 500g

Yes. I did that. But they don't add up to 500g/s total, only 476g/s (rounded up).

[Ainsley4ever]

Your debate really made me wonder about which setup was the most efficient between Xenoborg's and the "conventional" setup (I'm not saying that Xeno's is bad, I just don't know how to call it: p) shown by Saturnus. To a point that I spent the last three hours (really) doing calculations. And I want to show you my results.

So we have two setups to separate: Xenoborg's electrolyzer and Saturnus' electrolyzer (for this one I used a setup a little different from his but with the same efficiency).

Spoiler

Xenoborg's electrolyser:

(Don't mind the filters, they were just here for the test and weren't included in the results)

 

"Conventional" electrolyser:

We are trying to find out which of the two is the most efficient, i.e. which produces the most excess energy per 1kg of oxygen produced (excess energy=total energy from the output hydrogen - energy used to power the setup).

So I performed experiments to obtain data on energy consumption per cycle, oxygen production per cycle, and total oxygen and hydrogen production for about 5 tonnes of water (these last two numbers are not exact, but it is not important because it's the ratios that interest us). I then inserted the data into a spreadsheet and calculated the rest to get what we were looking for.

Here are the results:

Spoiler

It's important to note that Xenoborg's electrolyser seems to produce less than the other one because it only outputs 527.5/600=0.879 kg of oxygen per second, versus 1kg of oxygen per second for the other one. But again, we aim for efficiency, so the ratios interest us and not the max output.

So as you can see Xenoborg's setup produces 0.299 kJ (or 299 W) of excess energy per kg of oxygen produced, while Saturnus' produces 0.314 kJ (or 314 W) of excess energy per kg of oxygen created, which means that Xenoborg's setup is a tiny bit less efficient that the other one.

BUT let's look at the ratios. The theoretical perfect ratio hydrogen/oxygen is 1/8, or 0.125. The "conventional" build easily achieves that ratio, but not Xenoborg's, which means it suffers from hydrogen deletion. This is what makes it less efficient that Saturnus', which is actually quite unfair because if we look at the power usage per 1kg of oxygen created, we can see that the "conventional" setup actually uses more energy than Xeno's.

So in conclusion, Saturnus' electrolyser is currently more efficient than Xenoborg's, just because of the unfair hydrogen deletion bug. So if it gets patched in the future, you can expect Xeno's setup to be the most efficient one.

But all of this is based on my math, which isn't the best in the Multiverse. So if I made a mistake somewhere (which is quite possible), feel free to correct me please... I will include the spreadsheet so you can check it out and correct me if needed : The Spreadsheet of Truth.xlsx

 

And Xenomorb, if you want to make your electrolyser run 100% of the time, make the 2x4 room where the electrolyser is located wider (4x4). This way, it will have more space to breathe and will never overpressurize

 

 

EDIT : I made a big mistake in my thinking. Actually if the hydrogen deletion bug was deleted, the two setups would have the  same efficiency in terms of excess energy per kg of oxygen generated... Thanks @Saturnus for correcting me.

[avc15]

I know it's been round and round. I'll weigh in that most people optimize for mass efficiency (as close to 0% deletion as possible) and power efficiency (biggest component being max mass per packet pumped out)

If one particular electrolyzer doesn't get full utilization, that's a minor concern compared to wasting power and/or water. In fact, it's impossible to sustain 100% electrolyzer output because you can only pump out however much your dupes will breathe, in the long run.

But yeah, "how do I get max output from my electrolyzers" is a natural place a lot of people go first. Just keep it up, you might decide that power and fuel waste matters more to you sometime down the road.

Also the different types of equipment aren't matched to one another. You have to choose whether you're going to build for throughput (requires extra under-utilized pump systems and hvac loops) or power efficiency (requires that you accept less than perfect throughput from each electrolysis module)

[Angpaur]

More than 2 months ago I posted on this forum same electrolyzer design (link), because I was quite excited about it as it was compact and had good performance in oxygen generation. Later I noticed that 2 pump electrolyzer setups are deleting hydrogen, so I stopped using it.

