Jump to content

Waterfall storage for LOX/LH2


Recommended Posts

Amongst the cell to cell heat transfer rules, we have clamping that limits the per-tick transfer to (T1-T2)/4, and a lower limit of at least 0.1 DTU.

This is easy to take advantage of and keep gases in gaseous state. For example, if 1mg H2 at 70 degrees meets a cell that is -260C, the temperature delta is 330C and the DTU delta is 0.792; (T1-T2)/4 forces the DTU delta to be 0.198 and the hydrogen should liquify. However, 500mcg of H2 will have a per-tick DTU delta of 0.099 (less than 0.1) and is therefore safe from temperature changes.

This means that using a small-enough quantity of gases in a waterfall liquid storage allows us to completely ignore their temperature.

When building this in survival, you have to keep in mind that the above rules only provide safety from cell-to-cell transfers. You cannot have buildings in the gas cells; even insulated ceramic pipes will cause (slow) heat transfer. No drywalls or wires in the cell, and absolutely no tempshifts nearby.

Edit: if you're doing this in space and you must have drywalls, make them out of insulation for those two tiles. 

image.thumb.png.0fff3232d5ffae0f4f40196e5d2d1bb1.png

The construction process is a bit tricky but doable:

Pump very small amounts of mixed gas into a pipe section. The safe quantity depends on the gas, but since H2 is safe below 500mcg, pretty much anything else is OK if less than 1mg.

image.thumb.png.3ff22ba45b6184d2e554155fdfcf6d9d.png

Save two packets of the mixed gas for later use by deconstructing segments around it:

image.png.ad85742ef8b64a41aa70b98fa455cb96.png

Prepare the waterfall chamber for receiving these gas packets. I built two dead-end pipe sections and a dead-end bridge from insulation. Insulated ceramics *should* be fine. You will still get some heat transfer and will have to work quickly with ceramics, but if you have access to insulation, just use it. You will get the materials back.

image.png.1d69e62b31b6bb8b4888e78ff4a856c8.png

Start up your LOX/LH2 machine and liquify some gas. You don't need a lot, just enough to ensure these liquid seals:

image.png.a5f94f8b5651973e815dc11583170603.png

When this is done stop the process, and ensure there's no gas left anywhere. Send in the small packets of gas you saved earlier, deconstruct some tiles to allow dupe access and have them deconstruct the dead-end pipes:

image.png.ee7772e8ea331ad87b36a4cd72f07e64.png

Do the top one first, then the lower one and the bridge. Sweep the materials out. If it's ceramics, you may want to hurry.

Yay! Clean up what you have to, and seal the chambers back up:

image.png.c6b6935a47cf22df5b6006fbf5c7c9b5.png

It works:

image.thumb.png.3d0f31665ca367311648318dadc8d6f2.png

Yes, you could use door compressors for liquid storage. This is smaller, and much harder to set up, therefore also much cooler.

 

Link to comment
Share on other sites

Does any excess gas (eg CO2) escape into the condensing chamber along with the waterfall compressor chamber  when you break open the pipes?  I'm assuming the tiny drop on the step prevents it from escaping.

Are you certain that drywalls can't be used, this would exclude this from being made in space.

Incidentally, you can improve your condensing chamber design.  Have 2 tiles space beteen the top row of tempshifts and lower tempshifts.  This creates an insulating vacuum between the hot gasses at the top and the cold liquid at the bottom,

Like this.

Untitled.png.61af17bf1257ac26638cdb800d5c2eac.png

Link to comment
Share on other sites

Not if you pre-seal the two tiles that are going to be gas. I probably wasn't clear but you want to have droplets of liquid h2/o2 here:

image.png.a5f94f8b5651973e815dc11583170603.png

The droplets form a liquid lock which keeps the gas in and dupes are free to come and go. 

You want vacuum in the top gas tile, and you'll have some liquid in the lower one. The liquid will be displaced to the left when you deconstruct the lower pipe section and the gas pops out. This is why the order of deconstruction is important.

You don't have to do anything special to make this happen, just liquify some H2. If you build the waterfall storage similar to mine, the seal will be there automatically.

