Jump to content

Is there a simple automation mechanism to fill a gas/liquid tank?


Recommended Posts

I've just been pinching off the output with a valve, and that works, but the tanks lack a much needed fill sensor. We've got one on the fridge ....just saying.

Anyway. I'm trying to automate the process of opening the outlet once the tank is full...or at least cause flow to bypass the tank once it has been filled this would be a great start.

My applications thus far have used pipe element sensors and so on, what I'm really looking for is a genuine pipe clog detector ( I can wish for as much ) so I can let things pass out of the tank or around a full tank and on to other systems...once the tank is full. 

I already have several ideas and one partial solution - but I'm hoping this has already been done in some clever fashion I haven't come up with or thought of. 

So far I've used an SR+filter ( set longer than a buffer/clock timer ). And this works, but only for my time limited and intermittently running carbon skimmer. This got scrapped for something more trivial considering the circumstances.

-  I have a Carbon Skimmer(CS) set to run on a clock timer at night ( 60 seconds ). On the outlet of the CS is a pipe ( connected to a tank where there is a liquid element sensor(LES) on a relatively short run of pipe. The LES is set for PW and turns on accordingly. Off of this piece of automation is a filter gate set to 65 seconds ( for tolerance ) - then filter output automation line runs to a powered liquid shutoff on the output line of the liquid tank.

Theoretically, the tank should fill up until the that LES sensor stays on for more than 65 seconds ( deducing the tank is full or the PW has backed up), At which point the filter off the LES holds the shut off valve open until the supply line becomes clear of liquid again. A nice caché mechanism, but it's entirely bound by the parameters of a clock sensor and accommodating filter. It doesn't take bypassing into consideration. This setup isn't demand oriented, more a storage solution for later

In other words, any other wonderful automation advice would be appreciated as I'm looking to expand this into a bypass-when-full type system rather than a passthrough.

Link to comment
Share on other sites

Simple solution: Use a bridge to force the content of the pipe to move that way. If the exit is full, the content will flow further down the line, where you place an element-sensor. If this one triggers, you know the tank is full, and something else can jump in. After the sensor, a second bridge linked to the input of the tank, so the content in this pipe-section can move out, once there is room again.

What you do, if the sensor triggers, is up to you. But thats the simple way. And yes, it only works with pure content. Mixed ones are a bit more tricky, and require other sensors or more complex builds.

Link to comment
Share on other sites

More simple solution: lay a pipe through the tank inlet directly, without a (prioritizing) bridge. A content will flow into the tank, then, when the tank is full, it'll flow further down the line - a sort of return pipe. And an element-sensor should be placed on the return pipe.

IMHO: TL;DR. I'm not so good in English, but my automation skill looks good enough to solve problems. If you only could post a screenshot of your piping, it could probably be much faster to understand how to help you. I mean: post a screenshot of status-quo, describe current behavior, describe desired behavior.
On the other side, looks like you wrote it being not at the computer and haven't tried to build the system you've just tried to explain in words. So it's possible you'll understand how to solve your problem yourself while building it.

Link to comment
Share on other sites

10 minutes ago, The Plum Gate said:

@SharraShimada, I'll get back to you on that.

I think @Saturnus has the setup element I'm looking for. Thank you.

Do tanks hold more than one type of liquid? Is that a thing ( is it a FIFO buffer, lol )?

Yes they do.  The nice thing is that they will sort-of consolidate the liquids, and give you mostly full packets on the output.  It can't be a FIFO buffer because you can alternate 1g packets of many gases, but still fill it to 150 kg, and get mostly full packets out the other end.  Not sure the exact algorithm for output, but I think it tries to give you a mix of gases.

Link to comment
Share on other sites

Btw, I can't remember if this is still the case, and can't check at the moment but it used to be that you could use pipe temp sensors as "any element" sensor as they would default to 0K if there was no element in the pipe, so effectively setting the temp sensor to active if above 1K would detect the presence of any element.

Link to comment
Share on other sites

16 minutes ago, Saturnus said:

Btw, I can't remember if this is still the case, and can't check at the moment but it used to be that you could use pipe temp sensors as "any element" sensor as they would default to 0K if there was no element in the pipe, so effectively setting the temp sensor to active if above 1K would detect the presence of any element.

Iirc they changed it to remember the last temperature if there is nothing in the pipe.

Link to comment
Share on other sites

23 minutes ago, Sasza22 said:

Iirc they changed it to remember the last temperature if there is nothing in the pipe.

If that is the case that means I can use it as single bit memory to save a particular binary state between saves without having to use miniature water clocks.

Link to comment
Share on other sites

7 hours ago, Saturnus said:

There's an even simpler solution that requires nothing more than a single bridge.

Bridge the output of the reservoir into the pipe feeding the input. Once the tank is full the input will overflow.

I've used a pair of sensors and a memory toggle.  The 'set' filter trips when there is no liquid flowing into the reservoir, turning the pump on.  Once liquid reaches the sensor, the 'set' turns off, but the memory toggle stays on.  The other sensor is on the output and detects when liquid is leaving the reservoir.  It trips the 'reset' on the memory toggle when no water is flowing out.  In this method, I only turn the pump on when the reservoir is empty, and I turn the pump off when it is full.

