Jump to content

Transient Hot Wire Thermal Conductivity Testing of Fluids

mayhems

We have started with THW (Transient Hot Wire) testing liquid coolants with real data of how thermally conductive liquids are. This is an unbiased way of testing however does not take into account long-term usage or any other issues or in system usage. This is pure analytical, research, development data which is extremely handy for working out information when developing new fluids or looking at what's on the market. There will be no comments on which is best (this is not our aim) nor is it our aim to have a go at coolant makers its simply real data so users can see data that is never published to users and forgoes all the "Claims" they make. There is a lot more work to do and to be added, this is a Work in progress project.

This list will grow with time as we test each coolant we can get our hands-on.

Tests carried out to ASTM D7896-19 Standards.

  • W/mK stands for Watts per meter-Kelvin. It’s also known as ‘k Value’. The comparison of thermal conductivity can be measured by the ‘k’ value. The k value, or Thermal Conductivity, specifies the rate of heat transfer in any homogeneous material. If a material has a k value of 1, it means a 1m cube of material will transfer heat at a rate of 1 watt for every degree of the temperature difference between opposite faces. The k value is expressed as 1 W/mK. The lower this value is, the less heat the material will transfer.
  • AVG Drift is the result of the drift calculation that was conducted prior to measurement.
  • Delta T Is the total change in temperature over the course of the measurement
  • Test Time is he time in seconds over which the test was conducted Ambient Temperature The temperature of the sensor at the time of the test (in excel sheet)
  • Conductivity The thermal conductivity of the sample (in the excel sheet)
  • Current is The current provided to the sensor during measurement (in excel sheet)


I have spoken with the company who we have bought the thermal testing equipment and shown my results and they are extremely close to their own testing and within their 5% tolerance. They have confirmed everything I am doing is correct. -> thermtest if you read the site it will help you understand the results below.

There is a 5% tolerance in testing and within a 2% rest tolerance. Any reading that I cannot retest within a 2% tolerance is disregarded and redone until this is achieved.

Mayhems - Ultra-Pure H20
Test Date: 24/02/21
AVG W/mk reading : 0.599 λ (W/mk)
AVG Drift: 0.2 °C
AVG Delta: 2.8 °C
Freeze Protection: 0°C
Excel raw data file: Mayhems Ultra Pure Excel
PDF Report round-up: Mayhems Ultra Pure PDF

EKWB - Cyrofuel - Clear
Test Date: 25/02/21
AVG W/mk reading : 0.515 λ (W/mk)
AVG Drift: 0.5 °C
AVG Delta: 7.7 °C
Freeze Protection:- 3°C
Excel raw data file: Cyrofuel - Clear Excel
PDF Report round-up: Cyrofuel - Clear PDF

EKWB - Cyrofuel - Solid White
Test Date : 25/02/21
AVG W/mk reading : 0.555 λ (W/mk)
AVG Drift: 0.3 °C
AVG Delta: 0.4 °C
Freeze Protection:- 3°C
Excel raw data file: Cyrofuel - Solid White Excel
PDF Report round-up: Cyrofuel - Solid White PDF

ThermalTake - P1000 White
Test Date : 25/02/21
AVG W/mk reading : 0.530 λ (W/mk)
AVG Drift: 0.3 °C
AVG Delta: 2.5 °C
Freeze Protection:- 4°C
Excel raw data file: P1000 White Excel
PDF Report round-up: P1000 White PDF

Alphacool - Cape Kelvin Catcher Clear
Test Date: 25/02/21
AVG W/mk reading : 0.585 λ (W/mk)
AVG Drift: 0.2 °C
AVG Delta: 4.2 °C
Freeze Protection:- 1°C
Excel raw data file: Cape Kelvin Catcher Clear Excel
PDF Report round-up: Cape Kelvin Catcher Clear PDF

