Jump to content

Aquatuner Loop Breaking Pipes


Recommended Posts

Like the title says. I've a a loop of two aquatuners to cool water for my farms, but occasionally a packet gets past my automation and is cooled below 0, breaking the exit pipe. How are you guys building your loops?

image.thumb.png.ec13e89116ccae6348423653711c4a11.png image.png.35ec78fd36cac244c081117bb4e22701.png

The first sensor is set to above 40C, the second is set to above 25C, the third to below 26C.

2 minutes ago, Prince Mandor said:

There is a sensor in game, to detect temperature of liquid in pipe. Put this sensor just before AquaTuner and let it turn off AT, if liquid is already cold enough. And make some bypass for liquid to go past AT

Already have that.

1 hour ago, bleeter6 said:

Like the title says. I've a a loop of two aquatuners to cool water for my farms, but occasionally a packet gets past my automation and is cooled below 0, breaking the exit pipe. How are you guys building your loops?

image.thumb.png.ec13e89116ccae6348423653711c4a11.png image.png.35ec78fd36cac244c081117bb4e22701.png

The first sensor is set to above 40C, the second is set to above 25C, the third to below 26C.

Is your loop completely full ? If there's gap between packets, it could happened that the AT doesn't run when it has to.

Then you ended cooling always a packet that doesn't necessarily need to be cooled.

1 minute ago, OxCD said:

Is your loop completely full ? If there's gap between packets, it could happened that the AT doesn't run when it has to.

Then you ended cooling always a packet that doesn't necessarily need to be cooled.

Ya it is. I have an input pipe to a bridge that adds packets if there's space for it.

1 hour ago, bleeter6 said:

image.thumb.png.ec13e89116ccae6348423653711c4a11.png

I also don't see the first image and I don't see one here either. 

One thing I would be careful of is that the alternative route for the liquid doesn't back up.  That is the usual cause for this.  Even one second of the flow stopping can cause this.  You can use bridges to create a buffer.

1 hour ago, Zarquan said:

I also don't see the first image and I don't see one here either. 

One thing I would be careful of is that the alternative route for the liquid doesn't back up.  That is the usual cause for this.  Even one second of the flow stopping can cause this.  You can use bridges to create a buffer.

Does this image work?

image.png.1b3c4ffb0d75fc1890d5a828778b1cae.png

1 hour ago, bleeter6 said:

Does this image work?

image.png.1b3c4ffb0d75fc1890d5a828778b1cae.png

These systems are notoriously touchy.

My bet is that the T-junctions are the issue.  You may want to rework the pipes to avoid them.  These can cause alternating flow from the two inlets.  In this design, the alternative flow (the path that skips the aquatuner) can be blocked if the aquatuner turns on by this alternating flow, thereby allowing a packet that is too cold to sit on the inlet of the aquatuner while the next packet triggers the sensor, letting it in to the aquatuner.  Generally, you want to avoid T-junctions, and use bridges instead to merge two pipes in a flow-sensitive context like this one.

It is also possible that the pipe is overfilled.  This can happen because bridges teleport the liquid from the inlet to the outlet in such a way that the liquid is never actually in the inlet of the bridge.  If the outlet is full, then the liquid sits on the inlet until the outlet is empty, causing there to be an extra packet of water in the pipe.  This can cause a chain reaction where there is too much liquid for smooth flow, as each length of pipe needs 1 second to start moving again, causing stop and go "traffic".  When the game is running, is there any stop-and-go movement in the pipes?  It can be hard to tell if this is happening in a picture.

ALSO:  There is a bug in the physics.  You should look at this thread:

Spoiler

 

@Tonyroid made a video where he makes a stable aquatuner loop.  It does not take in to account the bug I mentioned above though.

Spoiler

 

 

2 hours ago, bleeter6 said:

Does this image work?

image.png.1b3c4ffb0d75fc1890d5a828778b1cae.png

Yes.

I also think the junction is the issue.

You should try this kind of post-AT junction (not this exact layout, but the bridges principle and order) :

unknown.png

Also keep in mind no bridge output should end on the sensor. The reunification of the flow should take place before the sensor.

11 minutes ago, OxCD said:

Yes.

I also think the junction is the issue.

You should try this kind of post-AT junction (not this exact layout, but the bridges principle and order) :

unknown.png

Also keep in mind no bridge output should end on the sensor. The reunification of the flow should take place before the sensor.

So I should move my bridges one tile to the right, and connect on the tuner output?

2 hours ago, Zarquan said:

 

@Tonyroid made a video where he makes a stable aquatuner loop.  It does not take in to account the bug I mentioned above though.

 

It's true. Even when I've thought I had these loops figured out, I didn't have these loops figured out.

2 hours ago, bleeter6 said:

So I should move my bridges one tile to the right, and connect on the tuner output?

Yup, but my advice is (flow direction) AT output first, then bypass bridge output, and then sensor of the next AT.

Bypass bridge output first could be problematic as far as I know.

I should also precise that I always use 2 bridges, as shown in my pic, and I never had any issue anymore. That doesn't mean it's the only good way, that doesn't mean connecting directly the AT's output to the bypass bridge's output will not also work, but that means  you should be careful because... because we never know ! (and I'm worried by default ^^ especially with closed setup)

6 hours ago, bleeter6 said:

 

image.thumb.png.4058ba239451038144e6faa85354501d.png

Can you explain the meaning of automation logic? What logical element is it?

Your sensor connected to input and to output, does it really planned as "electrical OR", or they connected by mistake?

5 hours ago, Prince Mandor said:

Can you explain the meaning of automation logic? What logical element is it?

Your sensor connected to input and to output, does it really planned as "electrical OR", or they connected by mistake?

image.png.89d0a5e99226560e7da0b6d6e26a1298.png

Top two are input, bottom is output. Temp sensor for right input, negative element detector for the left, just in case.

13 hours ago, bleeter6 said:

 

image.thumb.png.4058ba239451038144e6faa85354501d.png

On this screenshot you have an automation wire connecting sensor directly with output, and wire connecting sensor with input of AND.

So wire on top of screenshot has no meaning. If sensor green, result be green no matter what, because it is directly connected to output by wire. If sensor red, then AND gate will be red.

So, can you explain what you mean by this logical circuit?

25 minutes ago, Prince Mandor said:

On this screenshot you have an automation wire connecting sensor directly with output, and wire connecting sensor with input of AND.

So wire on top of screenshot has no meaning. If sensor green, result be green no matter what, because it is directly connected to output by wire. If sensor red, then AND gate will be red.

So, can you explain what you mean by this logical circuit?

What are you talking about? No sensor is connected to an output.

9 minutes ago, bleeter6 said:

What are you talking about? No sensor is connected to an output.

The pipe sensor is connected to the output of the AND gate.  You can see the green wire between the sensor and the output, and it is also connected to one of the inputs of the AND gate. 

I believe the effect of the logic circuit will be as follows:  If the automation wire coming in from the left is true (green), when the sensor is activated, the output of the AND gate will remain true until the incoming signal from the left becomes false.  This will likely lead to broken pipes eventually.  Additionally, the aquatuner will turn on with a green signal from the sensor regardless of the state of the wire from the left.

I think the intent is to be able to turn off the aquatuners off with the signal from the left, probably to prevent it from draining power in the event of brownouts.  If this is the case, then the connection from the pipe temp sensor and the output of should be removed.

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...