Jump to content

I want to make metal pipes


Recommended Posts

12 minutes ago, Reaniel said:

Saturnus, you can actually prime the temps in vents and pipes by painting the area with 10000kg of whatever gas at the temperature you want.  It'll prime your system instantly.

I can also let it prime on it's own, and make a cup of coffee while it does. :D

Link to comment
Share on other sites

I ran my test again. This time with 2 tiles of 10kg P-O2 at 280K and 380K respectively to see why there's such a huge difference from the previous test I ran with liquids which clearly showed wolframite to be far inferior. And sure enough. If the static medium is a gas instead of a liquid then wolframite performs better.

Besides from wolframite always performing best when the static medium is a gas (remember, the previous test where the static medium was a liquid, wolframite always performed the worst) then granite performed the best at high flow rates, and sandstone performed the best at low flow rates.

2017-09-17 (2).png

Link to comment
Share on other sites

4 minutes ago, Saturnus said:

I ran my test again. This time with 2 tiles of 10kg P-O2 at 280K and 380K respectively to see why there's such a huge difference from the previous test I ran with liquids which clearly showed wolframite to be far inferior. And sure enough. If the static medium is a gas instead of a liquid then wolframite performs better.

It seems to be more of an issue with the mass of the static medium (per tile) as oppose to the state of the medium.

Link to comment
Share on other sites

6 minutes ago, Reaniel said:

It seems to be more of an issue with the mass of the static medium (per tile) as oppose to the state of the medium.

I disagree with that being an accurate interpretations of the results. You can run your own tests, and show the results if you like.

You're also seemingly ignoring the most important finding. The lowest flow rate was the best in all cases. It cooled down/heated up the static medium much more than a higher flow rate which is completely unintuitive.

Link to comment
Share on other sites

Additional tests with heating and cooling water:

I've added a small chamber of 10kg water to see how heating/cooling in a single tile perform.

We first start with heating:

WaterS1Setup.thumb.png.70bd5166a6e65e2bf6e794fa8974361c.png

Result:

WaterS1Result.thumb.png.56cf44b5a8af3c9cfae1325de1c9ba51.png

So just like the case with gas, Wolframite seems be superior in heating/cooling a medium when using a water/water system.

However, it took out less overall heat from its pipe content.

Test 2 fills the chamber with water at 1000kg per tile.

WaterS2Setup.thumb.png.75d27627d9c5cce7b9933fd515273c38.png

The result was... surprising...

WaterS2Result.thumb.png.09638a2938776cda75a5de22dc3d4c6c.png

The temperature of the end product are the same for all 3 systems!  I'm really not sure what to make of this.  Perhaps Risu or Kasuha can try to shed a light on this?

Same tests but for cooling instead.  It pretty much confirms the result of heating system.

WaterS3Setup.thumb.png.f88ca05d3677263e6f77172438435dfb.png

WaterS3Result.thumb.png.0ccf7f58eb7f719058302de0f8c8ddb7.png

WaterS4Setup.thumb.png.ec3662e2065acec261a32c15110a3a5c.png

WaterS4Result.thumb.png.771ec86d339e273adb79364423fbfe26.png

 

I guess I'm almost 95% a (re)convert to Wolframite pipes.

And I can't wait until we get Tungsten.

 

Link to comment
Share on other sites

8 minutes ago, The Plum Gate said:

Curious: Was it the same mass gas and liquid?

No. But I'm running it at 800kg/tile P-O2 now. It's exactly the same result as for 10kg/tile... it just takes longer.

It is the state of the static matter that is important. Not the amount.

Link to comment
Share on other sites

10 minutes ago, Saturnus said:

I disagree with that being an accurate interpretations of the results. You can run your own tests, and show the results if you like.

You're also seemingly ignoring the most important finding. The lowest flow rate was the best in all cases. It cooled down/heated up the static medium much more than a higher flow rate which is completely unintuitive.

I did do the test, and having more mass per tile in gas chambers make Wolframite's total heat transfer higher.

It's especially apparent when you use liquid, in that having massive "static medium" would negate any total heat transfer granite/sandstone has over wolframite.

