automation Useful Automation Gadgets

[Moggles]

@Layta61 This technique should work with a natural gas generator.

 

[Layta61]

41 minutes ago, Layta61 said:

I also make weight plate that tirggers the natural gas pump when dupe try to fill low batteries with manural generator but still not effective.
Any one have something like smart batteries ?

like i said i did that but i dont think its so efficient . just looking for more cleaner setup

[Moggles]

3 hours ago, Layta61 said:

like i said i did that but i dont think its so efficient . just looking for more cleaner setup

Without smart batteries a really clean set up seems difficult.

Here's another approach that works for consistent power usage. You'll have to tinker with the timings depending on your consumption.

I found that 60 seconds idle time followed by a 10 second burst from two hydrogen generators will power a couple of fridges on one battery without wasting power. Battery level peaks and troughs without ever hitting zero or maxing out. For extra charge power I used a power shutoff to cut the wire to the consumers while the generators are active.

The same thing will work with natural gas generators you've just gotta twiddle the dials until you reach equilibrium .

Spoiler

 

[Moggles]

@mickaux Just an update after a lot of fiddling around I got a set up for a work shift airlock similar to yours.

Thanks to all the other posters too for showing how to make use latches and stuff. It's all gone in to making my perfect "get out there and do work" air lock. Nisbet will be out there until 70 o'clock. She seems happy enough to stand around in 90C oil waiting for a job.

Seriously exo suits combined with lockouts is a game changer for me. It eliminates all the pain of seeing them return to base to do some random task or eat. She wont be eating until the door opens in the evening. In fact the mess hall wont be open til 70% either, hence the food warning.

I think now I have the timing sorted I'll try to connect it to a more sophisticated automated water lock or something.

Spoiler

[mickaux]

@Moggles I'm glad you found a good use for it! Actually, closing the airlock behind dupes seems to be a far better setting - you don't have to worry about a buffer gate time, and it eliminates an irrigating issue with dupes just going in to grab some garbage they dropped while going to sleep and coming back before the door lock

[TheScaryOne]

I'm trying to wrap my head around automation, but not getting anywhere. I want to have three automatic doors in series, and have the middle one open after the other two are closed, but have it close before one of the side doors to prevent destroying gas. If it could be made directional (to force gasses into or out of a room with a pump) that would be even better, but fine if not. Is this possible without having the doors also be able to send an active signal?

The purpose of this would be to create a vacuum temperature tight airlock between biomes.

 

Edit: Here is a lock that works as such:

 

[bzgzd]

36 minutes ago, TheScaryOne said:

Is this possible without having the doors also be able to send an active signal?

If you put weight sensor under the door you can know when they are open or closed (0kg or 400kg).

36 minutes ago, TheScaryOne said:

I want to have three automatic doors in series, and have the middle one open after the other two are closed, but have it close before one of the side doors to prevent destroying gas.

Automatic and closed is the problem here...
You want doors to be on "Auto" settings so that duplicants can see path through them and same time control their behavior with automation. (sometimes open, sometimes auto closed)

But when automation is used on doors it can currently make them only "Open" or "Locked"...
I think people complained about that as soon as automation update was released. Some method to make doors "Auto" with automation is missing.

So simple airlocks like that are not possible because when you lock doors with automation duplicants would not try to go through them.

 

[Ketmol]

You can get the same result as from the RS-NOR gate ( the combination of several different logical circuits) using only one single XOR GATE

[Nativel]

On 20.11.2017 at 11:05 PM, Layta61 said:

I also make weight plate that tirggers the natural gas pump when dupe try to fill low batteries with manural generator but still not effective.
Any one have something like smart batteries ?

No smart batteries, and only way to not loose natural gas is to, well it's too much to type so I've create video for you:

 

[Arash70]

They should teach this topic in university :))

[TheScaryOne]

@Saturnus Is there a benefit to your rolling compressor design here: https://imgur.com/2t7YdMU versus the one I got off of Reddit here: https://imgur.com/a/nLTWa (the buffers are set 2,4,6 from bottom left to top right in the actual build picture, the filter is 2). I guess I could just build both in debug and find out if you don't answer.