Also, the location of the LOX hydro sensor is important. The liquid H2 pump can't touch the lower seal tile directly so it's safe, but for LOX my liquid pump's suction cross covers that tile. The sensor prevents it from creating vacuum there and ruining the setup.

19 minutes ago, Craigjw said:

Are you certain that drywalls can't be used, this would exclude this from being made in space.

Sadly yes, unless made from insulation. The thing is, while you're liquifying your initial batch of H2/O2 to create the seals, the drywall will also take on that super low temperature, and will insta-cool the barrier gas when you release it from the pipe. If you build the drywall from insulation it should be fine though.

22 minutes ago, Craigjw said:

condensing chamber design

I suppose I missed out on that, but I had to fit in piping to allow 3 ATs to cool the room somewhat evenly, and it's a fairly small space.

Link to comment
Share on other sites

LOL, three AQ's for just Hydrogen is too many.

You are probably spending a large amount of energy on phase changing already liquefied Hydrogen, don't pre-cool your Hydrogen with your liquid Hydrogen, it causes the liquid Hydrogen to boil off and phase change back into gas, then you have to expend more energy to re-liquify it.

You should pre-cool your gases.  The O2 coolant is useful for this, as even with full 1kg/s of O2 liquification, your AQ is still being underused most of the time.  The closer you can get the temperature of the gas entering the condensation chamber, the better.

Keep your gas pipes outside of the condensation chamber.

 

Link to comment
Share on other sites

Nothing is boiling back up :) If any did, the waterfall storage wouldn't last long. I mean I am kind of a noob compared to some people here, but not that much of a noob. The gas pipe layout was considered, and I decided to run them like that on purpose. It's ceramic piping, the incoming hydrogen cools down a whopping 1 degree in it, any warming it does to the liquid is easily countered with 1 radiant pipe segment with SC in it.

I am also doing 3000g/sec H2 and 3000g/sec O2, both start at 99C due to the 99C water that goes to the electrolyzers, so two SC ATs wouldn't be enough for hydrogen. The AT handling O2 is about 75% utilized, not much left for pre-cooling there. 

And before you ask, this is sustainable for a while. 20kg/sec water is being electrolyzed, and I have about a hundred tons of hydrogen saved up from earlier.

Link to comment
Share on other sites

What I did was to over cool the Super coolant to below hydrogen freezing temps, then had  2 loops in each of the O2 & Hydrogen condensers.  I have a temperature sensor on the coolant pipe as it exits the system and if it's too cold, I use a flow valve at say 3kg/s to introduce new coolant to the loop.  The upper cooling loop is set to near freezing temperature, while the lower cooling loop is set to keep the liquid at a liquid temperature.

Even with Ceramics, running the gas pipes through it is not ideal.  If for any reason it stalls for a long period of time, the contents will freeze.  Just be aware of that, I have fallen into this pitfall previously.

What I'm saying is, that by not having the vacuum buffer, you are getting temperature creep from the top layer to the bottom, requiring more effort to cool the system.

Link to comment
Share on other sites

The easiest thing to do for hydrogen is to have one temp sensor in the liquid, and aim it at whatever you want your final temperature to be. Have that control the AT(s), and ignore coolant temp completely. Supercoolant won't freeze in an AT no matter what, so you're safe. This runs the risk of hydrogen solidifying, but it shouldn't happen if your input is at a constant rate.

33 minutes ago, Craigjw said:

If for any reason it stalls for a long period of time, the contents will freeze. 

Well my only other alternative would have been to run the gas pipes through steam, which would have worked against what I am trying to do. So a small risk of a broken pipe vs. actually putting more load on the system, it was an easy decision. Yes, a different layout, etc, etc, but this is pretty compact this way, I wasn't looking to add dead vacuum just to route pipes. :)

36 minutes ago, Craigjw said:

by not having the vacuum buffer, you are getting temperature creep from the top layer to the bottom, requiring more effort to cool

It can't be more effort though, can it? The only external temperature inputs are the SC and the H2, and if the liquid H2 interacts with the gas, it's not energy that's being added to the system. The gas cools, the liquid warms a bit; the former is desired, the latter can be handled with no extra effort.