Mayhems - XT-1 Nuke Premixed Clear
Test Date: 26/02/21
AVG W/mk reading : 0.502 λ (W/mk)
AVG Drift: 0.3 °C
AVG Delta: 2.5 °C
Freeze Protection: -12°C
Excel raw data file: Download
PDF Report round-up: Download

Mayhems - X1 Premixed Clear
Test Date: 26/02/21
AVG W/mk reading : 0.504 λ (W/mk)
AVG Drift: 0.2 °C
AVG Delta: 3.1 °C
Freeze Protection: -8°C
Excel raw data file: Download
PDF Report round-up: Download

Mayhems - Ultra-Pure H20 / Mayhems Biocide+ / Mayhems Inhibitor+
Test Date: 26/02/21
AVG W/mk reading : ? λ (W/mk)
AVG Drift: ? °C
AVG Delta: ? °C
Freeze Protection: 0°C
Excel raw data file: ?
PDF Report round-up: ?


--== More tests to come and be added to the list ==--

More will be added as we test more. This will take time. We are also looking for ppl to send us samples of coolants we only need 50ml all testing is free of charge. We are aiming to start with just clear and white coolants and will look at colors are a different date and we may also start going into alternative coolants.

Finally, all this data is mine and mine alone, if you wish to use this data in any future work please reference Mayhems (me) as the protagonist. Basically, you can use my data any way you wish but make sure you say where you got it from.

Updates: 26/02/21
Added real freeze protection data (not taken from websites of the makers of coolants) actual testing in-house to find real freeze points. This is important as it helps to understand more why a coolant's λ (W/mk) may be lower. E.g The benefit of freeze protection may be more important than the heat load capacity of the fluid.

 

Link to comment
Share on other sites

Link to post
Share on other sites

To what extent is the thermal conductivity of the liquid important? I ask because my understanding is that the pump and flow is what is primarily the way in which energy is transferred from A to B (I.e. water block to radiator), and by far the rate limiting step is heat being removed from the radiator into ambient air. 
 

Is there a big bottleneck in the heat transfer between nickel (waterb lock)and the coolant, and then coolant to brass/copper (radiator).

 

While the numbers are appreciated, ultimately what it looks like is that they are all give or take similar to water, and so some more context for the general reader would be appreciated.
 

What do these numbers mean and what is an impactful difference. what is the significance of a high delta - did it just get nuked more (higher current?) or is it a reflection of some property of the coolant.

Link to comment
Share on other sites

Link to post
Share on other sites

8 hours ago, For Science! said:

To what extent is the thermal conductivity of the liquid important? I ask because my understanding is that the pump and flow is what is primarily the way in which energy is transferred from A to B (I.e. water block to radiator), and by far the rate limiting step is heat being removed from the radiator into ambient air. 
 

Is there a big bottleneck in the heat transfer between nickel (waterb lock)and the coolant, and then coolant to brass/copper (radiator).

 

While the numbers are appreciated, ultimately what it looks like is that they are all give or take similar to water, and so some more context for the general reader would be appreciated.
 

What do these numbers mean and what is an impactful difference. what is the significance of a high delta - did it just get nuked more (higher current?) or is it a reflection of some property of the coolant.

The numbers are pure unadulterated numbers and there is no way any one can give you perfect numbers when you take into account all factors in play. E.g Pumps, Rads, Flow, Pressure and most of all users ect. These number are relative only to the products tested and are there real true reading. This data set is still being worked on and as you will understand is a slow process and very much unbiased. It doesn't tell everything you need to know as the puzzle is very large.  However to you and me its a good data set that has never been done before as just like thermal compound is relative and has a λ (W/mk) rating but once again is just part of a long list e.g adding a cold plate, adding an air cooler or a liquid cooler. Just because a Normal user may not fully understand doesn't make it less relevant.

 

If you read the excel sheets and check the tabs at the base you'll see all the charts and all the data relevant to some of your questions. 

 

This is a work in progress project though. But thank-you for your feed back and have updated some info above.

