Water cooling system with Peltier thermoelectric support

[Deous]

Peltiers are great, as long as you can figure out how to get what you want from them.

Yes exactly ) Peltiers should never try to take the entire heat overload from the system - it is critical in efficient design.
They can be used to cool down the water in the cpu/gpu loop and additionally add delta to cpu block.
I mentioned once that if you put peltier on top of water block it will give better results then underneath it
assuming you put some good dissipating device on top of tec. Usually a second loop or extra pathway to radiator or coil.
It is amazing that tecs work fine despite their one side can get really hot while other one is freezing.

 

[Silver Surfer]

Yes exactly ) Peltiers should never try to take the entire heat overload from the system - it is critical in efficient design.
They can be used to cool down the water in the cpu/gpu loop and additionally add delta to cpu block.
I mentioned once that if you put peltier on top of water block it will give better results then underneath it
assuming you put some good dissipating device on top of tec. Usually a second loop or extra pathway to radiator or coil.
It is amazing that tecs work fine despite their one side can get really hot while other one is freezing.

Do you actually have anything experimental or operational, or are you just theorizing and entertaining the concept?

 

[Deous]

Do you actually have anything experimental or operational, or are you just theorizing and entertaining the concept?

Read my posts. I am yet building tec wc for pc. Haven't seen interesting tec solutions so far

 

[Silver Surfer]

I have read your posts, and looked at the picture you posted, that's why I asked my last question.

How do you plan implementing what you posted?

 

[Conumdrum]

Why do you guys even bother with replies? You're feeding the troll methinks.

 

[Lochekey]

Why do you guys even bother with replies? You're feeding the troll methinks.

Because it is like,

 

[Alaric]

Because it is like,

View attachment 194947

 

[Tír na nÓg]

Read my posts. I am yet building tec wc for pc. Haven't seen interesting tec solutions so far

Well, look better!

Silver Surfer built a very nice TEC/WC setup (check his build log on OCN).

 

[bob4933]

I studied termodynamics btw

If you studied "termodynamics" (sic) you would know there is no such thing as a free lunch. Your thought process is irrevocably incorrect, and it's clear you have no real desire to learn from people that actually use this stuff in the manner you're describing.

Is your next project showing us perpetual energy? Because without that; your little idea isn't going to work.

 

[denali]

Yes exactly ) Peltiers should never try to take the entire heat overload from the system - it is critical in efficient design.
They can be used to cool down the water in the cpu/gpu loop and additionally add delta to cpu block.
I mentioned once that if you put peltier on top of water block it will give better results then underneath it
assuming you put some good dissipating device on top of tec. Usually a second loop or extra pathway to radiator or coil.
It is amazing that tecs work fine despite their one side can get really hot while other one is freezing.

Hi Deous,
Just wanted to chime in to say I am on the same page as you with using a TEC. I know what you are saying about using them efficiently; they can be just as efficient as any other chilling method if used maximally. So I'm very interested in what you have to say and what you're up to. I'm working on plans to cool a 8700k on a Maximus X Formula (so I can cool the VRM) and using TECs in a kind of heat exchanger, chilled side goes to CPU and hot side to a radiator, all in one case, Thermaltake View 71. Nevermind the naysayers, lets go for it! I don't have a huge budget, but I bet it'll be cheaper than all the chiller nonsense. My main concern is protecting the board from dew. I'll compare notes once I see a reply. PMing as well.

 

[Mr.Scott]

The 'naysayers' have experience in TEC's that you do not.
Ignorance is bliss I guess.

 

[denali]

The 'naysayers' have experience in TEC's that you do not.
Ignorance is bliss I guess.

Actually, with proper experience in investment due dilligence research I was able to learn that the most experienced TEC cooling blogger here and the most experienced peltier enthusiast elsewhere seem to agree with our thermaldynamic lexiconed Deous that efficient TEC utilization methods are widely unknown and their misuse is widespread. Since you haven't had anything constructive to say, rather destructive, I think I'll defer. Oh ... and I actually do have experience with TECs and TEGs, thanks for asking.