However there is a one, unique electrolyzer build with just 2 pumps, which is not deleting any hydrogen. It is even more interesting as it is changing ratio of generated hydrogen and oxygen and it creates more hydrogen at a cost of oxygen. You can find it here

[Saturnus]

23 minutes ago, Ainsley4ever said:

So in conclusion, Saturnus' electrolyser is currently more efficient than Xenoborg's, just because of the unfair hydrogen deletion bug. So if it gets patched in the future, you can expect Xeno's setup to be the most efficient one.

I think your conclusion is inaccurate. If there was no mass deletion both set ups would be equal. The whole idea behind the conventional set up with it's larger area and initial building material use is to avoid mass deletion.

If there was no mass deletion, I and everyone else would be using the compact version but it's a known fact that has been document several times on the forum already that the stacked design deletes roughly 5% hydrogen.

[avc15]

Still another option is the basic one-pump open-air module. Open air electrolysis gives you the option of using an aquatuner (or even wheezes) to cool your oxygen instead of hvacs, for a massive power savings. Honestly this could be the 2nd most efficient exploit-free way (open air with wheezeworts being the first?)

First, you build a fair number of these around your base. They are small, but you'll only get partial output from each one because, natural circulation of air. Only the H2 gets pumped, the O2 just falls out across the pressure gradient on its own. Set the pump control as low as you can while still getting full packets (actually, using a latch might even provide more efficient results)

The oxygen gets cooled by an *aquatuner* instead of hvacs. Here's how; I'm going to start at the coolant and work my way back to the aquatuner. Just use flow valves and shutoff valves on a temp control and send coolant in through radiant pipes. Place a couple temp shift plates to draw heat out of the coolant:

So, whenever the air temp goes above 23C we circulate some -40C petroleum around. There are two tiles of radiant pipe, the rest is insulated. Tempshift plates even out the heat so we don't just get one cold spot. Set flow control to 1000g/s because that's about right to pull around half the delta T out of my coolant.

Now return the coolant to a cold tank which regulates to -40C through an aquatuner directly above:

The hot tank is full of Pwater, and it regulates its temp by pumping its contents out through a sieve, replenishing with pwater from the washroom. (I don't like this, I think it's a bit munchkin but it's definitely NOT an exploit). The hot tank also has a 2nd aquatuner for my cold water supply to the base.

oh and yeah, that geyser arrangement is natural, I didn't mod it. seed 

Certainly you can see some inefficiencies especially with my liquid loops but the point is this, hydrogen from these electrolysis modules provides nearly all the power needs for my base because no hvacs - only aquatuners, so we save about 95% of the power we'd have spent pumping oxygen, and about (edit after analysis) 70% of the power we'd have spent on cooling.

My coal gens only turn on when there's a spike in demand. For instance, when both aquatuners AND the sieve are all running at once, or my dupes are crushing rock to replenish sand (haven't surfaced yet), or I'm refining. I generally have extra hydrogen in storage available on demand. In fact it's a problem that these tanks keep filling up. I will have to keep them from topping off, or build more tanks.

 

 

[Ainsley4ever]

26 minutes ago, Saturnus said:

I think your conclusion is inaccurate. If there was no mass deletion both set ups would be equal.

You're right! How silly of me... I corrected the mistake, thanks for pointing it out.

[xenoborg]

Thanks for the math @Ainsley4ever

Had to go back and double check my math, but yeah max hydrogen per cycle was my goal, And while the hydrogen bug exists Saturnus does indeed produce slightly cheaper 1kg hydrogen.

At 100% hydrogen efficiency though, 75.03749999733578kg of hydrogen would be Saturnus max output So it would cost him 5.479526903409611KJ per 1kg, If mine were at 100%, so 65.93750000590784kg It would cost me 5.353099525586726KJ per 1kg If I've done that right?

So yeah, slightly worse performance for now for less space I suppose.

 

[avc15]

oh, your goal is maximum hydrogen power, and to hell with the extra oxygen. That didn't really come across in your original post, it just looked like you were trying to get maximum utilization.

Which is usually unimportant, in most applications. But in this one, might change things.

I don't have any comment on the relative usefulness of any approach for this purpose. Just validating this one point.

