How to make a battery sensor

[Saturnus]

A request I often see people writing, or saying on streams, is they wished there was a way to check battery status and use that to turn on extra generators or something. There is. You can use the temperature inferred method.

The principle is that if you battery bank is large batteries as it normally is, you can have a small battery in a thermally sealed room with a constant cooling so that the room temperature at some point stabilize. Batteries drain at the same rate so the small battery will run empty when the large batteries are at 75%. Abattery without charge in it does not give off heat, so when the small battery is empty the temperature in the sealed small battery room will drop. You can then just place a sensor in the room that turn on external circuit to enable more generators to run.

Here's a save file and pic to illustrate the concept.

EDIT: For some reason when I load the game file again. Stable temperature drops significantly. Maybe because I ran it on turbospeed to stabilize temperature in the first place so it'll not work out of the box. You need to turn on the master switch. Set the sensor temperature to about 36.5C. And then let it run for 2-3 cycles... not on turbo speed. Check the temperature again. Set the sensor to 0.2C under that, and throw the switch.

EDIT2: If you experiment with the principle yourself you might find the heat output of a single tiny battery too fidgety as it requires very tightly balanced cooling. Try instead with 4-8 tiny battery. The heat production is much higher and thus exact cooling balance becomes much less significant. 

temp inferred battery sensor.sav

[CfSapper]

Ohhh this is nice! and bloody brilliant! +5 Internets to you!

[Sevio]

That's an interesting concept, what kind of response time would this sort of system give?

Another way I was thinking of which does unfortunately drain extra power could perhaps be using the liquid pump clock I used for my tepidizer oil refinery experiment.

If the pump and vent could be set up in a way that a powered pump sucks up the liquid before it touches the hydro switch, a drained battery could then be detected with the hydro switch turning on or off. The power drain could be reduced by increasing the length of the liquid pipe, at the cost of increasing response time.

[Saturnus]

2 minutes ago, Sevio said:

That's an interesting concept, what kind of response time would this sort of system give?

Once you've reached a stable battery room temperature the response time can be set to extremely short if you build everything in the room from wolframite and you set the trigger point just below the stable temperature. Literally just a few seconds. Check the concept file.

Normally the cooling water would be from lavatories and showers or anything else that has rock solid output temperature, so no power is wasted on that part normally but here I needed something that would drain the batteries to illustrate the concept, so hence the water pump.

Normally you'd have a lot more than a single battery in your battery bank, so what is really needed is not a fast reaction switch. But more an indication that battery charge state is dropping. What you do with that information is up to you. You could can have it activate a flip-flopped secondary power generation, or just light indicating battery state warning, or something else. Entirely up to your needs.

[Sevio]

So it looks like we can develop a useful battery sensor with this temperature method then.

An unfortunate flaw I see with the existing sensor system is that to switch a large section of your power grid based on one particular sensor (let's say a "master switch", you have to route the power for all of that to the sensor and then back to where the consumers are... As the game develops, more cases will probably turn up where you might want to have consumers and sensors be far apart from each other and this would be the point where a "redstone" update could really shine. Either that or avoid the need to have sensors and consumers be far apart if redstone-like mechanics are too far out of scope for the game.

[Saturnus]

Well, not really. The sensor could just as easily remotely trigger a gas filter that release gas in a small room with an atmo-switch that does the actual switching somewhere else on the map.

[Whispershade]

Bah! You beat me at posting battery meter. I had a scheme I thought about posting friday, but I dont run debug so didnt get around to it. It was along the lined of @Sevio pump clock.

Interest work, good job. If anyone is interested, I will a picture up of my idea later.

[Saturnus]

I'm still not seeing how your ideas will detect battery state of a large battery bank as my idea does. Both of your ideas work on the principle of draining a battery and then something happen. However, that does not help you to detect the battery state of a large battery bank because if you had a small battery connected you'd still get power from the battery bank.

I think you should post your ideas here because to me, it doesn't sound like it's going to work the way it's intended. The only way to do it is by inferring it from temperature fluctuations.

The old "battery drained" switch is btw much better implemented by a liquid valve as it uses less power. When the valve lose power, the bypass it's forming is broken and water flow through a vent to trigger a switch. That's been around for a long time.

