I've developed a method for displaying circuit network data on the power graph. The design uses sets of 1, 2, 4, 8 etc. lamps to draw an amount of power governed by the input signal. There is 1 lamp connected to the 'AND 1' circuit, 2 to the 'AND 2', 4 to the 'AND 4', etc. The power graph for the isolated power circuit plots the circuit value. To isolate the power grid I'm using a steam engine, so the thing requires steam input to run.
The first approach I tried was switching on and off various devices with power switches. The initial inrush of current to charge the device's internal battery causes a big spike in the graph. Logic gates have 1kW draw and a small internal battery, so make reasonable drains - the spike is small but still visible. Lamps have 5kW draw, but seem to have effectively zero internal battery, so the output on the graph is very clean. Switching power for lamps sometimes leads to power network partitioning issues where the graph will lose its history. Letting lamps turn themselves on and off using their own circuit network logic solves this problem.
There are 127 lamps, this is the effective resolution of the output trace on the power graph. You could increase it to 255 for more granularity.
In this example I'm graphing the fluid level of tanks as I pump water from one to the other.
Prior art in this thread 29735 using PWM on a power grid and here viewtopic.php?f=193&t=29998&p=189895 using production statistics.
plotting circuit network on power graph
Re: plotting circuit network on power graph
Very creative! +1
Re: plotting circuit network on power graph
Do fluid pumps have a buffer? And do they have to move fluids to draw power? Wiki just says 30kW draw. 6 times as much at double the area used, so might have potential to reduce footprint of the blueprint up to 3 times if it works.
Beacons and radars don't buffer energy (beyond a single tick) I think, and use quite a bit of power per tile all the time. You could put those behind power switches to get varying loads. And to avoid having networks (that you want to measure on) have power switches that merge networks you can separate the power switches from your measure network with accumulators that are powered on the measure side and have load and switches on another network on a consume side. If load is never higher than accumulator throughput they are always full and just draw as much as whatever is on the other side. If two power poles both cover the same accumulator then energy can flow through the accumulator from one network to the other even when they are not connected. You would need 1 accumulator for every radar though I believe so at least 9+4 tiles for 300kW.
Beacons and radars don't buffer energy (beyond a single tick) I think, and use quite a bit of power per tile all the time. You could put those behind power switches to get varying loads. And to avoid having networks (that you want to measure on) have power switches that merge networks you can separate the power switches from your measure network with accumulators that are powered on the measure side and have load and switches on another network on a consume side. If load is never higher than accumulator throughput they are always full and just draw as much as whatever is on the other side. If two power poles both cover the same accumulator then energy can flow through the accumulator from one network to the other even when they are not connected. You would need 1 accumulator for every radar though I believe so at least 9+4 tiles for 300kW.
My mods: Capsule Ammo | HandyHands - Automatic handcrafting | ChunkyChunks - Configurable Gridlines
Some other creations: Combinassembly Language GitHub w instructions and link to run it in your browser | 0~drain Laser
Some other creations: Combinassembly Language GitHub w instructions and link to run it in your browser | 0~drain Laser