Twisted tubing for liquid cooling ?

[Call_Me_Daddy]

Hi Guys,

 

 

You know what ? I like to think, like a lot. Feel free not to read this thread if you don't like very hypothetical ideas.

As some of you might have seen, I'm currently building my very first custom watercooling loop and I REALLY LOVE understanding what I do and using my engineering skills to try & do it the way I think is the best.

 

This topic is definitely something that I'd be doing with my own system because it would require a precision that I wouldn't be able to achieve and for a minor (if existing) cooling performance improvement.

 

So ! As far as I've understood, the point of cooling a broadband computer is keeping its internal components temperature as close as possible to the ambiant temperature. Air cooling transfers heat to a rad from which heat will be extracted by one or more fans. Liquid cooling has exactely the same purpose, except that waterflow will help transferring the heat more rapidly and provide a more efficient cooling.

 

So the whole point is taking warm water to some metal parts from which heat will be extracted by blowing.

 

I've seen lots of things being done to increase this performance. Some of them may be only marketing and low efficient in real conditions but let's take an example : I owned a small form factor reservoir, the phanteks R160 which has an integrated heatsink on its back. The reservoir mounts directly to a fan so that it dissipates more heat.

 

Which finally bringed me to this idea : how else could we increase even a little heat dissipation from a watercooling loop at a low price ?

 

And there I came with this idea. In the industry, tubing is very often used as a heat dissipation component.

Some of them have a heat sink circulary mounted all around them. This, in nickel + copper version would IMO be a VERY great idea but increase the cost and space usage way too much in a situation where cost and space management are very important to us. They would be a nightmare to bend cut and fit. I think it would still be interesting to see how it performs - Video idea ? ( : - There's an example image of this tubes :

 

 

Which finally led me to this idea : Some mid range heat transfer tubes are just regular tubes that were twisted. They dont take more space, they aren't much more pricey to produce, don't reduce water flow efficiency but they maximize the contact surface with air so more heat is being dissipated with a same tubing length. Here's an example :

 

I think this would be easy to manufacture both with PETG or Copper like Linus did on one of his video and could increase - even if very low increase - the heat dissipation and therefore the cooling performance.

 

And it's so easy to make and visually speaking that I'm surprised that GAMIIIIIING stuff manufacturers didn't do it yet as the century idea. With catching titles like this : « up to twice the heat dissipation performance compared to classic tubing ! » with nice charts with no legend.

 

 

 

Thanks for reading !
I'd be glad to have your thoughts about this.

 

This is not at all trolling, just sharing some ideas with low impact but lots of thinking to be made.

 

 

Vincent

[Bananasplit_00]

We have had stuff like reserators that were a reservoir with finns, didn't do amazing things for temps so it died. This would look cool but most people prefer seeing their coolant in the tubes which any material that could transfer enough heat to make a difference won't allow. PETG isn't a great thermal conductor

[Call_Me_Daddy]

11 minutes ago, Bananasplit_00 said:

We have had stuff like reserators that were a reservoir with finns, didn't do amazing things for temps so it died. This would look cool but most people prefer seeing their coolant in the tubes which any material that could transfer enough heat to make a difference won't allow. PETG isn't a great thermal conductor

I totally agree with you.

Even though I loved their copper tubing build people prefer (they still don’t have much choice from pc cooling retailers) clear tubing.

 

But even if they are bad at conducting heat,  considering absolute values, let’s say that you win 50% increase in performance. Companies like aorus would love to brand this increase haha. 

 

Have a a look at this chart about their z390 master mosfet heatsinks (picture attached)

 

no unit, no value, wrong base reference (what motherboard don’t use heat sinks now) ?

They even didn’t put this chart on their Z390 pro product page as they must be so close that it wouldn’t be significantly different.

but that’s still +30% dude  

 

 

jokes apart I just mean that it would be a non surprising play from manufacturers to advertise a % improvement even if the value is still very low 

 

 

[Tristerin]

Twisting tubing is expensive.  I work in the Titanium industry and produce tubing for desalinization plants.  The cost increase is insane (and Titanium is easy to twist due to its properties).