[Mr.Trueba]

What an unnecessary fight for the "efficiency". 
Guys, you need make some friends and better enjoy the game. 

[avc15]

different people have different goals. Max efficiency is a thing that will keep you interested well past your first few play-throughs for casual fun.

[mathmanican]

For me this thread has been very useful.

1. I learned that a gas pump can work with 2 different gasses and get higher than 500g in one packet (corrected a misconception - thanks @R9MX4).
2. I also learned that two gas pumps can't empty a room fast enough to achieve a full 1000g/s average (thanks @Saturnus), despite the above find. Hence no electrolyzer setup that uses only two pumps will ever fully be able to empty the 1000g/s created by an electrolyzer without some power loss (sure - it can be power positive, but could be improved). 
3. I built another setup, adding a tile of empty space on both sides of the compact design by @xenoborg. It actually made the mass loss worse. The bottom pump also started getting packets of hydrogen occasionally, so even more power loss.  This probably has implications on open builds (probably already documented if I search enough in the forums). 
4. I tested a submerged build (liquid covering the electrolyzer). It converted the liquid stuff perfectly to hydrogen/oxygen, without any mass loss (didn't have debug installed, so I might have missed the -888/112 total that @Saturnus noticed). The averaged calculuation actually turned 1000kg water to 1000.1 kg of gas (guessing round off error). 

[avc15]

35 minutes ago, mathmanican said:

1. I built another setup, adding a tile of empty space on both sides of the compact design by @xenoborg. It actually made the mass loss worse. The bottom pump also started getting packets of hydrogen occasionally, so even more power loss.  This probably has implications on open builds (probably already documented if I search enough in the forums). 

The very compact one-pump open-air setup I posted above never loses its hydrogen-oxygen boundary after it's primed. It does delete about 5% of the hydrogen it produces but it's very stable.

The lesser tradeoff is in power consumption vs hydrogen production. And that you can't really get the electrolyzer to go above around 80% of its rated production. Neither of those are really big things, but this one is: You can't properly cool an open air system without quite a lot of abyssalite, you need to have some advanced tech unlocked, and you need at least a moderate amount of refined metal available. You could do it earlier if you'd dug up a bunch (~~10) wheezeworts but you might be pushing a deadline as you use up your last bits of algae.

So, producing 5% less hydrogen isn't a big deal since I'm consuming about half as much total power. But, you can't really build that system easily and cool it too unless you delay switching to electrolyzers for a really long time. Which is a different challenge.

[Mariilyn]

Interesting. I've been using the 3 x 10 layout of this kind of SPOM for an eternity (I like to call them MSPOM - short for Mechanical SPOM,  because of the filter usage) but never bothered looking for the hydrogen loss. I verified in my colony and the number for quick one cycle of testing corresponded to what was said (62,8 kg hydrogen + 531,9 kg oxygen produced for one cycle). Not really an issue though, and the difference in efficiency between both SPOMs is trivial. 

5 hours ago, avc15 said:

Open air electrolysis gives you the option of using an aquatuner (or even wheezes) to cool your oxygen instead of hvacs, for a massive power savings. Honestly this could be the 2nd most efficient exploit-free way (open air with wheezeworts being the first?)

What exactly is preventing me from cooling my base with an aquatuner? Your demonstration was for a cooling method, not an oxygen production module. I usually build one early-game because it's quicker to make, but I think the other more organised SPOM variants are more adequate for the late game. Even in an open-air setup, most of the oxygen needs to be pumped regardless for the exosuits or for surplus storage.

[avc15]

1 hour ago, Mariilyn said:

What exactly is preventing me from cooling my base with an aquatuner? Your demonstration was for a cooling method, not an oxygen production module. I usually build one early-game because it's quicker to make, but I think the other more organised SPOM variants are more adequate for the late game. Even in an open-air setup, most of the oxygen needs to be pumped regardless for the exosuits or for surplus storage.