Link to comment
Share on other sites

Just now, Reaniel said:

I did do the test, and having more mass per tile in gas chambers make Wolframite's total heat transfer higher.

It's especially apparent when you use liquid, in that having massive "static medium" would negate any total heat transfer granite/sandstone has over wolframite.

And I found and showed the result showing the exact opposite.

It is not the amount of the static matter. It is the state of the matter that matters (pun intended).

I increased pressure to be the same 800kg/tile for the gas and liquid. And again, wolframite performed worst of all with liquids, and best of all with gases.

Link to comment
Share on other sites

2 minutes ago, The Plum Gate said:

...This is very strange behaviour for a physics sim - if what both of you are demonstrating is accurate. or infallible, etc.

So the takeaway from this is to use granite when wanting to radiate from pipes in liquids and wolframite for gases.

And most importantly... use the lowest possible flow rate you can get away with in your set up. The slower the flow rate, ie. the less water than actually flowed through the system, the better it heated up or cooled down the static medium. This, to me, quite unintuitive finding was the most important factor of all for the overall cooling or heating. Far more important than the material choice.

Link to comment
Share on other sites

Just now, Saturnus said:

The slower the flow rate, ie. the less water than actually flowed through the system, the better it heated up or cooled down the static medium.

I found this particular part, the "dwell time", to be more relevant when I was designing a water cooling system from a geyser.

Link to comment
Share on other sites

1 hour ago, The Plum Gate said:

I found this particular part, the "dwell time", to be more relevant when I was designing a water cooling system from a geyser.

Yes. Notice how the output water is 7.0C and 7.7C (low flow on the right). That's heated up 0.1C and 0.8C. Seems fair enough. 10 times less water flow through the system so it's heated up more. But notice the temperature change in the static medium. The high flow system had cooled that by 85.1C whereas the low flow system with 10 times less water flowing through the system had cooled it by 86.7C. It's quite astonishing that 10 times less water can cool the same amount of liquid more just because it's flowing 10 times slower.

EDIT: The run with 800kg/tile P-O2 is trending to show the same numbers as the 10kg/tile P-O2 test. No reason to complete the 800 cycle run as I can already conclude it will have the same result. The static medium amount is irrelevant, it's just takes longer.

wierdness.png

Link to comment
Share on other sites

21 minutes ago, Saturnus said:

And I found and showed the result showing the exact opposite.

It is not the amount of the static matter. It is the state of the matter that matters (pun intended).

I increased pressure to be the same 800kg/tile for the gas and liquid. And again, wolframite performed worst of all with liquids, and best of all with gases.

That's because you're comparing apples and oranges.

Different gases and liquids has different specific heat and thermal conductivity.  I mean, for crying out loud, same element in different state has different thermal conductivity.  And the deltaH of each system, calculated through temperature change of the content, showed that the mass of the static medium plays a significant factor in how efficient it runs in terms of total heat transfer (deltaH).

 

Link to comment
Share on other sites

44 minutes ago, Reaniel said:

That's because you're comparing apples and oranges.

I thought that was blindingly obvious. But I am comparing apples to apples, and oranges to oranges. I have no idea what fruit your tests are. 

As I have shown, it has completely different results, and completely different conclusions. To reiterate, with liquids (more specifically polluted water) wolframite performs the worst in my tests. With gases (more specifically polluted oxygen) wolframite performs the best in my tests.

But much more importantly to both systems, reduced flow rate by a factor 10 increases efficiency by a factor of 22(ish).

Link to comment
Share on other sites

On 2017/9/17 at 2:59 PM, Saturnus said:

The amount of water didn't allow for a measurable difference to appear. You need to redo it with far more starting water and let it run longer.

Actually, running it longer doesn't make your result better.  It's much more accurate to infer deltaH from the resulting water temperature instead of trying to look at the .1 or .2 degree difference at high mass.

I built a copy of your testing contraption, and one thing I could never figure out is why you are testing 2 tiles of space instead of just 1.  You'd be able to correctly compare gas and liquids if you do just 1 tile, or do a horizontal tiles that'd allow you to do low liquid tiles across multiple tiles.

Also, having the liquid vent in the resulting water would throw you off on the final temperature of your water.

