Combo regulated steam turbine.

[JRup]

Hi!

If an ONI question lingers in the back of your head... build it and answer it. You'll sleep happier.

Today's treat is about the self cooled steam turbine, well sorta. It's the THermo Regulated Steam Turbine Tower Expansion Rehash. (Acronym not available!)

TL;DR;:  Cools steam up to 178ºC safely for an average net gain of 450W! Little to no space materials! (Restrictions apply...)

 

 

On the low end it will generate 675W and tops at about 710W so the gross average would be around 692W or so.

This array is feeding off magma so a lower chamber is recommended to tightly manage the temperature the steam turbine is fed. Automation for the upper chamber is set to 178ºC and lower is set to 200ºC. Steam pressures are 20kg and 100kg per tile, respectively.

Here are the stats for airlock activation averages (again upper and lower chambers):

 

And here are some overlays for curiosity's sake:

Liquid piping

 

Gas piping

 

Automation

The buffer gate is set to 2 seconds. The thermo regulator activates at 99ºC

 

Metals overview

The turbine here is aluminum but gold does equally fine. All radiant liquid piping is copper, radiant gas is steel. All insulated piping is ceramic except for the places the gas piping touches hot surfaces (insulation was used but not required)

Both radiant piping should follow the same route. I once tried that fancy counter loop thingamajigs and it didn't turn out so well.

I do recommend gold for the metal tiles in the main steam chamber, this way you can easily avoid overshooting the temperature by too much.

The thermo regulator is niobium... just for the bling. (Use steel instead, meh)

 

So there you have it... A not so fuel hungry arrangement that avoids the aquatuner and gives the much maligned thermo regulator a micro spot in the sun.

[Saturnus]

This is remarkedly similar to the hyperloop idea I posted on discord except for using a thermoregulator and only going up to 171C.

The hyperloop cools up to 200C steam easily. It's completely self-powered and self-cooling, in fact it generates power. How much depends on what it cools. At 200C steam the AT is on about 16% of the time so that's roughly 192W and the ST generates 850W in that case. At 137C it runs in normal self-cooling mode and the AT is never on.

The way it works is for the AT to assist the cooling of the ST output by running that in a loop and only turn on the AT when the ST nears overheating.

  

[JRup]

50 minutes ago, Saturnus said:

This is remarkedly similar to the hyperloop idea I posted on discord except for using a thermoregulator and only going up to 171C.

Neat!

I'd love to see the discord "thread" on that one. (Although I'll admit I hardly ever use discord...)

Regarding the self cooling part of your build,  have you considered adding a hydrogen atmosphere above the water layer? (I used to cool just the same as you but found out the additional part of the cooling I was missing there and how I got to 178ºC - 179ºC with just the output from the ST. But I digress.) I'm guessing that completing the full 10kg pretty much obviates the need for that.

For clarity's sake, I'll assume the layer of thermium on the bottom is the cooling target in your build.

 

[Saturnus]

You don't need the hydrogen if you loop the cooling water around in a closed loop. Basically the same idea, the hyperloop just adds the AT to take this self-cooling potential to 200C steam, and beyond if needed.

Yes, that is thermium. For testing purposes not much beats just pasting in a few million tons of thermium at the desired temperature.

[tuxii]

@Saturnus

What do you set the sensor to, 99-100C?

[Saturnus]

7 hours ago, tuxii said:

@Saturnus

What do you set the sensor to, 99-100C?

So after some additional testing I found the original set up was sometimes too slow to react to temperature spikes, and you needed different thermosensor settings depending on temperature range. So I redid the set up, and tested it at different temperatures and different materials used.

Above it's cooling Thermium at 200C, 165C, and 130C. And testing if there's a difference between using petroleum and water. There isn't btw.

What I've found is that if you just don't use lead radiant pipes or a lead ST then the metals used are pretty much the same and they all work with a thermosensor setting at 98C.

The AT uptime for each set up is 15%, 8%, and 1% respectively. Producing an on average surplus of 664W, 464W, and 264W respectively.

Here's the piping overlay in both display formats. Everything else should be pretty straightforward, ie. connect power ports together, connect thermosensor to AT etc

Note 33kg/tile water/petroleum is used on top (1 full bottle) and 100kg/tile water was initially dropped inside (3 full bottles).

[JRup]

56 minutes ago, Saturnus said:

So after some additional testing I found the original set up was sometimes too slow to react to temperature spikes, and you needed different thermosensor settings depending on temperature range

Thermo Sensors are slow to pick up changes. I prefer to stick a liquid thermo sensor on the final pipe after the steam turbine (or as close to possible to that...) This gives an accurate reading of the turbine's temperature as it must be the same as the water used to cool it.

I decided to mix up some stuff last night and moved on from a Thermo regulator to a gas shutoff... "How?" you'd say... well, idle hands and some "hydration" later I got to this, all liquid sensors are set to 99.4ºC:

It tiles... This uses 200kg of super coolant for the AT and the big loop is water (Set to 72ºC leaves the top pool at 77ºC all radiant copper and only aluminum for the cooling pool).