Nothing. It's a demonstration that SPOM is inefficient because it uses HVAC. Most of my bases can keep suits charged with 1 pump, anywhere from 6-20 dupes. If that's not you, don't use an open air setup. But  the cooling method goes with it - you have to cool somehow. You can put wheezeworts on the ground, done. You can use an aquatuner. If you were going to use hvac, you'd build a spom. (I don't think I'm being disparaging to the original author of spom here, but I'll listen to him - spom was an exercise, an answer to a question - 'how power-positive can I make a self-contained electrolysis system?')

Yes, you have to pump oxygen to exosuits. How much of your oxygen is consumed inside an exosuit vs in base?

I don't understand storing surplus oxygen. You get the same result just letting your electrolyzers sit idle instead. Build a few extra. They'll ramp up their output when needed.

Anyway, you don't have to use any build that doesn't meet your objectives. But, if you want an extremely power efficient base, and your dupes spend a lot of time in base, open air saves a tremendous amount of power because you can choose more efficient cooling methods.

Oh, and I take my time getting to high tech. 20 athletics makes a huge difference. Our objectives make a large impact on our priorities.

Sorry for the many edits. I had a few drinks tonight. I also think electrolyzers or the power content of hydrogen is over tuned.

[Djoums]

4 hours ago, avc15 said:

I don't understand storing surplus oxygen. You get the same result just letting your electrolyzers sit idle instead. Build a few extra. They'll ramp up their output when needed.

I can't speak for Mariilyn but in my case, storing O2 allows to have constant H2 production from my electrolyzers, no pipes backing up, no overpressure. You also don't need a "smart" cooling solution for O2 because its flow is constant, but making it smart isn't hard anyway.

In a way it's similar to venting it if it doesn't get any further use, maybe oxylite will change that if it becomes useful.

[itsYiyas]

23 hours ago, Ainsley4ever said:

 

  Reveal hidden contents

 

I'm a bit confused here. You tested Xeno's using 2 gas filters but he uses gas shutoffs with gas sensors? This is a colossal 240w deficit towards Xeno

[Saturnus]

2 minutes ago, itsYiyas said:

I'm a bit confused here. You tested Xeno's using 2 gas filters but he uses gas shutoffs with gas sensors? This is a colossal 240w deficit towards Xeno

Read the text. He clearly says they are not included in the calculation.

[itsYiyas]

Just now, Saturnus said:

Read the text. He clearly says they are not included in the calculation.

Oh right, since they are in the picture I assumed he was using them. Could have just used a valve loop instead.

[Ainsley4ever]

51 minutes ago, itsYiyas said:

Could have just used a valve loop instead.

You're right, and in an actual setup I would have done that. But this was just for testing and I was too lazy (and nothing, NOTHING beats my laziness) to build mechanical filters.

[itsYiyas]

2 hours ago, Ainsley4ever said:

You're right, and in an actual setup I would have done that. But this was just for testing and I was too lazy (and nothing, NOTHING beats my laziness) to build mechanical filters.

I'll be honest, my obsession with them is unhealthy.

By chance did you see this recent thread on reddit.

Even with a big pump it's looking like a 55-65% excess KJ per oxy KG on your table...

I use automation stop the entire operation when a gas flow is occurring for 5s+. 

Although the automation isn't required, if the hydrogen backs up then part of the system stalls and needs to be rebuilt. Alternatively, just consume all 170g/s of the hydrogen and it doesn't even matter.

[Saturnus]

7 minutes ago, itsYiyas said:

By chance did you see this recent thread on reddit.

I already commented in another thread how I think the two design he shows are very poor electrolyser designs. It's not that they don't work, they probably do but the materials needed to make it is ludicrous. In any normal survival mode game you need to have your oxygen supply sorted out fairly quickly. You are on a timer to get it set up before easy to access algae runs out. Once you have to search and dig for small algae patches to keep the oxygen flowing you're wasting far too much time.

So what you need is a design that requires very few exotic materials, very little refined metal, and definitely no plastic or diamond. I'd also be wary of being dependant on designs that includes cooling, especially wheezewort cooling, as that's a limiting factor as well. Not every map will have easy access to 4-6-8 wheezeworts straight away, and you might have other plans with them. In that respect the stacked design and the gas separating designs are the same. They can both be done with simple materials and a tiny amount of refined metal.