Ignorance isn't bliss. It causes much uneeded suffering. The only time ignorance is bliss is when everyone around you insists that you remain so so as not to disturb their ideas of what all-knowing is and the pecking order they've created with it. We're all ignorant and shouldn't be abusing each other about it.

 

[Silver Surfer]

The 'naysayers' have experience in TEC's that you do not.
Ignorance is bliss I guess.

Mr.Scott, Deous and denali could very well be the same person look at their join dates, just saying?

They literally have zero longevity here.

A mod could check the IP address to know for sure if the mods have that capability.

We called them Sock Puppets at Toms Hardware.

 

[denali]

Mr.Scott, Deous and denali could very well be the same person look at their join dates, just saying?

They literally have zero longevity here.

A mod could check the IP address to know for sure if the mods have that capability.

We called them Sock Puppets at Toms Hardware.

I'm not a sockpuppet. Why would anyone waste their time trolling a topic like TECs for CPU cooling? That just doesn't make sense. The coincident in join date is probably because we are both thinking about making a build. I do believe I had a login here before a very long time ago, and I'm wondering if they get purged. Please don't waste our time with conspiracy theory, I'm quite serious about this build. If I didn't have over 200 tabs open and felt more like searching through them I'd give you the links right now to show why I think a reasonably efficient TEC cooling system is possible.

Mr.Scott, Deous and denali could very well be the same person look at their join dates, just saying?

They literally have zero longevity here.

A mod could check the IP address to know for sure if the mods have that capability.

We called them Sock Puppets at Toms Hardware.

Isn't a sockpuppet someone who has multi userids on a forum for the purpose of mucking it up with others on one while seeming civilized on the others? Are mutiple userids allowed here?

- - - Updated - - -


Here is the Odd thing. You seem to have accomplished the very thing I am talking about here: http://www.overclock.net/t/1633988/chilled-water-cooling-vs-3-0-build-log Isn't that you??? You claim to have accomplished this with 300 watts of power on 3 TEC. Am I so wrong? Why are you coming at me with a pitchfork?

 

[Alaric]

Even though the technical end of this discussion is way outside my experience, I don't think I would be quick to discount Mr. Scott's input. Over the years I have seen him engage in discussion, as a peer, with some of the best and brightest OCF has to offer on every subject from storage to software to cooling.

 

[Mandrake4565]

Even though the technical end of this discussion is way outside my experience, I don't think I would be quick to discount Mr. Scott's input. Over the years I have seen him engage in discussion, as a peer, with some of the best and brightest OCF has to offer on every subject from storage to software to cooling.

to that!

 

[Lochekey]

Hi Deous,
Just wanted to chime in to say I am on the same page as you with using a TEC. I know what you are saying about using them efficiently; they can be just as efficient as any other chilling method if used maximally. So I'm very interested in what you have to say and what you're up to. I'm working on plans to cool a 8700k on a Maximus X Formula (so I can cool the VRM) and using TECs in a kind of heat exchanger, chilled side goes to CPU and hot side to a radiator, all in one case, Thermaltake View 71. Nevermind the naysayers, lets go for it! I don't have a huge budget, but I bet it'll be cheaper than all the chiller nonsense. My main concern is protecting the board from dew. I'll compare notes once I see a reply. PMing as well.

Let me start by welcoming you to the forums Denali if no one else has. Please do not let the controvesry of this one thread deter you form continueing at this site. With that said...


