aquatuner Is aquatuner really cooling by 14 deg?

[Tobruk]

I have this set-up here set to only enable aquatuner if temp of the incoming liquid is over 25 (!) degrees, yet it still damages the pipes. For some reason packets as cold as -7 (!!!) degrees are coming out of the machine. HOW? :O

Is it the delay in how the automation propagates? Maybe it takes longer for the signal to get to the aquatuner than for a wrong packet to enter?

[EterniaLogic]

This is the problem of not having a Liquid Reservoir in front of the Aquatuner to average out the temperature before it goes in. You probably have a few cold packets at 7 Celcius if it is hitting -7 Celcius.

My setups are generally pretty simple. Before the packets even go in, the temperature is tested and if it is too cold, it gets shoved into a liquid shutoff. There are several ways of doing it. (Liquid Temperature sensors outputs are delayed by 1 packet, so it is exact when it gets to the next pipe) You do not need a reservoir, but it helps keep everything even.

 

Here is an example, ignore the fact that my pipes are pulling 20kg instead of 10kg because I have a mod that does that.

Liquid Pipes example: https://steamcommunity.com/sharedfiles/filedetails/?id=2106240473

Automation: https://steamcommunity.com/sharedfiles/filedetails/?id=2106240726

Temperature Setting: https://steamcommunity.com/sharedfiles/filedetails/?id=2106241007

[Tobruk]

Just now, EterniaLogic said:

Liquid Temperature sensors outputs are delayed by 1 packet, so it is exact when it gets to the next pipe

So if there's a packet at 7C followed by another at 30C, the aquatuner gets green signal for the 30C packet while 7C is processed? My temp sensor is right next to the aquatuner intake :/

Otherwise I can't see how a 7C packet can go through the temp sensor

[Glassyfo]

26 minutes ago, Tobruk said:

So if there's a packet at 7C followed by another at 30C, the aquatuner gets green signal for the 30C packet while 7C is processed? My temp sensor is right next to the aquatuner intake :/

Otherwise I can't see how a 7C packet can go through the temp sensor

Yes.

[Edoc_]

2 hours ago, Tobruk said:

Is it the delay in how the automation propagates? Maybe it takes longer for the signal to get to the aquatuner than for a wrong packet to enter?

It could be because the signal is delayed. I use 2 not gates to create a little delay between the main signal and the signal going through thoses gates.

[Chipplyman]

I would guess that this is because the automation input on the aquatuner controls "should I pull from my input buffer, chill, and place in my output buffer" and not "should I pull liquid from my input pipe and place in my input buffer."

As in EterniaLogic's design, your temp sensor automation needs to control whether too-cold liquid can even get to the aquatuner, not whether the aquatuner should run. If a pipe is running past the aquatuner liquid input, it will contain hot liquid that couldn't be cooled e.g. because the aquatuner output is blocked or some other automaton logic says "I don't want any heat transfer now."

[psusi]

I've always just used a temp sensor on the pipe right before the input to the AT and had it disable the AT ( with a pipe bypass if the AT is disabled ) and never had a problem.  I normally run polluted water and have set the sensor as low as 4 C at least.  I also normally run it in a closed loop with no reservoir but also without multiple branches so you don't ever get some hot and some cold packets alternating.  The temperature of each incoming packet of pwater is about the same since it's all coming from the same radiant pipe heat exchanger.  It might be a problem if you have some packets coming from a hot source merging with packets coming from a colder source and so the temperature between each packet varies widely.

[Tobruk]

Just now, psusi said:

I've always just used a temp sensor on the pipe right before the input to the AT and had it disable the AT ( with a pipe bypass if the AT is disabled ) and never had a problem

Which is what I have done? Look at the images.

[psusi]

3 minutes ago, Tobruk said:

Which is what I have done? Look at the images.

The left one looks that way, but not the right one.  Where is the damage taking place?

[Tobruk]

Just now, psusi said:

The left one looks that way, but not the right one.  Where is the damage taking place?

The right aquatuner is protected in a different way. The damage is occurring right after the output of the left aquatuner.

[Christopherus]

Is your overflow pipe sometimes stucked? If so, the AT will proceed the water that should habe gone to overflow but couldn’t and stayed in the intake pipe of the AT. 

