Aquatuner Coolant Bypass

[Lilalaunekuh]

14 minutes ago, DreamApart said:

The latter is so much prettier. I was using the former design all the time and I can't believe I survived. You've saved my life!

Keep in mind that my comparison was depending pipes inside the "hot chamber".

I did compare a "bad" version of a bypass with the version using the valve on top.

=> So if your happy about finding a version without a gap after the bypass was used, I would recommand the one @Saturnus posted earlier.

 

I like the design with the valve on top, but it means routing pipes across the hottest part of the chamber and requires a "suitable" material for the valve.(Most likely just steel ...)

=> I don´t like it

[BLACKBERREST3]

I think I cracked the code! Can someone try to explain to me why I can't break this? maybe it has something to do with the length of the pipe segments. The packet still slips by when cycled, but at least there's no gap. 

i'm gonna keep cycling for another 10mins, but I just can't seem to get it to overfill. Edit. I need a refresher on bridges and pipe priority cause I can't figure out why this works. Maybe it has 1 pipe segment of buffer So even if you do get it stuck in the AT, the blob that sits in the bridge fills the gap that was created or something like that.

[BLACKBERREST3]

This can't break either, but I can see where the gap was created. I don't use forums a whole lot, so I'm gonna practice my @ calling.

Hey! @Tonyroid, see if this double bridge bypass works for you.

[nakomaru]

The reason that won't overfill is because it is a 1 packet capacity reservoir. If there's ever a gap from the tuner storing 10kg, the buffer fills it before the top off. If there's ever overflow from the aquatuner storage, the input goes in the reservoir. Careful to build the reservoir pipes insulated from your hot room.

[BLACKBERREST3]

3 hours ago, nakomaru said:

The reason that won't overfill is because it is a 1 packet capacity reservoir.

Whats funny is that I had the same thought as a solution before I built it, but then I just never made that connection in my mind cause I didn’t know if it was even possible to create a buffer out of pipes. pipes are not my strong suit in this game. Are there any topics I can read about pipe priority?

[nakomaru]

It seems like you probably already understand all the basics such as these: 

For everything besides bridges and valves, the output only works when the pipe is entirely empty. So a shutoff and a valve both fully opened function differently.

The rest are clever applications like the mechanical filter, the in line packet stacker, bypasses, full/empty detection, etc...

[ZanthraSW]

18 hours ago, nakomaru said:

The reason that won't overfill is because it is a 1 packet capacity reservoir. If there's ever a gap from the tuner storing 10kg, the buffer fills it before the top off. If there's ever overflow from the aquatuner storage, the input goes in the reservoir. Careful to build the reservoir pipes insulated from your hot room.

If you use a reservoir pipe section along with an overflow alternate path with a normal pipe junction (so packets alternate going in) you can keep all packets moving. Here is an example where no packets remain still for more than 1 second. In steady operation, the bridge on the left is not used. Only when the extra packet is inserted by the aquatuner will that allow the packet to bypass. It gets shuffled into the loop, and the bridges on the right store the extra packet..

[WanderingKid]

This is my usual approach.  I've been using it the last five upgrades since I ran into this head first and froze a pipe accidentally.  I use the shutoffs and bridges to buffer while the aquatuner spins up.  I've exaggerated the piping for clarity, it can be done much more tightly.

Here's an example of very tight piping designs for smaller spaces:
 

Spoiler

 

 

 

 

[Chintam]

I haven't dabbled with bypass loops before so I did some investigation.

So my understanding is that the loop can become blocked when the aquatuner is paused while the packet of liquid is in the machine, which in turn creates a space in the loop for another packet to enter. When the aquatuner is re-enabled, this results in an extra packet of liquid in the loop causing blocking. 

So the conventional solution would be to create a buffer which can store/fill the extra packet. I believe I may have found a bufferless solution which utilises automation.

From what I've seen so far, a bridge is used to allow the loop to be topped up if a packet becomes missing.  I've replaced the bridge with a liquid shutoff which only activates if two or more packets are missing. This is achieved with a liquid element sensor and a FILTER gate such that it will only activate if two or more packets are missing. The filter time has been set to 1.7s. The logic chain is fairly simply: Liquid element sensor -> NOT GATE -> FILTER GATE -> liquid shutoff. 

 

This solution is probably not novel but I haven't seen this design in this thread so I thought I'd mention it. 

 

[Derringer]

The most bulletproof bypass I've been able to come up with uses a liquid valve (not a shutoff) beyond the aquatuner and directly below the steam room, and automation controls the aquatuner. If the aquatuner stops for any reason, including loss of power, flow continues circulating through the bypass. The actively-cooled and bypass pipes have precisely the same length, so there should never be stuck packets or inter-packet gaps. This also appears to be insensitive to whether the aquatuner is flipped left to right, and fits neatly into an arrangement of 4 tile high rooms.

I've added a coolant reservoir here to stabilise the output temperature, with the extent of output temperature swing largely determined by the length of pipe from the aquatuner output to the reservoir input. It has the added benefit of allowing the pipes to be drawn empty by using the door to disable reservoir output. If a reservoir is undesirable, the thermo sensor can be placed on the exposed pipe elbow in the steam room for near-immediate control of the aquatuner.

Spoiler

Aquatuner is steel, pipes inside steam chamber are ceramic, all other insulated pipes and tiles are igneous rock.

[BLACKBERREST3]

Nice build @Derringer, I never thought to use a reservoir on a door to empty pipes.

On 9/7/2019 at 10:23 PM, Tonyroid said:

 

I know this seems like a small thing but it's important to me to use things for my solutions that WORK and also that don't have extraneous flow-control hardware.

I feel the same, If we were to just solve the gap issue, then @ZanthraSW approach might be the answer to a liquid pipe only solution with no stagnant packets. I am not at my pc so I have not tested this yet for stagnant packets, but he says that there are none. If you do not mind having a 1-packet buffer that stays stagnant only when the AT is on, then the double bridge works all the same. If you are going to add a reservoir as a temp buffer, then that means you have the space to build on the buildings/automations layer in most cases, so whether you add a valve or shutoff is up to you and what design you choose. I also use the reservoir when I need accurate temps, but for smaller heat exchange builds, I will probably just use the double bridge or ZanthraSW’s design. As for the packet slipping when the AT spins up, I have a gut feeling this can be solved with valves, shutoffs, and sensors. However, I do not know what kind of circumstances that a simple reservoir design couldn’t fix. I feel sorry for the guy who is trying to build a bypass only on pipes and have all packets be in the same temp margin.

- ZanthraSW 

Spoiler

- double bridge

Spoiler

 

[TaeganRiles]

I'd like to share how I'm managing cooling loops going forward.  I like to call this a Liquid Loop Regulator.

Basically it's two Liquid Reservoirs, a primary and an overflow.  It's piped so that once it's filled to capacity it won't accept more liquid than the total system can handle.  When the output backs up it dumps the loop into the overflow reservoir to prevent pipe damage.

I've attached example images and blueprints (for use with the Blueprints mod).

Liquid Loop Regulator Test.blueprint

Liquid Loop Regulator - Left.blueprint

Liquid Loop Regulator - Right.blueprint