 

Edit: Both units worked flawlessly and compressed 8 tiles at 160kg to 4 tiles at 320kg without loss. Saturnus's design was more efficient at drawing out a larger percentage of gas to compress, but both designs pull to vacuum at about the same time. I'd guess Saturnus's would have higher throughput in an active system.

[qda]

Hello everyone, hope you're doing well !

First, this is my first post on any forum of any kind, please spare me for my english, for the fact that this post might not be within the right topic, and for any mistakes I'll probably make sooner or later.

With the recent addition of a memory bit, more space efficient logic circuits can be designed (a single memory bit/RS-latch used to take at least 12 tiles, instead of just 4 now). Below is a design of a 4-bit counter that takes 10x22 tiles (which I believe is of a size and complexity low enough to make it usable before ultra-late game and such as it doesn't take two entire screens, which is highly unpractical). The goal here, if anyone is interested, is first and foremost to share it with you all, but it would also be to tinker with the design and hopefully give birth to new design ideas and/or practical applications of a counter in ONI. I am not an experienced player, so here is just a few ideas I can think of with my basic knowledge where a counter can actually be useful :

- Limiting the number of dupes able to go in a designated areas, without having to specifically select them through door permissions.

- Preventing dupes to go to an outhouses area, for morbs spawning purposes, although I feel morbs are actually not really useful unless you get thousands of them.

- Whenever you want to go ham with atmo suits, and with different sets of docks and exosuit checkpoints, this often leads to really inefficient pathing issues which I feel could be partially resolved with counters and duplicant checkpoints.

- And potentially many, many more, but my imagination is kind of limited.

So here is the counter, animated, and with a display made for fun to highlight the 4-bit counting.

Note that while the display takes more or less the whole screen, the counter itself is just the 10x22 box on the left. Also, the gif doesn't have the reset switch, as presented below, but it doesn't make any difference animation wise and I am way too lazy to make another one. The 4 wires surrounding the counter and the display is the 4-bit bus. Now, let's get to the counter design itself :

First, on top are two switches, the one on the left being the trigger for the counter (in a real game that would be the sensor of your choice), and the one on the right being the reset switch for the counter. Hitting this button, as you could have guessed, resets the entire counter to its initial state (i.e. a value of zero, or all bits being 0). As far as my knowledge goes, this architecture is not sensible to timing, meaning you can reset the whole counter at any time without breaking it (which isn't something straightforward since the counter itself is somewhat sensible to some timings within it).

Then, just below the trigger/reset switch are the oscillator (for easy design purposes) and the pulse generator (which you would use with a sensor in a real game), both of which are working great to feed the counter. Then, from top to bottom are the 4 stages having the consecutive bits as their outputs (1,2,4,8). As you can see, each successive bit stage doesn't take much more place than its predecessor, which means that the total counter's surface would be almost linear (vs. the number of bits), while the actual count is exponential (up to what number/number of bits could we go ?). I won't describe it here more than the image does, but if anyone is interested, I'll happily discuss the design and the minor tricks needed to make it work as intended. Just for those who might be afraid of this, the basic idea for a given bit is to route every bit output to the refresh port (all through AND gates) of the current memory bit, and to route every previous bit output along with the opposite of the current bit output (also through AND gates) to the input port of the current memory bit.

In order to make it work in a real game situation, one just has to feed a pulse trigger to the counter (any kind of sensor going through a pulse generator), and feed the max number wanted to the input port of a memory bit, and the reset switch to the refresh port of the same memory bit.

About the architecture itself, I've tried for a bit (no pun intended) to optimize the surface the whole counter takes, but this is by no means optimal and probably can be improved. However, I feel like the height/width ratio cannot be tinkered much with, due to the configuration of the input/output/refresh ports on the gates being fixed.

On a side note, if you disable the reset/refresh port of the first bit, you get an odd counter, effectively counting twice every pulse/trigger thus dividing by half the number of values one can count. Don't know if this might be useful or not for some purposes, maybe someone could find one.