Separate hydrogen loops are provided for each ST.

[Saturnus]

23 minutes ago, JRup said:

Thermo Sensors are slow to pick up changes. I prefer to stick a liquid thermo sensor on the final pipe after the steam turbine (or as close to possible to that...) This gives an accurate reading of the turbine's temperature as it must be the same as the water used to cool it.

That was my initial thought as well but it's complicated by the fact that you need to place the liquid thermo sensor (relatively) far away from the actual ST and that the liquid temperature doesn't accurately reflect the ST temperature. The thermosensor set up as depicted in my post above gives the fastest and most accurate control possible.

The thermosensor setting can be micromanaged but it really doesn't matter too much as the difference in AT up time is in the 1/10ths of a % difference, 1% difference maximum. The 98C thermosensor setting works for all materials used as radiant pipes and/or the ST (except lead) and works for all temperatures from 125C to 200C. No need trying to optimise beyond that.

[gabberworld]

i want point out also that Depleted uranium have slow heating. in my eyes its the best material for build the Steam turbines

[Saturnus]

19 minutes ago, gabberworld said:

i want point out also that Depleted uranium have slow heating. in my eyes its the best material for build the Steam turbines

It's worth pointing out that I have not tested DLC specific materials, as I don't own the DLC and have no intention of buying it. But looking at the stats it would be worse than even lead to use for both radiant pipes and the ST itself due to it's absolutely terribly TC. For an accurate temperature control you'll want at least decent TC (to transfer it's heat to the water).

[gabberworld]

17 minutes ago, Saturnus said:

It's worth pointing out that I have not tested DLC specific materials, as I don't own the DLC and have no intention of buying it. But looking at the stats it would be worse than even lead to use for both radiant pipes and the ST itself due to it's absolutely terribly TC. For an accurate temperature control you'll want at least decent TC.

i not talk about pipes really, it can be that Aluminium is best not the Depleted uranium, but i not know why yet. either its a bug or devs trolling us with that slow heating text

need todo some more heavy tests

 

[gabberworld]

this is list off material from worst to best

its the cold test without any cooling builded, but i nothise one thing both the Thermium and Aluminium takes cooling from bottom tiles from steam turbine and in end they may go very hot

i tested them with igneous rock tiles as that is most common tiles for this stuff

Lead Gold Tungsten Niobium Copper Iron Cobalt Steel Depleted Uranium Thermium Aluminium

[Saturnus]

13 minutes ago, gabberworld said:

its the cold test without any cooling builded, but i nothise one thing both the Thermium and Aluminium takes cooling from bottom tiles from steam turbine and in end they may go very hot

You can't really use that test in this context though, and regardless when I see a list where "steel" is ranked higher than "iron" I know it's not been a very rigorous test.

[gabberworld]

22 minutes ago, Saturnus said:

You can't really use that test in this context though, and regardless when I see a list where "steel" is ranked higher than "iron" I know it's not been a very rigorous test.

it shows what needs the most cooling energy and what less. with one materials you need run cooling loop more often than others

soo i used now in this test with the Insulated Tile as in this test aluminum and Thermium cant take coolng from bottom tiles

the Depleted uranium wins with huge advantage

while all others are already hit 100c mark. that one runs at 94c

Lead

Gold

Tungsten

Niobium

Copper

Cobalt

Iron

Steel

Thermium

Aluminium

Depleted Uranium

 

 

[JRup]

3 hours ago, Saturnus said:

That was my initial thought as well but it's complicated by the fact that you need to place the liquid thermo sensor (relatively) far away from the actual ST and that the liquid temperature doesn't accurately reflect the ST temperature. The thermosensor set up as depicted in my post above gives the fastest and most accurate control possible.

The thermosensor setting can be micromanaged but it really doesn't matter too much as the difference in AT up time is in the 1/10ths of a % difference, 1% difference maximum. The 98C thermosensor setting works for all materials used as radiant pipes and/or the ST (except lead) and works for all temperatures from 125C to 200C. No need trying to optimise beyond that.

Did some messing around with the build itself. The thermo sensor triggers after half a degree is surpassed. So when setting it to above 98ºC we're actually getting a positive after 98.5ºC hence the impression the sensor is slow...

I did some fiddling around with the liquid sensor and 98.7ºC was a safe bet.

And even though I enjoy some Slayer every once and then, there was too much headbanging with the hyperloop... So I tried out a reservoir to see how it went. Results are very consistent still.

I must admit this is a fun design to mess around with.

[gabberworld]

when you input heat is higher than 200 its recommended also use the steam turbine valve closure automation, usually needed when you  take heat from high temps like magma

[JRup]

43 minutes ago, gabberworld said:

when you input heat is higher than 200 its recommended also use the steam turbine valve closure automation, usually needed when you  take heat from high temps like magma

I usually prefer to cut heat input instead of stopping the turbine. If I have to stop the turbine it would be because I'm using a smart battery for that purpose, for example.