I believe you are misunderstanding what most of us were posting about in this thread originally. We are not trying to discourage the OP from experimenting with TEC colling but at the same time we would try to give people realistic expectations in new adventures so that they do not become frustrated as easily when something does go wrong. As to the OP stating that he can run peltiers more efficiently than what is done when the peltier is cpu mounted with an air cooler attached. No one will disagree with that statement as it is done using a WC setup alkll of the time but the OP does not seem to have comprehended that the basis of the system he is suggesting has been done many times before. As to his new revolutionary idea of being able to gain more effeciency from the TEC coolers than others have achieved. If he is able to do this I say all the better but at the same time we would like for him to explain what he is trying to do that is different from other systems and back it up with some actual technical support instead of just "I studied thermodynamics and I will explain proper implementation of TECs to you later" some of the basic concepts of peltier cooling seem to have been ignored by the OP such as the Peltier/s have to carry the full thermal load of the system otherwise ambient conditions will override the temperature drop obtained by the TEC.


If you are interested in doing your own setup there is plenty of people here that would be willing to help/ learn from what you are doing. I would suggest you start your own thread with a log of your build as it will serve both as a learning tool for people in the future that are looking to build a similar system as well as allow you to gain suggestiuons and feedback from some of our members on what you are doing. I truly would be pleased if somone is able to design a more efficient way of using peltiers to cool a computer system down but at the same time I would also like to see the science behind what is going on so that others can implement said system instead of just "trust me it is better" think about it from a scientific perspective a new discovery always goes before a peer review before it is accepted as having any actual value to the general community.



Oh and just to lighten up the mood for everyone have a cute holiday cat picture.

 

[Mr.Scott]

Indeed. I am wide open for better TEC implementation, but haven't seen hide nor hair of anything useful from the 1 or 2 new posters, other than 'I will explain termodynamics' or I have an idea.
I'd even entertain a rough draft of what you're trying to accomplish. Until that time something/anything revolutionary is produced, you guys are just talking through your hat. Put up or shut up.

 

[Alaric]

Note, I didn't say he was subtle.

 

[denali]

Thanks for the replies, friendly and blunt. You had me when you grabbed me by the kitten pic.

So, after taking a couple of hours of quality time with my 4yo and the assembly manual for his new playmobil treehous (proper german spelling) I'm gonna take a few minutes to see if I can find my golden key, the holy grail document proving without a doubt that TEC cooling can be done as efficiently as any other method, in my current tabs.

This is not the document Ive been looking for, but it seems along the same lines. Also, I AM JUST BEGINNING on this topic. I do admit this. I'm not done researching, not even close. I've just begun. I almost hung up this pursuit just yesterday when I began to consider TEC efficiency, as it does concern me as my electric bill at home her in Houston is already astronomical without another constant power draw other than a server, a desktop, and a bunch of other appliances that run full time because I work out of my home and my 80yo MIL is with us ... every day ... all day long ... never lea ... I digress. So I get it.

But I need SILENCE, and I'll pay a premium for it, as long as it's not too crazy, too much stupid tax. And I've always wanted to do this. This is gonna be Fun! Let's play, not fight. I really am a caring guy and I know most of you probably just have your troll defenses up, but I just wanna be friends, learn, discuss openmindedly, and have fun. I appreciate the vast amounts of "institutional knowledge" you all share, and it's the reason why I'm here. I hope eventually you'll appreciate that I may have a tendency to, yes, break the rules, in a desparate attempt to meet goals, and I am wierd enough picking through the garbage as I do to often see things others on the path may have walked by.

So here is a chip off of the holy grail: http://www.overclock.net/t/110820/tec-efficiency/30#post_23311124
What I've been looking for elsewhere is a site many of you probably are aware of already something like "everything you ever wanted to know about TEC". What I see "experts" (physicists themselves) saying is that TEC's are Extremely finicky and not very stable by themselves. Find the top of their efficiency curve is very difficult, and getting them to stay at the top of their efficiency curve is evern more so, dependent on a handful of difficult to read variables. That if you can get a handle those things you can get efficiencies on par with freon pumped coolers. That's all I am asking for. And it actually seems it has been done here. So other than semantic differences I don't see why all the fuss for declaring to embark on the same quest.