Link to comment
Share on other sites

Here are additional tests I've done with a rebuilt testing contraption.

This is part one of the results, which just compares different mass of external medium and how it'd actually affect heat transfer of the pipes.

Setup:

Spoiler

I've rebuilt my original design to avoid having the vent affecting the final water temperature.  Also, building horizontal rows like this allows me to have low mass liquid tiles interacting with the pipes and compare directly with gases.

I also added a liquid valve to test how flow rate would affect heat transfer to the pipe system.

PipeTestSetupPipes.thumb.png.33bc5bcd7f7ca6b08d80ef1cfbe4dd88.png

PipeTestSetupThermal.thumb.png.47e782d71165140c0f950aae7902806c.png

 

Test 1: Pipes exposed in vaccum to test how well the pipes transfer heat with no external medium.

Spoiler

DeltaH is calculated by simple H = MCT

Pipes, like buildings, have only 1/5 of mass in calculating heat transfer.

So for each row with 20 pipe tiles, I'm simply using DeltaH = (1000 x 100 / 5) x (0.134) x (Average of Max/Min - 5) x 20

For the DeltaH of Water, it'd be (1000 x 1000 x 4) x (4.179) x (Average Temp of 4 water tiles)

PipeTestR1.thumb.png.fb311ab5537ec6a6aa60c0fa3c6dafed.png

 

Test 2: 1kg per tile of water as external medium:

Spoiler

 

PipeTestR2.thumb.png.03909cffd5987f849524280a7e2143e9.png

 

Test 3: 10kg per tile of water as external medium:

Spoiler

PipeTestR3.thumb.png.b398dd477bb82ae0492fcc388a1b97e5.png

 

Test 4: 100kg per tile of water as external medium:

Spoiler

PipeTestR4.thumb.png.77f7d10f567c3808dcc83bfa507f7d02.png

Test 5: 1000kg per tile of water as external medium:

Spoiler

DeltaH calculation in the pipe sections becomes pointless with high mass for external medium because of the small temperature change.

The most accurate (relatively speaking) way of calculating DeltaH remains using the final temperature of content.

PipeTestR5.thumb.png.88794873c32041b5a51e607f694d0550.png

 

From the results of test1~5:

As you can clearly see, the amount of external medium you have outside of the pipe greatly affects the heat transfer effectiveness for the pipe system.  When you have very high mass per tile of external medium (such as cooling or heating liquids), all 3 are nearly identical in terms of amount of heat removed from the pipe content.  I have no idea why this is happening except that this is what's happening.

In addition, while Sandstone and Granite both takes more heat from the pipe content in almost all cases, Wolframite pipes are clearly superior in transferring heat to the external medium.  

(I'll post more test results later with different elements as well as liquid oxygen lines)

Link to comment
Share on other sites

Part 2 tests different materials for the external medium.

Test 6: 10kg per tile of oxygen as external medium:

Spoiler

PipeTestR6.thumb.png.285bc9166d15717d0149c9e47598468d.png

Test 7: 100000kg per tile of oxygen as external medium:

Spoiler

PipeTestR7.thumb.png.2097a47c86522432dd8f8fc1905f1872.png

Test 8: 10kg per tile of hydrogen as external medium:

Spoiler

PipeTestR8.thumb.png.9ce0fce5fbb3f6ee639798787482d515.png

Test 9: 10kg per tile of helium as external medium:

Spoiler

PipeTestR9.thumb.png.97b315c7a77b6438fd4330f36134dafb.png

Test 10: 10kg per tile of propane as external medium:

Spoiler

PipeTestR10.thumb.png.a29137866f76d643e390befe0b6f8982.png

Test 11: 10kg per tile of crude oil as external medium:

Spoiler

PipeTestR11.thumb.png.f51247f1241b5ec71008e1437d86009d.png

Test 12: 10kg per tile of mercury as external medium:

Spoiler

PipeTestR12.thumb.png.4c8cff194fed8176ff2c1351acced09f.png

From tests 6 and 7:

This confirms what part 1 had tested, where if you have high mass per tile of external medium, it will make the heat removed from internal medium roughly the same in all 3 types of pipes.  State of the external medium isn't a factor.

