Prioritization Rules for Auto Sweepers?

[Qubit]

Long build up, but let me sketch the backstory so as to be clear: 

Recently I've been experimenting with various (magma) volcano taming steam turbine setups.  I've got the basics down-- my magma is getting dispersed and cooled at a rate where it turns into simple igneous rock debris rather than clumping into actual rock blocks that could choke the volcano, so, technically I can ignore it and just enjoy the occasional bursts of electricity.   

I was bothered, however, by how slowly my setups have been able to drain the (still substantial) heat left over in the igneous rock debris.  Some of the magma from each eruption makes it beyond the volcano's neutronium base to sit on the metal tile floor of the heat chamber.  That debris gets drained down to 250-350 degrees fairly quickly.  But much of the magma cools to debris right on the volcano's neutronium pad,, and even though the room has a pure hydrogen atmosphere, those neutronium stranded rocks can still be 800 degrees dozens of cycles after an eruption, only managing to lethargically spin the turbines. 

So upon reading on this forum that using conveyor belts to route solids through tiles provides for crazy good heat transfer, I decided I'd try conveying my leftover hot igneous debris through the metal tiles in the heat chamber to Goose the system a bit.  And it worked...extremely well, to a point.

After the volcano goes dormant, I send my dupes into the heat chamber to perform a one time sweep of my still very hot debris to a set of sweep only storage compactors in a heavily insulated chamber adjacent to the main heat chamber.  Once staged, these rocks get swept by an autosweeper onto a conveyor line that winds through the metal tile inner "liner" that runs around the perimeter of the heat chamber (and which contacts (via doors) the steam turbines).  So far so good -- the first load of rocks picked up by the sweeper quickly transfer their heat into the heat chamber liner and I get a fresh power spurt out of the turbines. 

After the rocks finish their loop around the heat chamber, I'm having them deposited via a chute back onto the floor of the staging chamber, where they are in range to be picked up again by the auto sweeper.   

That's where my problem begins.   Although the sweeper has access to a lot more rock than I can fit on my conveyor rail (some sitting in storage compactors, some now accumulating as a fresh debris pile next to those compactors), the sweeper will continuously keep picking up the same set of rocks -- namely the increasingly cold ones coming off the rail and landing on the floor, while ignoring the still relatively energetic rocks waiting for attention in the storage compactors.

So, long build up....what is the recommended approach for de-energizing, say, all 30 tons of igneous rock debris that a given volcanic activity cycle might produce when my conveyor rail can only process about 5 tons at a time, and seems to want to continuously *re*-process the same five tons?   I mean, I've resorted to monitoring the situation and having my dupes occasionally intervene to carry the "chilled" section of the debris out of range of the sweeper so that it is forced to move on to putting hotter rocks on the rails, but...that is way too much tedious micro to be worthwhile.  

Does an auto sweeper follow a known set of prioritization rules when picking which, amongst multiple eligible sweep targets, to sweep?  If so, is it positional (my debris is landing to the right of my storage compactors)?  Or is floor debris prioritized over compactors?   Or...am I thinking about this the wrong way?

Thought/recommendations appreciated.

Thinking about it a little bit more, if it were *only* 25t of debris, I could probably just make sure that my rocks all started off as loose debris sitting in a single tile (and have the conveyor chute output onto the same tile.)    That way, I think the entire stack of rock would gradually get processed.   But complicating matters is that the full eruption cycle is, I believe, actually producing a bit more than 25 tons, so I don't think I'm ever going to be able to rely on getting all the rock i want processed collected in a single 25t stack.  

[nakomaru]

I'm not sure why you are storing the cooled rock with the hot rock. I think it's because it takes more than one pass to cool?

I don't think you can get sweepers to prioritize what they are picking up, so instead restrict their access.

How about you drop off the rail system rocks in one pile that the sweeper can't see via a door. When you want the sweeper to again access that rock, open the door. Or drop it on a door and have the door open to drop it into range again. Maybe you could open it every 900 seconds to process 18t batches. Shutoffs can be used to split into different piles based on timers too.

This may not be an option, but generally I would be increasing the thermal capacitance and number of turbines so the steam room can maintain 170~195C, and one pass of the rails should bring it to that stable temperature. Alternatively, only allow rails when below your desired temperature. Then I would finish off the cooling by passing it through an aquatuner chilled thermal battery. Volcanoes can erupt into more than 100kg of steam per tile, so I prefer to keep them submerged in my turbine's high pressure steam room.

[Darkin Coaled]

6 hours ago, Qubit said:

..I send my dupes into the heat chamber to perform a one time sweep..

That bit kept me reading your wall of text..

Anyhow, as long as the output is less than 25t plus whatever fits on the rails you should be golden. Stick a conveyor chute above the compactor. Don't know what the complexity of having debris while your guys are still sweeping would be, but you might just want to run the sweeper as they are working. Once they finish, simply empty the compactor to the floor. This way you have a pile of temperature averaging debris that the sweeper can then pick from.

[Qubit]

Thank you, both.  

Nakomaru - yes, you're right, with the current setup I need to circulate the debris through the heat chamber multiple passes to drain the rocks down to a temp I'm comfortable with.   Using timer driven doors to separate the cooler debris from warmer and periodically allowing them to remix may be the ticket.  And I hadn't considered jiggering with the steam chamber to try and improve thermal capacitance as a way of reducing the need for extra loops.  

Darkin - yea, I'm not sure if everything will fit in one 25T pile + what's "in circulation" but since that would be a very simple solution, I need to rule that out.  

Feedback appreciated!