Link to comment
Share on other sites

Link to post
Share on other sites

3 hours ago, mayhems said:

The numbers are pure unadulterated numbers and there is no way any one can give you perfect numbers when you take into account all factors in play. E.g Pumps, Rads, Flow, Pressure and most of all users ect. These number are relative only to the products tested and are there real true reading. This data set is still being worked on and as you will understand is a slow process and very much unbiased. It doesn't tell everything you need to know as the puzzle is very large.  However to you and me its a good data set that has never been done before as just like thermal compound is relative and has a λ (W/mk) rating but once again is just part of a long list e.g adding a cold plate, adding an air cooler or a liquid cooler. Just because a Normal user may not fully understand doesn't make it less relevant.

 

If you read the excel sheets and check the tabs at the base you'll see all the charts and all the data relevant to some of your questions. 

 

This is a work in progress project though. But thank-you for your feed back and have updated some info above.

Here are more points that I would like to see addressed:

 

With respect to the data and for purposes of consistency:

1a - You want to compare the thermal conductivity between the coolants, however not enough parameters are being controlled at the moment. Thermal conductivity has a dependence on temperature, and so since the coolants are being heated to different degrees (a consequence of variable current?) and ending up at different final temperatures, the thermal conductivity are not comparable between datasets. I am 

 

This really needs to be corrected, or as a bare minimum the temperature quoted: (e.g. 0.599 λ (W/mk) @ 293 K)  for the dataset to have any value whatsoever. Otherwise the values are not comparable between coolants. 

 

1b - Can you depict the experimental setup? How are you controlling the temperature of the starting point? Is this conducted in a waterbath of some kind,  is the coolant stirred to mimic the effects of having a pump?

 

With respect to the relevance of data

 

2a - Thermal compound is functionally very different from coolant. Thermal compound sits statically between elements and purely transfers its heat through itself when required. And thus thermal conductivity is a direct and relevant metric for this case. However, as mentioned above, coolant moves heat primarily through actual movement of the liquid via a pump. The coolant does not move its heat from one source to another through conduction, and so is this metric really meaningful to depict the performance of the coolant? An extreme analog of this would be that you could theoretically quantify the calorific value (Kcal per 100 ml, for example) of all the coolants, but the whether one coolant would fatten you up more if you were to drink it is absolutely inconsequential and so should not be reported as a metric implying it has some relevant.

 

2b - Related to 2a, ti is implied in the opening paragraph that the thermal conductivity can be used to infer something about the manufacturers claims. This needs to be clarified further as to exactly what you intend on revealing. Does having a lower thermal conductivity have an measurable impact on the performance of the coolant? If it does, how low is too low, and if it doesn't then why is it being quantified in the first place. 

 

2c - Although you say you are unbiased, you do need to acknowledge that there is a conflict of interest here as you are a representative of a commerical coolant company. 

Link to comment
Share on other sites

Link to post
Share on other sites

4 hours ago, mayhems said:

The numbers are pure unadulterated numbers and there is no way any one can give you perfect numbers when you take into account all factors in play. E.g Pumps, Rads, Flow, Pressure and most of all users ect. These number are relative only to the products tested and are there real true reading. This data set is still being worked on and as you will understand is a slow process and very much unbiased. It doesn't tell everything you need to know as the puzzle is very large.  However to you and me its a good data set that has never been done before as just like thermal compound is relative and has a λ (W/mk) rating but once again is just part of a long list e.g adding a cold plate, adding an air cooler or a liquid cooler. Just because a Normal user may not fully understand doesn't make it less relevant.

 

If you read the excel sheets and check the tabs at the base you'll see all the charts and all the data relevant to some of your questions. 

 

This is a work in progress project though. But thank-you for your feed back and have updated some info above.