 

Second - you are now going to make the point of your loop obsolete.

 

The point of a loop is to not build up temperatures inside of your own case.  You are going to now use the tubing itself to dissipate heat inside the sealed unit you don't want heat building up in.

 

Now, everything you are saying is doable - will have some sort of "result" - but then comes in the terms "efficiency" "cost" and whatnot.  Not only that but twisting tubing creates a stress on the seam required to make tubing (unless its extruded and not welded from strip sheet metal) - that seam will be the first thing to corrode.  Now if its extruded...the cost is MUCH MUCH MUCH higher. 

 

I love to think, talk out my ideas, and do stuff that only gives a 1% improvement as well (regardless of cost).  However the only reason I would have to say this is a terribad idea is that you are now just dissipating heat into the case.  And unless you have airflow in the case its going to just build up.  Then weve gotten to a point where its not yielding any results except hotter temps in the case, more noise because I had to add more fans... etc

 

Because I Could just buy a box fan, 4x360mm rads and keep my temps at 1-2 degree C above ambient with rubber tubes (see Jayz2cents videos on this) if the goal was the best temps possible.

 

 

The last thing is - the performance is WAY far off from 50% increase.  That's a huge margin of success - and it would be a standard at that point.  You are more looking at likely a 1% change in temps, with a production of heat on all the rest of your computer parts now.

[Call_Me_Daddy]

8 minutes ago, Tristerin said:

Twisting tubing is expensive.  I work in the Titanium industry and produce tubing for desalinization plants.  The cost increase is insane (and Titanium is easy to twist due to its properties).

 

Second - you are now going to make the point of your loop obsolete.

 

The point of a loop is to not build up temperatures inside of your own case.  You are going to now use the tubing itself to dissipate heat inside the sealed unit you don't want heat building up in.

 

Now, everything you are saying is doable - will have some sort of "result" - but then comes in the terms "efficiency" "cost" and whatnot.  Not only that but twisting tubing creates a stress on the seam required to make tubing (unless its extruded and not welded from strip sheet metal) - that seam will be the first thing to corrode.  Now if its extruded...the cost is MUCH MUCH MUCH higher. 

 

I love to think, talk out my ideas, and do stuff that only gives a 1% improvement as well (regardless of cost).  However the only reason I would have to say this is a terribad idea is that you are now just dissipating heat into the case.  And unless you have airflow in the case its going to just build up.  Then weve gotten to a point where its not yielding any results except hotter temps in the case, more noise because I had to add more fans... etc

 

Because I Could just buy a box fan, 4x360mm rads and keep my temps at 1-2 degree C above ambient with rubber tubes (see Jayz2cents videos on this) if the goal was the best temps possible.

 

 

The last thing is - the performance is WAY far off from 50% increase.  That's a huge margin of success - and it would be a standard at that point.  You are more looking at likely a 1% change in temps, with a production of heat on all the rest of your computer parts now.

That’s definitely the kind of opinion that I was hoping for. 

 

I didn’t think that it would be both such an increase of costs and a a low increase of performance. I’m more into networking and electronics engineering. How about copper twisting ? 

 

I should have told about it but I was considering a case scenario with rad and intakes at the front and exhaust at the top and rear of the case so air circulation would be good enough but yes that’s still definitely an issue for static air cases. 

 

 

Thanks anyway very  much for this helpful feedback  

[Tristerin]

43 minutes ago, Insorior said:

That’s definitely the kind of opinion that I was hoping for. 

 

I didn’t think that it would be both such an increase of costs and a a low increase of performance. I’m more into networking and electronics engineering. How about copper twisting ? 

 

I should have told about it but I was considering a case scenario with rad and intakes at the front and exhaust at the top and rear of the case so air circulation would be good enough but yes that’s still definitely an issue for static air cases. 

 

 

Thanks anyway very  much for this helpful feedback  

The performance delta is negligible enough that the desalinization industry doesn't rely on twisted tubing (because cost) - same with the pool and spa industry (when it comes to temp controlled high end stuff) - I have one customer who RnDs twisted tubing still, but the issues its causing is the already stressed seam that corrosion attacks first becomes weaker with the added stress (so you now have to do an interior and exterior weld on the tube - doable but double cost for the welding portion) of stretching the material - coupled with making it thinner by stretching the metal when being twisted which causes the walls to be thinner meaning your input stock now has to be heavier walled (cost increase, more metal used) to have the same corrosion resistance over time that non-twisting achieves.