From tests 3, 6, and 8~12:

Different external medium will affect how effective the pipe system is.  The rule of thumb seem to be that the higher the specific heat and/or thermal conductivity the external medium has, the more effective it will make the piping system.

Again, state of the external medium isn't a factor at all, as the comparisons of Helium (gas) and Mercury (liquid) shows.  These two have identical specific heat, and their results closely mirrors one another.

It also show that while thermal conductivity of the external medium play a role, specific heat weighs more in determining the effectiveness of the system.  Helium and Mercury have only minor differences in their results despite having a huge thermal conductivity difference.  You may also see that when comparing Helium (high thermal conductivity, low specific heat) and Propane (low thermal conductivity, high specific heat), where Propane wins outright in the DeltaH numbers.  Of course, when you have about the same specific heat (like Hydrogen and Propane or Helium and Mercury), thermal conductivity will be the tie breaker.

 

 

Link to comment
Share on other sites

21 hours ago, Reaniel said:

In addition, while Sandstone and Granite both takes more heat from the pipe content in almost all cases, Wolframite pipes are clearly superior in transferring heat to the external medium. 

Judging from your tests, it's because you run it with small batches and not until stabilization. In your tests, pipe temperature capacity isn't much smaller than medium temperature capacity, which it should be. This is why wolframite, which has lowest capacity, transfers the best - it has the lowest "flat" cost, which makes it win despite (or regardless of) everything else.

Instead of testing pipe material in a continuous system, you only really get the results for an one-off system made to last a few cycles. This isn't really useful for trying to figure out the role of heat capacity in heat transfer, it can only be used to answer a question like "I need to cool 500kg of polluted oxygen, what pipe material should I use?".

On 9/17/2017 at 8:46 PM, Saturnus said:

And most importantly... use the lowest possible flow rate you can get away with in your set up.

Currently all materials in a given form occupy the same volume. For example, a packet of fluid in a pipe occupies the whole pipe tile for purposes of heat transfer.

This is why large blocks of ice (from compactor) can survive for hundreds of cycles in magma and why dupes walking through "tall puddles" (water on pwater) lose as much heat as fully submerged ones.

Link to comment
Share on other sites

12 hours ago, Coolthulhu said:

Judging from your tests, it's because you run it with small batches and not until stabilization. In your tests, pipe temperature capacity isn't much smaller than medium temperature capacity, which it should be. This is why wolframite, which has lowest capacity, transfers the best - it has the lowest "flat" cost, which makes it win despite (or regardless of) everything else.

 

I didn't let the system "run until stabilization" because it's pointless and waste of time.  There's nothing you can't infer from short tests that would require running a 50 or 100 cycles.  The "stablization" part can be tested/simulated with a short test by simply decrease the temperature difference of the external and internal medium.  (Which I've actually tested but haven't posted it yet).  This renders the pipe's heat capacity concern irrelevent because I've already done tests that'd show the effect (how increasing the mass of external medium will affect the overall performance).

As for the whole "which is best" debate, it depends on what your purpose is when using a pipe system for heating/cooling.  

If your primary objectively is to cool/heat the external medium, then wolframite is the best material because it transfers the most heat to the external medium when everything else is constant.  It doesn't matter if you're running it until stabilization or not because wolframite system will reach that point faster than granite/sandstone by transfering a higher portion of heat onto the external medium at all time, even as it gets closer to the "stabilization point.

It's a different matter when you're trying to cool or heat the internal medium.  In that case, Sandstone/Granite is the better choice because it transfers more heat to/from the internal medium than wolframite in "most" cases.  

Finally, it depends on if your external medium is a gas or liquid, because this will affect the possible mass of your external medium.  For gasses, you'll probably get something around 1~1.2kg per tile most of the time,  while liquids will give you 500~1000kg per tile.  Again, if what you're trying to achieve is changing the external medium's temperature, wolframite is the best of the 3.  If you're trying to change the internal medium's temperature, Sandstone/Granite is a lot better in gasses, but not much so in liquids.

Link to comment
Share on other sites

14 hours ago, Coolthulhu said:

Currently all materials in a given form occupy the same volume. For example, a packet of fluid in a pipe occupies the whole pipe tile for purposes of heat transfer.

