Jump to content

Testing Gas Shutoff. Doesn't seem to work when throughput is limited.


Recommended Posts

I've noticed that the gas shutoff sometimes mixes gasses when the throughput is limited so I made this little test to demonstrate it. Any reason why? Do I have to use gas filters to remedy this specific situation?

Throughput can be limited when oxygen pressure is reached in one's base for example. That is simulated here by limiting the throughput of the oxygen to the tank.

Link to comment
Share on other sites

Yes

remember there is a packet of oxygen in front of the hydrogen at the shutoff

normally the packet will continue on past the shutoff when it turns on

however if the pipe is blocked it will go though the shutoff

this is why most players don't make shutoff filters if there is any risk of pipes backing up

Link to comment
Share on other sites

And that mechanic right there made me want to bang my head on my desk a few times when I was designing the Spaghetti Monster.

Took a total of 7 shutoffs and a couple clever gas loops to deal with that problem when sorting 2 gases, and being able to deal with any other gas that might come into the system.

Any time a single throughput shut-off valve gets blocked and then re-opens, it's inevitably gonna let other stuff through. Highly obnoxious.

Link to comment
Share on other sites

automatic with shutoffs is fragile, it depends on both side pipes won`t be blocked, if you want to set up for energy savings, you should add some buff container or other logic or pipe blocked handle, actually, with automation, you could make electrolyte self-sufficient

Link to comment
Share on other sites

I ended up using a valve for each gas and a feedback loop to handle the problem.

Spoiler

image.png.ef3c73c7d387a72c575b3eb9105a7164.png

Top left senses oxygen and sends it right.  If it isn't oxygen, or the pipe is full, then it goes through the lower right element sensor which is set to polluted oxygen.  If its PO2, then it goes to the right below the O2 pipe.  If it is Oxygen, or the pipe is full, then it goes through a feedback loop.  

In this set-up, if either the oxygen or the PO2 backs up too much, the whole system will shut down until the pipes on the other sides of the valves opens up again.  

Link to comment
Share on other sites

4 minutes ago, KittenIsAGeek said:

I ended up using a valve for each gas and a feedback loop to handle the problem.

  Hide contents

image.png.ef3c73c7d387a72c575b3eb9105a7164.png

Top left senses oxygen and sends it right.  If it isn't oxygen, or the pipe is full, then it goes through the lower right element sensor which is set to polluted oxygen.  If its PO2, then it goes to the right below the O2 pipe.  If it is Oxygen, or the pipe is full, then it goes through a feedback loop.  

In this set-up, if either the oxygen or the PO2 backs up too much, the whole system will shut down until the pipes on the other sides of the valves opens up again.  

Would you mine giving a gas piping layout view?

 

Link to comment
Share on other sites

30 minutes ago, ruhrohraggy said:

Would you mine giving a gas piping layout view?

 

Certainly.   In the first shot, everything's working as it should.  In the second, I removed a bridge that was sharing the PO2 vent and I broke the O2 pipe so that the valve would back up.

Spoiler

image.png.ef70c1d2fae54368f83cafa3dd4878e1.png

image.png.24c6dc4fe108756309a14d30336a8d2b.png

Edit: Strangely it didn't actually post... Its posted now, though!

Link to comment
Share on other sites

Ah, ok.

Don't think the bridge really affects the system at all anyway as long as the vent isn't blocked...

Though, (and I built this system and tested it) if the p.oxygen vent / line happens to back up, you get some bleed-through in the other direction, which, based on what it looks like you're doing with it, is probably no big deal. Just a more energy efficient gas sorter, with no need for absolute 0 tolerance for gases going where they shouldn't. (And you probably couldn't back up a high pressure vent leading to a puft ranch with p.o2 anyway...but I guess you could try :P...)

It's definitely a nice little compact feedback design though, I like it.

Also, is the area with the left gas pump in it going to be sealed forever? where it's only p.o2 and o2?

I ask this because this thing won't do well with a 3rd gas entering it's system at all, and it won't take much of a 3rd gas to bork this sorter due to the short loop. Any gas that isn't p.o2 and o2 will get cycled into the loop with nowhere to go and it will jam up and require Duplicant attention.

If that's a non-issue, or you're solving this problem by room control or other mechanical means, then no biggie, just something to consider.

Link to comment
Share on other sites

1 hour ago, KittenIsAGeek said:

Certainly.   In the first shot, everything's working as it should.  In the second, I removed a bridge that was sharing the PO2 vent and I broke the O2 pipe so that the valve would back up.

  Hide contents

image.png.ef70c1d2fae54368f83cafa3dd4878e1.png

image.png.24c6dc4fe108756309a14d30336a8d2b.png

Edit: Strangely it didn't actually post... Its posted now, though!

Why not make the right shutoff one tile higher... 

 

It looks terrible 

 

and more important, the bridge should from the pump to the shutoff system, this will make the system loop forever even if all output is blocked. 

Link to comment
Share on other sites

26 minutes ago, dallion said:

Why not make the right shutoff one tile higher... 

 

It looks terrible 

 

and more important, the bridge should from the pump to the shutoff system, this will make the system loop forever even if all output is blocked. 