[TheMule]

Something I don't get in the picture... the gate above the left AT is a AND gate yet its output is green while one of the inputs is red? Is it possible there's an extra wire behind it? Or behind the AT?

[werecat]

Something unrelated that is dragging down your efficiency is the liquid bridge hopping from inside the steam room to where your turbines are. That bridge is actually bringing heat from inside the steam room into the steam turbine room, which is probably not what you want. The same thing is happening with one of your automation bridges into the lower room, but more efficiently since it is made of refined metal.

[Saturnus]

Put a NOT gate between the sensor and the aquatuner. And only a NOT gate. Reverse the ON condition of the sensor. And let anything else than may shut down the aquatuner be a positive signal on the sensor side. That should solve your issue.

@Tobruk Apologies for not noticing it right away. You pipe temp sensor is directly on top of a liquid bridge output. That means it will never work regardless of what you set it to.

[TheMule]

5 hours ago, Saturnus said:

Put a NOT gate between the sensor and the aquatuner. And only a NOT gate. Reverse the ON condition of the sensor. And let anything else than may shut down the aquatuner be a positive signal on the sensor side. That should solve your issue.

@Tobruk Apologies for not noticing it right away. You pipe temp sensor is directly on top of a liquid bridge output. That means it will never work regardless of what you set it to.

Really? I thought the input of of a bridge is the part where there's never a packet (unless the loop is clogged). I can't tell for sure 'cause I tend not to put sensors over bridges anyway.
Still it doesn't explain the AND gate.

[Nilfsen]

Double check if its not or gate?

As with red and green input it gives out green. So, either you've made screenshot exacly when its toggling or its not and gate.

 

Any mods that could affect logic?

[Trippleass]

I'd double check whether the sensor works correctly, since it's put right on the bridge output and I never do that for some reason. The reason might be it doesn't work.

[Tobruk]

52 minutes ago, Nilfsen said:

Double check if its not or gate?

As with red and green input it gives out green. So, either you've made screenshot exacly when its toggling or its not and gate.

 

Any mods that could affect logic?

I've reworked the setup since the previous replies to avoid the sensor on bridge's end and adding a liquid reservoir, which works, but now I won't be able to check why the AND gate was firing green when it wasn't supposed to in the original setup.

[TheMule]

3 hours ago, Nilfsen said:

either you've made screenshot exacly when its toggling or its not and gate.

That's a 0.2s (1 tick) window. Still it's possible. But since the problem seemed to be the AT activating when not supposed to it was a good candidate.

[Saturnus]

22 minutes ago, TheMule said:

That's a 0.2s (1 tick) window. Still it's possible. But since the problem seemed to be the AT activating when not supposed to it was a good candidate.

One tick is 0.1s. Not 0.2s.

[TheMule]

1 hour ago, Saturnus said:

One tick is 0.1s. Not 0.2s.

Thanks. So the window is even smaller.

 

Is 0.2s a thing of the past? I'm sure I've read that numerous times. E.g. first match on google: 

 

[beowulf2010]

1 hour ago, TheMule said:

Thanks. So the window is even smaller.

 

Is 0.2s a thing of the past? I'm sure I've read that numerous times. E.g. first match on google: 

 

There's 2 different "ticks". For everything that isn't automation, there are 5 ticks per second. For automation, this is doubled to 10 ticks per second. 

[cpy]

Bad design input temperature check have to be done right before entering AT. You need to be sure that last thing that check temperature can't be influenced by something else. Negative result must not go through AT input slot, it must be routed elsewhere. You have to have blocked pipe via automation to AT unless you are 100% sure you want to shove that liquid in it.

[Tobruk]

Just now, cpy said:

Bad design input temperature check have to be done right before entering AT. You need to be sure that last thing that check temperature can't be influenced by something else.

Although I agree with you about the blocked pipe (let's say via a shutoff valve), the sensor is placed right before it enters the AT.

[TheMule]

2 hours ago, beowulf2010 said:

There's 2 different "ticks". For everything that isn't automation, there are 5 ticks per second. For automation, this is doubled to 10 ticks per second. 

ooh makes sense