I don't think @For Science!expressed doubts about the measurements per se except maybe precision. Neither do I. The question is how much thermal conductivity of the coolant influences a loop's performance. Don't get me wrong: more data is always welcome especially with watercooling where valid data is often scarce. I can't remember the equations on the spot to calculate heat transfer in relation to surface, I need to look into some of my old textbooks and scripts but my gut feeling tends to the side of the coolant's k property not playing much of a role within a reasonable property range. It's a gut feeling and I myself or someone else should probably calculate how much influence a difference of .084 W/mK (~16%) has on real world loop applications.

Use the quote function when answering! Mark people directly if you want an answer from them!

Link to comment
Share on other sites

Link to post
Share on other sites

Each coolant has a diffrent electro conductivity, so before each test run a automatic work out of what Current is needed (mA). If you test all coolants across the same current you will get diffrent reading. Also before each test we test direct against the DFU sample that we have and if its not correct or the same with in 2% then its re done until this is achieved to give us the correct base line reading. Then tests are run and the reading taken, A second set are also run to confirm the readings to make sure we get the correct data.

 

All work is done to ASTM D7896-19 Standards and also I have passed my work onto the supplier of the equipment to verify its being done correctly and all has been confirmed and with in the 2% error of correction. Our data is also being passed onto uni's to be verified and VSG is one of them who now works up the road from me. (test equipment is in the PDF's)

 

Equipment used for testing.

·        Thermtest Measurement Platform MP-2 ATSM and ISO complaint

·        Thermtest THW-L3 (Transient Hot Wire)

·        DIUF Water (ASTM TYPE II) 0.589 W/mK @20°C calibration fluid.

·        Methanol Alcohol 99.95% pure Cleaning fluid

 

If you wish to conduct your own testing and mythology you are welcome to and you can question my results if you wish how ever they are all stated above and will be updated as I work along. All I am doing it creating a database of the information for ppl to use as they wish. It will be updated as we add more information.

 

 

Link to comment
Share on other sites

Link to post
Share on other sites

28 minutes ago, mayhems said:

Each coolant has a diffrent electro conductivity, so before each test run a automatic work out of what Current is needed (mA). If you test all coolants across the same current you will get diffrent reading. Also before each test we test direct against the DFU sample that we have and if its not correct or the same with in 2% then its re done until this is achieved to give us the correct base line reading. Then tests are run and the reading taken, A second set are also run to confirm the readings to make sure we get the correct data.

 

All work is done to ASTM D7896-19 Standards and also I have passed my work onto the supplier of the equipment to verify its being done correctly and all has been confirmed and with in the 2% error of correction. Our data is also being passed onto uni's to be verified and VSG is one of them who now works up the road from me. (test equipment is in the PDF's)

 

Equipment used for testing.

·        Thermtest Measurement Platform MP-2 ATSM and ISO complaint

·        Thermtest THW-L3 (Transient Hot Wire)

·        DIUF Water (ASTM TYPE II) 0.589 W/mK @20°C calibration fluid.

·        Methanol Alcohol 99.95% pure Cleaning fluid

 

If you wish to conduct your own testing and mythology you are welcome to and you can question my results if you wish how ever they are all stated above and will be updated as I work along. All I am doing it creating a database of the information for ppl to use as they wish. It will be updated as we add more information.

 

 

The question is: how relevant is λ in the context of a moving fluid in a cooling loop? A cooling loop represents a system with forced convection. Properties like density and kinematic viscosity are vital information or otherwise calculations are impossible. You need those properties first to calculate the Reynolds number in such a system (plus the channel dimensions and flow rate). You need to know the Reynolds number to calculate the Nusselt number. You'll also need to calculate the Prandtl number (ratio between kinematic viscosity and thermal diffusivity) for which you will need the density, specific heat capacity and viscosity.

 

On its own λ does not mean much. You should at least also measure density and viscosity. With that in mind calculations are possible to what extent smaller differences in λ may affect a loop's performance. However, the jet plates of most modern cooling blocks make these kind of calculations something way more complicated. Radiators are easier to calculate than the water blocks.