  1. It was made a piece at a time as I realized I needed to change it.
  2. It's functional, in the space provided, but no, it isn't pretty.
  3. It works just fine, but you're right -- I could re-organize the bridge priority just a bit.  However, since O2 is the most-likely to back up, and the fan only introduces 500g/s into the pipes, there aren't any problems.  If the O2 backs up for long enough, everything shuts down automatically and I haven't had any packets go where they aren't supposed to.  Charting the flow on pen and paper I can see now that there's a very slight chance that a badly-placed PO2 packet can lock the system up completely if it arrives on the O2 sensor just as the pipes stop flowing.  The fan hasn't picked up a single PO2 packet since the gas pressures stabilized in the room, so I'm not going to worry about it.  The only mode of failure where it could be a problem is if I happen to run out of sand or regolith for several cycles.  I could ALSO solve the problem by placing an element sensor on the feedback loop that turns the fan off if anything is flowing through it.  Then gasses would never stop moving, completely eliminating the potential problem.
45 minutes ago, ruhrohraggy said:

Ah, ok.

Don't think the bridge really affects the system at all anyway as long as the vent isn't blocked...

Though, (and I built this system and tested it) if the p.oxygen vent / line happens to back up, you get some bleed-through in the other direction, which, based on what it looks like you're doing with it, is probably no big deal. Just a more energy efficient gas sorter, with no need for absolute 0 tolerance for gases going where they shouldn't. (And you probably couldn't back up a high pressure vent leading to a puft ranch with p.o2 anyway...but I guess you could try :P...)

It's definitely a nice little compact feedback design though, I like it.

Also, is the area with the left gas pump in it going to be sealed forever? where it's only p.o2 and o2?

I ask this because this thing won't do well with a 3rd gas entering it's system at all, and it won't take much of a 3rd gas to bork this sorter due to the short loop. Any gas that isn't p.o2 and o2 will get cycled into the loop with nowhere to go and it will jam up and require Duplicant attention.

If that's a non-issue, or you're solving this problem by room control or other mechanical means, then no biggie, just something to consider.

The room is completely sealed.  PO2 (pre-filtered) comes in at the bottom and deoderizers turn it into O2.  The O2 goes out into the system through a priority bridge, so as long as my dupes are breathing at least 500g/s, it should never back up.  Any excess PO2 that isn't used by the system goes to the vent into the puft room.  Also, any PO2 packets that get picked up by the fan go out (with priority) that same vent.  I haven't had any packets go the wrong way since I put the feedback loop on.  I tested what happens when the system backs up, but it hasn't actually happened in the game.

Link to comment
Share on other sites

11 hours ago, ruhrohraggy said:

Don't think the bridge really affects the system at all anyway as long as the vent isn't blocked...

actually a bridge is important for preventing back flow

I would put it next to the shutoff so unwanted packets don't have room to back flow into the shut off

6.thumb.png.54e1555e179d3e2089dbd867e1f35aa2.png

Link to comment
Share on other sites

That's often what I do, though in the case above its after a second valve.  Anyway, just an update on my build.. for the past 10 cycles 100% of my base's oxygen comes from PO2 bottles.  I shut down my algae farm once pressure in the room exceeded 10kg.  Eventually the bottles will sublimate enough that it'll start pulling O2 from the algae room again and the atmo-sensor will unlock the room... but until then:

Spoiler

image.thumb.png.2e171d573181e00b6db93c4eee0a321a.png

So far, not a single PO2 packet has gone through the O2 pipe.  So my bypass and feedback are working correctly.

Link to comment
Share on other sites

On 6.12.2018 at 2:00 AM, Express said:

I've noticed that the gas shutoff sometimes mixes gasses when the throughput is limited so I made this little test to demonstrate it. Any reason why? Do I have to use gas filters to remedy this specific situation?

Throughput can be limited when oxygen pressure is reached in one's base for example. That is simulated here by limiting the throughput of the oxygen to the tank.

you can't see on layout, but on input of gas shut-off are sill some oxygen when senor find some hydrogen,

here you have compact design of filter room

remember, pump for oxygen must pump below 888g/s

so you can add two pumps, connect with pipes and add valve set on 800g/s

Link to comment
Share on other sites

All that's needed to make these type of sensor filters work correctly is proper bridging priority setup.  Try this setup, based on the OPs layout, and you'll see that it no longer allows oxygen to get past the valve because it has an overflow path past the valve that effectively blocks new gas from entering the valve in the event of oxygen backup. 

20181207105221_1.thumb.jpg.be03de7475da822c63a4007cb8e6b506.jpg

In order to handle the possibility of hydrogen backing up, additional bridges can be added to create the same type of loop on that side of the valve, as seen here:

20181207105210_1.thumb.jpg.7deae2fdef1d97bc78c4dff572b64cf6.jpg

Link to comment
Share on other sites

Here's my pipe blockage proof solution. It can naturally be minimized a lot but the key feature is that the two pipe block detectors controls the input gas shut off (and the pump if needed). The small "ears" are a buffer so that packets between the input shut off and the filter shut off can still be processed without causing problems.
image.thumb.png.6274bc2a17ce14beda085b4e04dbf08b.png


image.thumb.png.3d4a095ab780b85eb7e893b3a14bf325.png

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...