Link to comment
Share on other sites

I'll forgive you regarding the gas pipes :D

It's like the counter flow petrol boiler, you don't want the temperature creep between the two temperature zones.

Initially, I was inputting 60c H/O into my tanks and had the condensation chambers full of diamond tempshift and the result was terrible, it could never reach full capacity.  The problem was, that the hot gases were causing the liquid to evaporate off, so my conversion would be sporadic. I increased the conversion rate by making use of a vacuum zone, however it still wasn't 100%.

I used diamond tempshifts at the top and ceramic drywall below it.   This allow the top to be sub-freezing temperatures without freezing the pool below it.  Yes I know I am missing an insulated tile on the top right.

x.png.c1988ced6bad67df516e837da7b69ac8.png

 

My next approach was to pre-cool the gases.

This pre-cools my Hydrogen to -246 & Oxygen to -120, running on a single AT. I'm injecting 4000g/s SC into the Hydrogen cooling loop heat exchange and 350g/s into the O2 cooling loop heat exchange, leaving 5650g/s of SC not used.  The temperature on the O2 fluctuates a fair amount, but that's only because I'm injecting super coolant into it at -262c.

HO2_precool.thumb.png.c26a5ce5f0c804b15fc79bfcae8b3b36.png

So, knowing that this single AT pre-cools both gases down to -245 & -120 respectively, do you still think that your condensing pool is efficient?

This is the beast in it's entirety. 1 AT for H, 1 AT for O2 & 1 for cooling the turbines.  All AT's are hardly ever running.

HO2_1.png

Note that this is an old game, from a time when Wheezeworts were good.

Link to comment
Share on other sites

28 minutes ago, Craigjw said:

So, knowing that this single AT pre-cools both gases down to -245 & -120 respectively, do you still think that your condensing pool is efficient?

It is :) There's really not much to it, you need to extract so many DTUs from so much gas, and as long as you do it in a way that the input doesn't stall, and your output temp is what you want it to be, there's nothing to do about efficiency.

Your single AT doing the pre-cooling, it's pretty much full tilt, with 1000g/sec, right? From there it takes very little for both gases to liquify.

The only issue with injecting hot gases, like you were doing with 60C, is that it takes time for them to cool down. I don't know the exact building-to-tile mechanics (especially the weird multipliers that they use) but there's only so much cooling that, say, a -260 degree alu pipe is allowed to do on a tile of 60C hydrogen per tick. So unless you provide enough room for the gas to spread out you're going to get throttled at the vents. You circumvented this by pre-cooling the gas so it can liquify in one tick. I made the room just big enough so that the gas cools without throttling as it descends in the chamber. I would have made the room smaller if I could have, but this is what it takes based on a quick sandbox experiment. I think I used less room to do the thing in the end. :)

 

Link to comment
Share on other sites

I can't remember how active the AT on the pre-cooler was, it wasn't 100% of the time though.  The system used a loop of super coolant containing full 10kg packets which only loop around the heat exchange block, when I inject cold coolant into the loop, it displaces some from the loop.  The coolant displaced is sent for re-cooling.  I only ever cool 10kg packets, so if I inject a total of 4350g/s of into the loop, then 4350g/s of warm coolant leaves the loop and sent to the AT loop.  The AT loop continually cycles liquid in a tank, cooling as necessary.

From what I remember, the pre-cooler AT was under capacity.  This build only used Gold & Wolframite, as Aluminium wasn't available.

The vacuum doesn't just stop heat creep, it's effectively a big pump, removing all that hot gas and turning into liquid, allowing the vent to run without blocking up, in combination with pre-cooling the gases, means that gas almost immediately condenses and it can run at full 1000g/s input.

This system never backed up and two hydrogen rockets couldn't use the hydrogen fast enough.

Link to comment
Share on other sites

36 minutes ago, Lifegrow said:

Wealth of information here btw

That is a great thread, that's essentially where I got the idea for the shape of the storage. If it looks sort of like @mathmanican's geyser compressor, it's because.. it is. The hydro sensors acting as safety are from @Saturnus in that thread.