 

Edit: and of course specific heat capacity.

Use the quote function when answering! Mark people directly if you want an answer from them!

Link to comment
Share on other sites

Link to post
Share on other sites

Here would be examples of a few "coolants" that I would anticipate that would change the view of the usefulness of this metric:

 

Mercury - Would probably score very highly in terms of thermal conductivity. But nonetheless a very bad choice as a coolant

Liquid metal - Would probably score very highly in terms of thermal conductivity. But nonetheless a very bad choice as a coolant

Diluted propylene glycol  - Probably significantly lower thermal conductivity than water, but commonly used in PC coolants with no noticeable impact on thermals

Diluted glycerol - Same as PG

Saturated ammonia solution - Probably similar to water

 

So apart from my points above how the thermal conductivity cannot be compared with each other in their current state anyway (thus rendering any attempt of comparison between the coolants based on this number alone incorrect), overall I question what is trying to be demonstrated here. I think it would not be so farfetched that there is a motive here to demonstrate that one coolant is better than another due to having a higher thermal conductivity. I acknowledge that OP explicitly writes this is not the case, and I will only say that I agree with this 100%  and people reading this thread should not put value onto this number for any decision making purposes.

 

The data is of course fine and I am sure it is high quality. Coolant X has a thermal conductivity of Y at temperature Z. The thing I urge all readers of this thread is that there is little relevant information you can get out of this piece of data that will be able to discern anything related to PC cooling. As it stands, this is just a database of one many parameters that could have been quantified, and whether a coolant is high or low on this parameter bears no significance. An easy to get analogy would be if somebody measure  

 

Other pieces of data that could be quantified and would have similar value:

LD50 - How toxic a coolant is (i.e. how many mg of coolant required to kill 50% of test population)

Conductivity - How electrically conductive a coolant is

Scoville score - How spicy a coolant is

Calorific content - How fattening a coolant may be

Melting/Boiling/Flash point 

pH

etc etc etc

 

They all have a number which could be determined and for a specific use-case may even be relevant; thermal conductivity is one of these amongst others. But unlike for thermal pastes, it does not seem like the relevant metric to draw any kind of conclusion or worthwhile information. In fact, if you were to get some kind impression of a coolant from the thermal conductivity, then you are likely misinforming yourself ( abit like m0ar cores, and m0ar hertz = better, mentality).

 

Link to comment
Share on other sites

Link to post
Share on other sites

In the end is this going to work out as a percentage of the delta T of a CPU?  "" This coolant is 10% higher thermal conductivity so instead of the CPU being +20C delta over the coolant under 300W load it ends up being 18C?

 

You're basically testing the same liquidsso I can see a lot of variables "cancelling out" until it's down to the thermal conductivity of the fluid.

 

If the end game is to show that XTR with nano particles is higher thermal conductivity so it's better...I'm going to need to see how it reacts with microfin waterblock structures and whatnot anyways.  Because those particles are a difference vs. regular water.  It would honestly be more practical to "get to the point" and setup a simulated heat load with a couple waterblocks, run all these fluids in the same loop setup, and then show actual numbers.

 

Interesting stuff anyways...makes my head churn (pun not intended) thinking about how heat transfer with a fluid actually happens in a waterblock.

Workstation:  13700k @ 5.5Ghz || Gigabyte Z790 Ultra || MSI Gaming Trio 4090 Shunt || TeamGroup DDR5-7800 @ 7000 || Corsair AX1500i@240V || whole-house loop.

LANRig/GuestGamingBox: 9900nonK || Gigabyte Z390 Master || ASUS TUF 3090 650W shunt || Corsair SF600 || CPU+GPU watercooled 280 rad pull only || whole-house loop.