[Whispershade]

19 minutes ago, Saturnus said:

The old "battery drained" switch is btw much better implemented by a liquid valve as it uses less power. When the valve lose power, the bypass it's forming if broken and water flow through a vent to trigger a switch. That's been around for a long time.

Yeah. The pump is used as an inverter for the battery out signal to open the filter path to the generator. The batteries are set up with a transformer as a diode to partition the bank in the ratio of the whole bank's capacity.

I hadnt seen anyone post it and a lot of people annoyed they cannot direct the flow to their generators when batteries low like coal.

[Saturnus]

28 minutes ago, Whispershade said:

Yeah. The pump is used as an inverter for the battery out signal to open the filter path to the generator. The batteries are set up with a transformer as a diode to partition the bank in the ratio of the whole bank's capacity.

The problem is that still doesn't work the same way. I also used to do it like that, and have done so for months but it doesn't work as a charge state indicator for a large battery bank.  All it does is drain the battery in front of the transformer before those after the transformer, so they're not directly connected to each other, so you don't actually know anything about the state of the battery bank.

What that method is useful for is if you have two large battery banks. And they're separated by a transformer then you know when the first part is drained and you can kick in power generation. This means you toggle power generation on and off more or less completely whereas the method I show here works on the continuous power generation principle with top off power generation enabling when batteries start to drain.

[NurdRage]

This is bloody brilliant.

[Whispershade]

4 hours ago, Saturnus said:

The problem is that still doesn't work the same way. I also used to do it like that, and have done so for months but it doesn't work as a charge state indicator for a large battery bank.  All it does is drain the battery in front of the transformer before those after the transformer, so they're not directly connected to each other, so you don't actually know anything about the state of the battery bank.

What that method is useful for is if you have two large battery banks. And they're separated by a transformer then you know when the first part is drained and you can kick in power generation. This means you toggle power generation on and off more or less completely whereas the method I show here works on the continuous power generation principle with top off power generation enabling when batteries start to drain.

Yes. It detects that a battery has drained, but it can only do that if there is not something else feeding the system to keep the pump going. I do see what you are saying.

I don't believe showers and lavs would work because I heard they're now variable to the building temp. But I know carbon skimmers and potentially refined petroleum could for your purposes as they're steady state outputs.

Anyway, cool idea.

[Saturnus]

I've refined the system bringing it out of mere theoretical state to an actual proof of concept.

The two most important improvements are that the sensor room now has 2 tiny batteries so the heat generation is double that of a single one making temperature variations more pronounced. And I've made it a vacuum with only a puddle of crude oil on the floor, making it far more reliable.

You can increase the number of tiny batteries to 4 or maybe even 6 which makes cooling water temperature variations more or less irrelevant. It's already pretty stable with just 2 batteries as long as the cooling water doesn't vary by more than +-2C.

In a survival game the main battery bank would naturally be far far bigger than just 3 large batteries which gives the system amble time to react. In this test setup it's far more sensitive than it needs to be but making a larger set up would make testing an excruciating experience as I prefer to test on survival game speeds as running turbospeed often gives strange results. 

The system only has a single main line connection point, can be placed everywhere on the map, and since it's a vacuum inside there's no temperature exchange through the heavi-watt connection point.. The sensor connections are pretty self-explanatory.

temp sens 2.sav

[Fury1SOG]

Your battery sensor is excellent and works well with the new Automation sensors!

It is awesome to see the batteries slowly deplete and then get topped off automatically.

I use constant 40C  PH2O from a Carbon Skimmer as the cooling water. The batteries stabilize at around 65C fully charged. 

[MAD-Kuzia]

You can make an even more precise detector with Manual Generator (aka hamster wheel) and Weight Plate. Put hamster wheel(s) into dead-end corridor, put some Weight Plate(s) before them, set desired % recharge threshold and maximum priority. Connect Weight Plate exit to BUFFER-gate, and exit of BUFFER-gate  to NOT-gate, connect hamster wheel to NOT-gate, and power generating machines to BUFFER-gate.

When battery charge drops below adjusted percentage dups will come to operate hamster wheel and press on Weight Plate. This will disable hamster wheel and enable power generation. Timer on BUFFER-gate needs to be set experimentally, to avoid  running out of battery-juice till night, when no dups operate hamster wheel.

[Saturnus]

7 minutes ago, MAD-Kuzia said:

You can make an even more precise detector with Manual Generator (aka hamster wheel) and Weight Plate.