 

Next we can look at extruded (think, play-do) is shoving super heated material through a die that ensures there is no seam - but this process is very expensive compared to using sheet material, series of rollers to force the shape, then a tig welder (BASICALLY its much more than this though lol).  Then the next option is trepanning - super expensive.  Trepanning from what I know in the metals industry is usually a last resort option due to costs, the type of equipment required etc.  Its basically taking the solid metal bar, then drilling out the entire center for the full length of this bar now made into a tube.

 

So as for Airflow - I agree, intake and exhausting a rad will have airflow going, and as long as the tubing itself is in that airflow it will benefit - however (thinking out loud here) - first intake radiator is now introducing hot air into the case.  This hot air is now being pushed across the fins of the tubing, heating the air up even more, then that even hotter air is now being used to "cool" down your exhaust radiator.  Then we have to figure out if this delta we just created (more heat on second rad) is not nullifying the benefits introduced with twisted tubing. 

 

Great ideas, and if you had unlimited budget Id say do it cause...because. lol.  But it would take some RnD - maybe propose this to LTT or Jayz?  They have the sponsors and whatnot to do this!!!

[Call_Me_Daddy]

19 minutes ago, Tristerin said:

Great ideas, and if you had unlimited budget Id say do it cause...because. lol.  But it would take some RnD - maybe propose this to LTT or Jayz?  They have the sponsors and whatnot to do this!!!

Why not ! Im quite new to these forums. Is there any proper way to make video suggestions ?

 

 

[Tristerin]

Id start here and see if it gains traction - they have the monies, industry hookups etc to see if this may be more worthwhile than my naysaying is!  There IS a reason twisted tubing is used in heat dissipation...but is it Gamer worthy?  Only 1 way to find out!

 

 

[Call_Me_Daddy]

1 minute ago, Tristerin said:

Id start here and see if it gains traction - they have the monies, industry hookups etc to see if this may be more worthwhile than my naysaying is!  There IS a reason twisted tubing is used in heat dissipation...but is it Gamer worthy?  Only 1 way to find out!

 

 

 

Haha you're strong aswell at making bait-like titles

 

thanks for the link ! I already had a few cool ideas I'll be glad to share them there.

[Tristerin]

23 minutes ago, Insorior said:

 

Haha you're strong aswell at making bait-like titles

 

thanks for the link ! I already had a few cool ideas I'll be glad to share them there.

Ill check back later and support your post - Im curious now as well.  Im the kind of guy who buys shit just to test it (and it pisses my wife off!) to see if I can get that 1% gain from whatever Im trying to gain LOL. 

[Cb2394]

Could you also pair this with a negative pressure inside the line? If you can get the liquid to phase change on the block you might be able to cool it fast enough with this method and a good rad for it to condense nearly instantly, theoretically dispersing more energy then just moving water over the block. Just s thought.

[Call_Me_Daddy]

2 hours ago, Cb2394 said:

Could you also pair this with a negative pressure inside the line? If you can get the liquid to phase change on the block you might be able to cool it fast enough with this method and a good rad for it to condense nearly instantly, theoretically dispersing more energy then just moving water over the block. Just s thought.

I have absolutely no skill into case pressure. My own pc runs in push-pull in a pretty neutral way but that sounds really interesting although again I’m not sure how measurable the gains in cooling would be.

[Cb2394]

9 minutes ago, Insorior said:

I have absolutely no skill into case pressure. My own pc runs in push-pull in a pretty neutral way but that sounds really interesting although again I’m not sure how measurable the gains in cooling would be.

Negative pressure inside the custom water loop, not in the case itself. Like if you pull up a little water in a syringe, seal the end, then pull the plunger on the syringe more you can cause the water to boil. If I remember the science correctly, the more negative pressure, the lower the freezing and boiling point.