Server Router (Untangle): 13600k @ Stock || ASRock Z690 ITX || All 10Gbe || 2x8GB 3200 || PicoPSU 150W 24pin + AX1200i on CPU|| whole-house loop

Server Compute/Storage: 10850K @ 5.1Ghz || Gigabyte Z490 Ultra || EVGA FTW3 3090 1000W || LSI 9280i-24 port || 4TB Samsung 860 Evo, 5x10TB Seagate Enterprise Raid 6, 4x8TB Seagate Archive Backup ||  whole-house loop.

Laptop: HP Elitebook 840 G8 (Intel 1185G7) + 3080Ti Thunderbolt Dock, Razer Blade Stealth 13" 2017 (Intel 8550U)

Link to comment
Share on other sites

Link to post
Share on other sites

2 hours ago, AnonymousGuy said:

In the end is this going to work out as a percentage of the delta T of a CPU?  "" This coolant is 10% higher thermal conductivity so instead of the CPU being +20C delta over the coolant under 300W load it ends up being 18C?

 

You're basically testing the same liquidsso I can see a lot of variables "cancelling out" until it's down to the thermal conductivity of the fluid.

 

If the end game is to show that XTR with nano particles is higher thermal conductivity so it's better...I'm going to need to see how it reacts with microfin waterblock structures and whatnot anyways.  Because those particles are a difference vs. regular water.  It would honestly be more practical to "get to the point" and setup a simulated heat load with a couple waterblocks, run all these fluids in the same loop setup, and then show actual numbers.

 

Interesting stuff anyways...makes my head churn (pun not intended) thinking about how heat transfer with a fluid actually happens in a waterblock.

Nah we already know that the performance of glycol is about the same as water despite it being 3 or 4 times less thermally conductive.

Link to comment
Share on other sites

Link to post
Share on other sites

1 hour ago, For Science! said:

Nah we already know that the performance of glycol is about the same as water despite it being 3 or 4 times less thermally conductive.

So that would imply that the dominant factor is probably the IHS's thermal resistance or maybe for GPU's the silicon itself since its' going through the die.  With water flowing through a block I imagine the "boundary" where heat transfer is actually happening is very thin right next to the metal.  

 

We already know thermal paste doesn't really matter very much because it's so thinned out.  And waterblock doesn't really matter very much since the best and worst is within a couple degrees of each other.

 

Also do people ever run pure glycol?  I thought even if someone runs 50/50 mix from an auto parts store it's only about 3/4 the thermal conductivity of water, and whatever is in X1 is diluted down 10/90.

 

EDIT: also this gets me thinking that if we get a fluid that is wildly great at heat transfer on it's own, maybe it would make direct-die (no waterblock / IHS) cooling actually work.

Workstation:  13700k @ 5.5Ghz || Gigabyte Z790 Ultra || MSI Gaming Trio 4090 Shunt || TeamGroup DDR5-7800 @ 7000 || Corsair AX1500i@240V || whole-house loop.

LANRig/GuestGamingBox: 9900nonK || Gigabyte Z390 Master || ASUS TUF 3090 650W shunt || Corsair SF600 || CPU+GPU watercooled 280 rad pull only || whole-house loop.

Server Router (Untangle): 13600k @ Stock || ASRock Z690 ITX || All 10Gbe || 2x8GB 3200 || PicoPSU 150W 24pin + AX1200i on CPU|| whole-house loop

Server Compute/Storage: 10850K @ 5.1Ghz || Gigabyte Z490 Ultra || EVGA FTW3 3090 1000W || LSI 9280i-24 port || 4TB Samsung 860 Evo, 5x10TB Seagate Enterprise Raid 6, 4x8TB Seagate Archive Backup ||  whole-house loop.

Laptop: HP Elitebook 840 G8 (Intel 1185G7) + 3080Ti Thunderbolt Dock, Razer Blade Stealth 13" 2017 (Intel 8550U)

Link to comment
Share on other sites

Link to post
Share on other sites

42 minutes ago, AnonymousGuy said:

So that would imply that the dominant factor is probably the IHS's thermal resistance or maybe for GPU's the silicon itself since its' going through the die.  With water flowing through a block I imagine the "boundary" where heat transfer is actually happening is very thin right next to the metal.