Still needs dupe inaction. The one I've made is fully automatic. No dupe inaction needed at all.

[Kabrute]

So i've been toying with your idea @Saturnus and if you ran all of your generators (backups included) through a single large battery, then through a transformer bank to the rest of your battery packs that one initial battery holds the least charge (not charging at all if batteries past the transformer need power) so this could use a large battery as the thermal lever and would run out % ahead of your battery bank where % is the ratio of batteries in the bank to this one.  say 1 cycle to drain that main battery with 4 batteries on the supply rail you have your 75% charge variance back   Generators Single large battery in cold room with sensor, transformers(max 5=20kj), normal battery bank, normal supply transformers

[Saturnus]

10 hours ago, Kabrute said:

So i've been toying with your idea @Saturnus and if you ran all of your generators (backups included) through a single large battery, then through a transformer bank to the rest of your battery packs that one initial battery holds the least charge (not charging at all if batteries past the transformer need power) so this could use a large battery as the thermal lever and would run out % ahead of your battery bank where % is the ratio of batteries in the bank to this one.  say 1 cycle to drain that main battery with 4 batteries on the supply rail you have your 75% charge variance back   Generators Single large battery in cold room with sensor, transformers(max 5=20kj), normal battery bank, normal supply transformers

Small battery have the same heat output as a large battery but in a lot less volume. That's one of the reasons why it's really good for this purpose. You simply get a much more responsive system.

[Kabrute]

tiny area, ok, sure, thought it was because it discharged sooner, my bad

[manu_x32]

Thanks for this idea.  I think I made a version that needs no power at all for cooling.

A morb on the left side to make sure the cooling room always has gas since wheezes delete it. Wheeze gets shot off by door (thanks for that trick too) to keep a relatively constant cooling temp. Battery room just has 2kg of hydrogen. tungsten plates to link the 2 rooms.  Only problem is gas pressure keeps increasing in the cooling room, not sure if that will be a problem long term...

I then use an RS-NOR latch (2 states), to shut off generators when highest temp has been reached, and turn them back on when lowest temp is reached.

[Kabrute]

Weezies shouldn't delete gas in a single gas environment, afaik. Its the interaction of lesser amount of gas trying to be forced into area with greater amount of different gas and has nothing specifically to do with weezies.

[manu_x32]

5 minutes ago, Kabrute said:

Weezies shouldn't delete gas in a single gas environment, afaik. Its the interaction of lesser amount of gas trying to be forced into area with greater amount of different gas and has nothing specifically to do with weezies.

hmm, maybe you're right, or maybe it was fixed.  People used to say that wheezes Hydrogen cooling rooms got depleted by wheezes gas deletion after a while, that's why they used to always keep a slow feed in them to continuously replete them.  I may have never tested this for myself, testing it debug right now doesn't seem to reproduce the problem...

[Kabrute]

as I said, it seemed more to do with different gasses mixing, in most of those rooms would be CO2 or something in a corner and when a weeze pulled that up and tried to push it into a cell of Hydrogen the CO2 would lose, but so would some of the hydrogen.  afaik this was and still is the cause of most deletions, doors being the other cause, not to mention random pressure leakage in general, there are lots of reasons to keep a self throttled continuous feed running to a room, maybe having your morb Out of liquid will hit his personal pressure cap and limit the gas pressure in the room.

[manu_x32]

28 minutes ago, Kabrute said:

as I said, it seemed more to do with different gasses mixing, in most of those rooms would be CO2 or something in a corner and when a weeze pulled that up and tried to push it into a cell of Hydrogen the CO2 would lose, but so would some of the hydrogen.  afaik this was and still is the cause of most deletions, doors being the other cause, not to mention random pressure leakage in general, there are lots of reasons to keep a self throttled continuous feed running to a room, maybe having your morb Out of liquid will hit his personal pressure cap and limit the gas pressure in the room.

I thought 1kg was not enough to have good cooling effect, that's why I put the water.  High gas pressure doesn't seem to have much of an impact so far.   I'm also trying a version with just Hydrogen in the cooling room, but I'm afraid of gas deletion from the door under the wheeze.

[TOOK14]

Why not just make it 1 room, ditch the door and balance the number of batteries to heat even or slightly more than what the wheeze can cool?