With passive cooling you have a nice 10 million RF battery. Tie the batt % full to control rod % and you reach a nice equilibrium at your power consumption. Rednet ports have this built in, so it is super simple to setup -- love it.

Active cooling the steam in/out flow is all over the charts. I was able to get a similar throttle to equilibrium with a turtle monitoring it on a computer port every tick. My pretty big reactor has 29k steam buffer. Ideally the steam buffer would start at 100k and maybe scale up to a million using whatever method you have now for scaling.

I would avoid doing a buffer outside the reactor as almost nothing can pull out fluids at the big reactor rates.

Extra utilities drums store 256k in a single block and the corners of a reactor aren't typically used with fuel rods in the center, so maybe make a case for that is where the steam is stored. If there was an internal block or outer casing block to increase steam capacity I would add them in a heartbeat, so I could even out the consumption numbers and monitor/adjust the rods less often.

With those changes you could add steam tank % to the rednet port like the battery %, so you don't need computer craft to monitor it.

This is something that has been repeatedly discussed in other issues, and there's even a duplicate suggestion submitted less than a day ago: #250.

See also: #199, #184, #137 and #131.

Please look for similar/related issues before opening a new one.

Sorry, I read the recent/open ones and thought about adding as a reply to 250, but the other posters would have no idea what I am talking about.

They were focused on the throughput issues and their issues trying to pipe steam/water around in the quantities needed. I use tesseracts to bypass those challenges or recognize you need around 10x the ports you are used to using for liquid transport.

My goal was to just monitor steam demand and steam production and dynamically adjust the reactor to the level being consumed. The best place to monitor that is the internal buffer tank in the reactor as that allows tesseracts to be used for unlimited point to point transmission of steam. The weakness is a turtle/rednet controller cannot track the steam demand and steam production reliably in the small internal tank.

My request was to look at the internal steam buffer that last 1-2 ticks and consider making it big enough that you could monitor say once a second and adjust the control rod and not be going from 100% to 0% and back.

When the buffer is 30k and my demand was ~10k monitoring getHotFluidAmount and getHotFluidProducedLastTick was very unreliable unless I looped at computercrafts max rate. Tested a setup with a 256k drum as a buffer an I was able to push the monitoring out to .5 seconds and the reactor would bounce from 60-70%. Added a second drum and at one second monitoring intervals the reactor would run ~64-68%. To me less inputs/updates are better. But to do that test I had to add use 512 speed upgrades on extra utilities liquid nodes.

My response to any and all requests regarding the turbine's input buffer size is the same: this will not be fixed by increasing the size of the buffer. It will be fixed by providing alternative methods for handling high-flow situations.

What if the battery size was 30k RF or current output level? The mere fact that it has a 10 million RF buffer makes measuring the batt % a reliable indicator of demand over time.

As the buffer shrinks batt % = control rod % becomes less accurate. My comment was it would be nice for steam to have have similar accuracy. In the end that is your call to make for this excellent mod. I think I have communicated my thought and I thank you for reading and responding.