Link to comment
Share on other sites

53 minutes ago, Saturnus said:

If that is the case that means I can use it as single bit memory to save a particular binary state between saves without having to use miniature water clocks.

Unfortunately, the actual behavior, which I find very frustrating, is that the temp is reported as 0 K *and* there is no way to toggle the sensor state.  This is different from all the other sensors, AFAICT.  However, it does stay in the last non-empty state until a new packet arrives.

Link to comment
Share on other sites

Spoiler

image.thumb.png.0e1dc099982a86385772c6ffcc3159d1.pngimage.thumb.png.2960110db5479bd687d572f7adae7d1d.pngimage.thumb.png.53767a87dedeedc3416b0f7610dae587.png

I apologize for the clutter, but.. the tank is currently filling.  The toggle is actually turning my aquatuner off, rather than a pump, but its the same idea I posted above.  Once the tank is full, liquid will flow out the overflow pipe and turn the memory toggle on, which will turn the pump off.  The memory toggle stays on until the output senor detects no liquid, then it flips the reset and the pumps turn off again.  Here's a SS after the tank has filled completely but emptied enough to clear the feedback pipe:

Spoiler

image.thumb.png.a429b45eae8cad9741672da5a503cb83.png

 

Link to comment
Share on other sites

@SharraShimada, this is what I ended up with - which is a much more simplified tank overflow sensor than I was previously tinkering with.

Well, I managed on some derivative of Saturnus' recommendation - and maintained the overflow feature. I'll take a closer look at the other examples. Clutter is king of efficiency in good setups.

16 hours ago, Saturnus said:

There's an even simpler solution that requires nothing more than a single bridge.

Bridge the output of the reservoir into the pipe feeding the input. Once the tank is full the input will overflow.

Ok, so I tried that and there was some hitching, and the PW was just routing through the tank as if it was a single section of pipe - I don't doubt it would have filled up but it would have been running liquids around constantly. Basically I misunderstood your idea at first. ( at least how I had is set up). And there was no where to stick a sensor - so I used the intake priority method. Theory of operation is that PW won't go past the tank inlet unless the tank is full - at which point the sensor trigger the valve on the other side to open and let some of the overflow out. If I'm getting this straight?

5c4a9de5e003c_20190124173427_1(2).thumb.jpg.8cc0febf0fb634604f6c6d250498fff6.jpg

The bridge sends PW to the input, if it ever can't go that way, it would continue on down and around - across the pipe element sensor - which will turn on the shut off valve ( I mean activate it, since it defaults to the off position I didn't need to negate the signal ). Currently, It doesn't look like the fill mechanism uses any electricity - only on overflow. I'll have to check the behaviour once the tank start to top off ( Going to take a long time at the current rate, i've gone 30 cycles or so and have a 1/4 tank or so running the skimmer at night only ).

5c4a9de6940c3_20190124173436_1(2).thumb.jpg.049bc82f0d37558edb77afca809865fe.jpg

 

Link to comment
Share on other sites

You may want to add a buffer to the shut-off valve, or it will close again quickly once your overflow line clears.  The tank inlet will alternate from top and bottom with your setup, so once the tank begins to drain only a small amount of liquid would get released before losing signal again.

Link to comment
Share on other sites

6 hours ago, Nitroturtle said:

You may want to add a buffer to the shut-off valve, or it will close again quickly once your overflow line clears.  The tank inlet will alternate from top and bottom with your setup, so once the tank begins to drain only a small amount of liquid would get released before losing signal again.

That's the effect I was going for with the setup.

3 hours ago, fishoutofwater said:

could this be what you need:

20190125161917_1.thumb.jpg.e4d96b48aa2f466822d0175ea64ab974.jpg20190125161922_1.thumb.jpg.554c3f778d34f9714245a8b2a1992965.jpg

 

I like the quotes setup above also.

This ( a buffer ) is a solution to the first quote. I like it.

Link to comment
Share on other sites

9 hours ago, The Plum Gate said:

I really wish they would fix the degrounding exploits. I see so many youtubers using doors to do the strangest things. Typically a door has a full frame around it, and it should be fine open or closed.

I hope not, currently that's the only way of disabling buildings that don't have an automation input 

Link to comment
Share on other sites

30 minutes ago, Azunai333 said:

Same here. I believe that it will stay this way. How else could you automate flower pots with wheezeworts? :D

Well, by manually assigning dupes to run over and unplant them, of course! ... ... wait, nevermind. That's a dumb idea.  Keep the doors, please. :)

Link to comment
Share on other sites

On 1/24/2019 at 7:48 AM, Saturnus said:

There's an even simpler solution that requires nothing more than a single bridge.

Bridge the output of the reservoir into the pipe feeding the input. Once the tank is full the input will overflow.

The tank is just a bridge with capacity.  You can have the pipe go in and out of the inlet.  Once the tank is full, it will proceed along bypassing the tank.  Works like priority flow of a bridge

image.thumb.png.5ac67981627d9acf1f1b9d5941882d80.png

The liquid will only go to the right hand vent when the tank is full

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...