What you're implying would mean that the amount of flow have no relevance and have the same heat transfer. This is an erroneous assumption.

If you scroll up to my tests above you'll find this is one thing I specifically tested for, and over 50 cycles to eliminate errors. It turns out the lower the flow rate in the pipes the faster the external or internal medium is heated up or cooled down. While this is logical that the internal medium would cool down or heat up faster the lower the flow rate then what you're saying would have it not so but in fact it is. What is perplexing however is that the lower the flow rate, ie. the less water flows through the pipes the better it is a heating or cooling the external medium, regardless if that is a gas or a liquid. And not just by a little bit, by a huge margin. That is completely unintuitive and completely contrary to what you're suggesting above.

Link to comment
Share on other sites

1 hour ago, Reaniel said:

If your primary objectively is to cool/heat the external medium, then wolframite is the best material because it transfers the most heat to the external medium when everything else is constant.

Careful analysis of your tests doesn't suggest anything of the sort.

From your last batch of tests, only tests 5 and 7 meaningfully tested heat transfer in continuous system. And both of those showed minimal to no difference.

Other tests only showed that wolframite pipes take less heat from the content and environment to stabilize. Which is nothing new. Deriving any other conclusions from them is erroneous as they were dominated by the effects from the initial state. The effects from initial state explain the perceived asymmetry between heating/cooling content and the environment - the sum of heat in the system has to stay constant (sans bugs), so any difference in "heat lost/gained by external medium" and "heat gained/lost by internal medium" necessarily has to be the heat transferred to pipes (or bugs, but Occam's Razor). And pipes can only absorb a fixed amount of heat.

Your discrepancies look like those from lack of stabilization, your test environment looks like one that doesn't stabilize, you handwave the stabilization away. If it looks like a duck, walks like a duck, quacks like a duck etc.

54 minutes ago, Saturnus said:

It turns out the lower the flow rate in the pipes the faster the external or internal medium is heated up or cooled down.

I remembered one bug that might be important here: apparently some liquid flows (in the world, not pipes) that cause mixing can cause heat to go up or down to match one of the liquid tiles mixed. In your tests you pump the liquids from below. Have you tried dripping them from above?

If not that, then I have no other explanation. Could be rounding, but that doesn't sound like it would explain the magnitude.

The part about equal volume was from sources - there is no multiplier for volume (called contact surface there).

Link to comment
Share on other sites

5 minutes ago, Coolthulhu said:

I remembered one bug that might be important here: apparently some liquid flows (in the world, not pipes) that cause mixing can cause heat to go up or down to match one of the liquid tiles mixed. In your tests you pump the liquids from below. Have you tried dripping them from above?

If not that, then I have no other explanation. Could be rounding, but that doesn't sound like it would explain the magnitude.

The part about equal volume was from sources - there is no multiplier for volume (called contact surface there).

It's the liquid temperature in the pipes I show just before the output to eliminate any heat destruction bugs. And the liquids are pumped from above, go through the medium and is then dumped below but as I said, it's the liquid temp in the pipe just before the vent that is shown, and it's abyssalite pipes.

I could have just dumped it into a common tank but I used it as a timer since I knew the flow rate I could just check to see the height to quickly gauge the time remaining from the sofa where I couldn't see the cycle counter clearly.

Link to comment
Share on other sites

1 minute ago, Saturnus said:

It's the liquid temperature in the pipes I show just before the output to eliminate any heat destruction bugs. And the liquids are pumped from above, go through the medium and is then dumped below but as I said, it's the liquid temp in the pipe just before the vent that is shown, and it's abyssalite pipes.

Yeah but pumping it from below could have some unpredictable effects due to the liquid being forced to split due to pressure - it wouldn't be the first bug resulting from combining small amounts of liquid of one temperature with large amounts of liquid of different temperature.

Have you tried dripping it or only pumping from below? I wouldn't bet that it is the cause of the bug, but it may just be it and I have no other explanation since it clearly doesn't fit the formula.

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.

Please be aware that the content of this thread may be outdated and no longer applicable.

×
  • Create New...