It's all about the surface area - on both sides. The die havsa certain surface area which connects to the IHS - which again has a certain surface area that makes contact with the coldplate. That coldplate has ideally a micronfin structure which again sums up to a certain surface area making contact with the coolant. The coolant runs through the radiator which has small channels the coolant needs to pass through - the more of those channels you can fit the more surface area comes into contact with the water. On the outside the radiator has a fin structure connected to those channels increasing the surface area that comes into contact with the air.

 

Certain factors are more important in this process than others. 

 

Watercooling loops represent a system of forced convection. Thermal conductivity has some part in the process of calculating heat transfer but it is just a part. Other factos are density, heat capacity, viscosity, surface area, whether it's laminar or turbulent flow, etc.

 

Let's call it an educated guess, but my bet would be that thermal conductivity plays only a small to negligible role in this scenario but we'd need to run some numbers. It would be enough to have a rough approximation of a loop and then calculate it once with the heat conductivity of water, once again with a 100% larger heat conductivity and to be sure once again with a 50% lower heat conductivity.

 

That will tell you by how much this property influences the overall performance of a loop.

Use the quote function when answering! Mark people directly if you want an answer from them!

Link to comment
Share on other sites

Link to post
Share on other sites

  • 1 month later...

Can you do a better comparison test on heat transfer fluids, 

 

remarks so far seem to be hung up on the thermal conductivity and miss the Benifit of higher thermal carrying capacity of higher specific gravity fluids and anti corrosion additives already in RV and vehicle glycols. 

 

I wondered if you can use rtf propolyne glycol for RV,s with the cheap 80$ rig. 
 

the anti corrosion, and other additives remove the electrolytic reaction between dissimilar metals. 
 

also the reason Glycol performs as well as water straight up despite lower conductivity is the higher thermal carrying Capacity of the higher Specific Gravity of the solution. 
60/40 ratios allow high conductivity AND a large portion of the 12X higher thermal capacity per cc. 
 

Same reason a car boils if you use only water. 
it cant carry as much heat To the radiator without higher flow than allows dissipation in the Rad. 
 

By all ive seen in heavy industry using a standard RV PROPYLYNE Glycol 60/40 mix should lower your baseline idle temp and maximum stress temps by 20%. 

I can't see how computer applications can have less effect by fluid thermal carrying than ive seen in heavy industry. 
 
 

Link to comment
Share on other sites

Link to post
Share on other sites

On 2/26/2021 at 2:15 AM, mayhems said:

The numbers are pure unadulterated numbers and there is no way any one can give you perfect numbers when you take into account all factors in play. E.g Pumps, Rads, Flow, Pressure and most of all users ect. These number are relative only to the products tested and are there real true reading. This data set is still being worked on and as you will understand is a slow process and very much unbiased. It doesn't tell everything you need to know as the puzzle is very large.  However to you and me its a good data set that has never been done before as just like thermal compound is relative and has a λ (W/mk) rating but once again is just part of a long list e.g adding a cold plate, adding an air cooler or a liquid cooler. Just because a Normal user may not fully understand doesn't make it less relevant.

 

If you read the excel sheets and check the tabs at the base you'll see all the charts and all the data relevant to some of your questions. 

 

This is a work in progress project though. But thank-you for your feed back and have updated some info above.

Thermal transfer within the fluid is more a hydraulic cooling issue where the tank IS the cooler. 
 

It becomes irrelevant in flow turbulence only millimeters from the surface 

but where it matters is drawing that heat from the plate (fins surface area) as fast as possible. 
BUT pure water lacks the heat carrying capacity of glycols so requires much higher flow or it can boil and vapourlock 

correct mixes can carry larger temp differential increasing the heat siphoning much higher than water. 

Link to comment
Share on other sites

Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

×