Temperature-proof small gas packets have also been around for a while, even though I don't recall seeing a setup that relied solely on cell-to-cell immunity. 

Link to comment
Share on other sites

6 minutes ago, biopon said:

That is a great thread, that's essentially where I got the idea for the shape of the storage. If it looks sort of like @mathmanican's geyser compressor, it's because.. it is. The hydro sensors acting as safety are from @Saturnus in that thread.

It's why I mentioned it - always nice to credit the giant shoulders you're stood on :p 

Link to comment
Share on other sites

On 1.11.2019 at 7:01 PM, biopon said:
On 1.11.2019 at 6:47 PM, Craigjw said:

Are you certain that drywalls can't be used, this would exclude this from being made in space.

Sadly yes, unless made from insulation. The thing is, while you're liquifying your initial batch of H2/O2 to create the seals, the drywall will also take on that super low temperature, and will insta-cool the barrier gas when you release it from the pipe. If you build the drywall from insulation it should be fine though.

I fail to see how you can't just rebuild the drywall before releasing the gases. Sure, some LH2 will get sucked into the void...
Just be careful with "hot" building mats. It's probably best to trap a single dupe to avoid confusion and reuse the same material again! Probably also watch where the dupe stands (he is a source of heat)
 

For setting the gases up, you could also 'empty canisters'. Seems like similar rules apply for low masses and canisters as they do for gases regarding (the lack of) heat transfer.

Edit: clarification

Link to comment
Share on other sites

1 minute ago, Yalp said:

I fail to see how you can't just rebuild the drywall before releasing the gases. Sure, some LH2 will get sucked into the void...
Just be careful with "hot" building mats. It's probably best to trap a single dupe to avoid confusion and reuse the same material again! Probably also watch where the dupe stands (he is a source of heat)
 

For setting the gases up, you could also 'empty canisters'. Seems like similar rules apply for low masses and (lack) heat transfer.

You could rebuild a regular drywall, that's a clever option. Empty canister though... it's tricky to get them to the right place. You'd pretty much have to have a canister emptier one tile over, and watch the dupes like a hawk, so you can cancel in the last second.

I haven't found dupe heat transfer to be problematic, but then again, I have a supercoolant pipe over the top bead partly because I was expecting trouble. :) I think warm sweater gives them a lot of extra insulation, it might make sense to have them wear one if you think they're going to be in places that are not covered by pipes.

Link to comment
Share on other sites

On 1.11.2019 at 6:04 PM, biopon said:

Prepare the waterfall chamber for receiving these gas packets. I built two dead-end pipe sections and a dead-end bridge from insulation. Insulated ceramics *should* be fine. You will still get some heat transfer and will have to work quickly with ceramics

A reason for using canisters.

26 minutes ago, biopon said:

Empty canister though... it's tricky to get them to the right place

Sorry, I was unclear, don't use the canister emptier, use the plumbing skill on a segment of pipe above the future gas lock to get the right amount. Once everything else is in place, empty the bottles. 

Spoiler

image.png.3aa94e70a0a57ad1727e357beb21a18f.png

image.png.a134f800adaf1c0b489b3bbb94785577.png

 

 

26 minutes ago, biopon said:

I haven't found dupe heat transfer to be problematic,

Just wanted to mention it, since they do transfer heat and it might cause problems depending on the setup :).

Spoiler

5dbfec2e71133_Dupeheattransfer.gif.f911ab6833845cdb79f8c5a57bc8f503.gif

 

Link to comment
Share on other sites

3 minutes ago, biopon said:

It's a neat idea but are gas canisters exempt from heat exchange?

 

47 minutes ago, Yalp said:

For setting the gases up, you could also 'empty canisters'. Seems like similar rules apply for low masses and canisters as they do for gases regarding (the lack of) heat transfer.

For low masses,  yes :) (I edited my original comment, hope it avoids confusion/makes things more clear...)

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.

Please be aware that the content of this thread may be outdated and no longer applicable.

×
×
  • Create New...