Again, this is by no means the best way (or more efficient, or anything, really) to do this, and this kind of counter may even have already be done, but since I haven't seen much information about it and I believe it could find some use in some niche real game situation, I thought it was a good idea to share it with you guys.

Thank you everyone

[Saturnus]

3 hours ago, qda said:

Again, this is by no means the best way (or more efficient, or anything, really) to do this, and this kind of counter may even have already be done.

Indeed. Go back one page where I posted the compact 2-bit counter for example. In another thread I posted the compact 4 bit counter as well.

[qda]

11 hours ago, Saturnus said:

Indeed. Go back one page where I posted the compact 2-bit counter for example. In another thread I posted the compact 4 bit counter as well.

 

Nice one ! Didn't even thought about it, this is a great and compact design, cheers to the ones who contributed to it.

However, I was trying to design some circuits with some real game purposes, and since yours is asynchronous and probably takes a lot of time to update the successive bits, two inputs within less than 1s (or even more ?) would make the counter jump values, or, depending on the timing, put it in a random state, which I believe makes it unpractical to most situations encountered in a real game. The one I showed earlier is fully synchronous, and can handle two inputs within less than a second without any issues, the lowest interval possible being 0.1s due to how the game works.

Also, if I'm not mistaken this synchronous architecture allows down counting easily, which could be useful : I'm trying to play with a two 2-bits balanced up/down counters setup that could indicate a signed (or relative) change of a value within a given period of time. The purpose I can think of are most of the cooling setups, which takes typically a few to a few tens of cycles, and require one to manually and regularly check the temperature to assess if it is effectively cooling or not. I would be happy to know with a just a quick glance if one of my setups is doing its job properly or not.

Anyway, thanks for the reply and the great design, and let's all have fun with automation !

[Saturnus]

3 hours ago, qda said:

Nice one ! Didn't even thought about it, this is a great and compact design, cheers to the ones who contributed to it.

However, I was trying to design some circuits with some real game purposes, and since yours is asynchronous and probably takes a lot of time to update the successive bits, two inputs within less than 1s (or even more ?) would make the counter jump values, or, depending on the timing, put it in a random state, which I believe makes it unpractical to most situations encountered in a real game. The one I showed earlier is fully synchronous, and can handle two inputs within less than a second without any issues, the lowest interval possible being 0.1s due to how the game works.

Also, if I'm not mistaken this synchronous architecture allows down counting easily, which could be useful : I'm trying to play with a two 2-bits balanced up/down counters setup that could indicate a signed (or relative) change of a value within a given period of time. The purpose I can think of are most of the cooling setups, which takes typically a few to a few tens of cycles, and require one to manually and regularly check the temperature to assess if it is effectively cooling or not. I would be happy to know with a just a quick glance if one of my setups is doing its job properly or not.

Anyway, thanks for the reply and the great design, and let's all have fun with automation !

You can test the set up. Though the outputs takes a bit of time to propagate through all the bits, it will still accurately count all pulses applied to the input. Even if they are 0.1s each because the input NOT-AND gate on each bit is an edge detector.

[Le_Bison]

 

Can someone help me?

I'd like to make an automation in which a dupe have to step on the weight plate twice before the system activates. The first click is to "set up" the logic of the button and the second click, actually executes it.

I'm using this in my kitchen setup. The dupe has to step on the plate twice to open the mechanized airlock door.

Spoiler

Thanks in advance.

[Oozinator]

On 25.4.2018 at 6:47 AM, Saturnus said:

Indeed. Go back one page where I posted the compact 2-bit counter for example. In another thread I posted the compact 4 bit counter as well.

@Saturnus?

[no.1]

This project takes care of the meteorite strikes. The first automation closes the doors when the meteorite alarm is first received and ignores any following alarms. First two buffers set the ignoring time (205s), and the filter gate is 0.1s, the middle buffer sets time for keeping the door closed (200s), atmo sensor is for manual closing of the doors.

The oscilator turns the tower on for 7s and then off for 25s saving power, and output of the radar tower is buffered for 32s (sum of oscilator gate times) and filtered for 15s (I have two towers). Works perfectly