[gabberworld]

19 minutes ago, JRup said:

I usually prefer to cut heat input instead of stopping the turbine. If I have to stop the turbine it would be because I'm using a smart battery for that purpose, for example.

i not talk about stop turbine

[JRup]

2 minutes ago, gabberworld said:

i not talk about stop turbine

A screenshot of what you have in mind can help.

[gabberworld]

4 minutes ago, JRup said:

A screenshot of what you have in mind can help.

funny that you ask but here you have

[JRup]

Just now, gabberworld said:

funny that you ask but here you have

Oh, yes! I totally forgot about that. I haven't found a use for it yet because I always plan and build to avoid this.

The map I'm currently using does not even have a hot steam vent to even fiddle around with something like this so I'm kind of stuck with my vanilla builds atm.

[gabberworld]

3 minutes ago, JRup said:

Oh, yes! I totally forgot about that. I haven't found a use for it yet because I always plan and build to avoid this.

The map I'm currently using does not even have a hot steam vent to even fiddle around with something like this so I'm kind of stuck with my vanilla builds atm.

things is about this setup is that you can input higher heat, and because you close valves it generates the less water output what means you heat source cooling down slower, what means it can run much longer

[blakemw]

 

On 11/11/2021 at 11:35 PM, Saturnus said:

This is remarkedly similar to the hyperloop idea I posted on discord except for using a thermoregulator and only going up to 171C.
 

  

I've also been using this kind of design (independently invented) for a long time. I call them "hybrid steam turbines".

What I normally do is have a Liquid Shutoff that is linked to a Thermo Sensor that is set to something like > 99.7 C (fractions are allowed to be entered, they are just rounded on display). This way the auxiliary cooling only kicks in if the self-cooling can't handle the heat load.

A fairly typical such design of mine, here for an Aluminium Volcano Tamer:

Aluminium Volcanos, at least when using an instant-freeze-to-debris design, is incredibly bursty in its heat output, since aluminium has such a low freezing point. A practical self-cooled design needs 3 Steam Turbines. On the other hand, the average heat output is only about 700-800 kDTU/s which can pretty much be handled by a pair of self-cooled Steam Turbines, especially if you store some of the aluminium for the dormant period.

Hence this is an ideal build to have auxiliary cooling to bring the temperature down.

Some build notes:
 

1. To avoid broken pipes it's necessary to let water trickle through even if cooling is not required. Here I use the simple solution of ORing the Themo Sensor with a Timer Sensor set to 1/600 to release a water packet every 600 seconds. It could certainly be done with more logic like using a Liquid Pipe sensor to actually detect if a water packet has been recently released.
2. A metal or diamond Tempshift Plate overlapping the Steam Turbine and Thermo Sensor does an exceptional job of keeping the temperatures in sync, there is no meaningful lag and the temperature deviation isn't more than 0.1 C or so. I use a Thermo Sensor threshold like 99.7 C with no problems with the Turbine overheating. You can also use an aluminium Conductive Wire Bridge to thermally "connect" the Thermo Sensor with the Steam Turbine.
3. This kind of build is straight up complexity for the sake of complexity, like in this case simply using three self-cooled Steam Turbines would be pretty much better in every way unless there's some dire need to save space. But I do find it extremely satisfying and often I include even more pointless logic and complexity to make the build a little more "efficient". 

[JRup]

2 hours ago, blakemw said:

To avoid broken pipes it's necessary to let water trickle through even if cooling is not required. Here I use the simple solution of ORing the Themo Sensor with a Timer Sensor set to 1/600 to release a water packet every 600 seconds. It could certainly be done with more logic like using a Liquid Pipe sensor to actually detect if a water packet has been recently released.

I've been looking into the general outline of the 'hybrid' design and have come to a simple workaround for a single turbine (I'd have to tear stuff down and rebuild to be able to test for more than that.)

No need to trickle, no need for timers. Don't toil, pre-boil! I've let the game and pipe rules do the tricky stuff.

Two reservoirs and 3 liquid valves  (top: 500g/s; mid & bottom: 1000g/s) handle the logistics easily for a single turbine, the liquid sensor governs the AT and is actually set to below the threshold of the thermo sensor:

 

My take is that as long as we're not cooling the turbines below 95ºC then it's all game.

 

2 hours ago, blakemw said:

This kind of build is straight up complexity for the sake of complexity, like in this case simply using three self-cooled Steam Turbines would be pretty much better in every way unless there's some dire need to save space. But I do find it extremely satisfying and often I include even more pointless logic and complexity to make the build a little more "efficient". 

Complexity you say? ... hmm presenting the testing phase of a gold tamer... The counter is to find out if we'd be exporting power from the build at all... Chilly -10ºC gold is coming out of this build...

I'm dividing the steam chambers to keep "low pressure" steam available for the turbine. Atmo sensors only go up to 20kg after all.

The lower chamber has about 125kg/tile at the moment... I'm considering lowering the pressure to allow for higher temps, maybe...

The volcano is in a vacuum and heat is being extracted by a super-coolant filled pipe. The blob of gold occurred during the startup phase but turned out to be a happy little accident. I have another build in the works with a coal tempshift plate in the lurk.