Aluminum and copper in the same loop...

[Top Nurse]

For years now Aqua Computer and other imports have taken some real bitch slapping here in this forum for having the audacity to put aluminum and copper in the same loop. Also for years all these users of these products haven't reported corrosion problems when used according to the manufacturers direction.

So now that Swiftech is using aluminum bolted right next to copper in their blocks I have a couple of questions:

Are you guys sorry for bagging on us all this time?

Are you going to start bagging on Swiftech now?

Comment:

Since the imports were correct about this issue do you think that other cooling philosophies (low and medium flow) that the same importers express in their products are now viable water cooling solutions? Or do we have to wait another 3-4 years before the domestic companies follow suit to be be believed?

Edit: Please keep this conversation to a low roar and no FLAMES!

 

[Ranker]

For years now Aqua Computer and other imports have taken some real bitch slapping here in this forum for having the audacity to put aluminum and copper in the same loop. Also for years all these users of these products haven't reported corrosion problems when used according to the manufacturers direction.

So now that Swiftech is using aluminum bolted right next to copper in their blocks I have a couple of questions:

Are you guys sorry for bagging on us all this time?

Are you going to start bagging on Swiftech now?

Comment:

Since the imports were correct about this issue do you think that other cooling philosophies (low and medium flow) that the same importers express in their products are now viable water cooling solutions? Or do we have to wait another 3-4 years before the domestic companies follow suit to be be believed?

Edit: Please keep this conversation to a low roar and no FLAMES!

No I'm not sorry. Swiftech uses military spec plating which is more stringent than the techniques used by any other manufacturer. It will never become an issue with corrosion. Even in the harshest of environments, the plating will not deteriorate to expose the aluminum. The aluminum will never come into contact with the water resulting in the possibility of corrosion.

Now, even having said that, Swiftech caught the mother of all hell from those of us on XS. Suffice to say, I think Gabe and co heard the uproars and won't be likely going down that same path down the future. It was stated it was done purely to make it easier to work on the looks and had nothing to do with performance, something most of us railed him for.

As Marci said over on Thermochill forums, its up to the manufacturers to idiot proof water cooling components in order to prevent the users from ruining their components, turning them off from water cooing altogether and/or making the company liable for the damages. Copper is the only way to go. There's no performance based reason to choose aluminum. The use of aluminum is simply a shortcut that hurts the user in the end. Many of us on these forums are quite competent enough to know that mixing Cu and Alu is bad. It's stressed almost everywhere. However, the majority of water coolers out there don't read [H], XS, overclock, etc. Out of the 15 people that I know whom watercool their PC's, only 2 of them are active on forums. Most of them picked up on watercooling only throught magazines like Maximum and PC-world and learned/purchased their components through only the brief knowledge they gained from that medium. They'll most likely mix blocks, ruin their system, blame the hobby in general, and withdraw their support for the hobby monetarily which hurts the entire scene altogether.

So absolutely a big flat no. The use of Aluminum is malicious towards the end user by harming him and his investment. Those who produce such a product need to get their shit together; even if it's the almighty Swiftech for just one experiment in design. This goes doubly so for all of the other manufacturers out there that continue to cut corners and use aluminum. Did I buy a GTX? Yeap. Will I ever buy another Swiftech product again if they continue to use Aluminum? Nope. Even with the strictest standards, military specifications (which basically means you can run it for 2 decades and the plating will never wear (something that can't be said for various companies who use dubious methods)), I could get similar performance from a D-Tek FuZion and have peace of mind.

Now for the second part of your question, it doesn't really matter what one's cooling philosophy is. One can do whatever one wants to one's system. However, within watercooling, there is only one metric that can objectively be measured for performance: temperature readings. To achieve the most optimal and efficient setup, there is only one that can claim to that throne: high flow systems. Now if you prefer looks, bling, eye candy or simply the fact that you want to be different or that you drive an import setup, then by all means, purchase whatever kit that fits your cooling philosophy.


From the legends, Marci and Cathar:

My stance on the use of aluminium is well-documented both here and on all other sites I frequent... that being, just avoid using it and thus avoid the complications that come alongside it. (ie: Is the anodizing good enough...? Has the enduser done sufficient research to know he HAS to use an additive...? Already there are now two factors out of anyone's control to account for - remove the use of aluminium, remove two uncontrollable factors, and thus remove the need for this entire topic which crops up with monotonous regularity every 6 weeks and has doneso for eons)

If we could have the faith that folks would use additive, then this wouldn't be a problem, but this whole thread highlights a lack of research, plain and simple. For nearing 10 years it's been common knowledge that you need to run anticorrosive additives alongside aluminium blocks in watercooling systems... to miss this fact simply demonstrates that adequate research prior to the purchase of the parts is lacking. This is what happens when watercooling becomes more mainstream. Parts availability becomes easy, and folks don't bother doing their research first.

This info is WIDELY spread across ALL forums that I frequent... it is covered in every watercooling guide I've ever read, and in more or less every watercooling sections' sticky by default... it's almost the first rule of watercooling it's that old.

Ignore the terminology I use in this next sentence, it isn't meant to be insulting... it's just a suitable way of phrasing it and I can't be assed to think of a more P/C way of saying it...

Manufacturers cannot account for the idiot factor (aka net-clue-factor to coin Cathar's way of saying it). If you can't account for the idiot factor, don't build items that an idiot cannot use correctly.

To release alu blocks is milking the market. Manufacturers know there is a HIGH risk of that aluminium block getting destroyed by what I'd say would be a good 50% of the watercooling market (the chunk that don't do their research and just dive in head first). This destruction isn't covered by warranty, so another sale is milked from the equation when that customer requires a replacement. A responsible manufacturer would include a note in with every block clearly stating that an anticorrosive additive MUST be used.

This is what happens when you mix copper and aluminum: http://www.xtremesystems.org/forums/showpost.php?p=2215997&postcount=1

 

[Jag]

To achieve the most optimal and efficient setup, there is only one that can claim to that throne: high flow systems. Now if you prefer looks, bling, eye candy or simply the fact that you want to be different or that you drive an import setup, then by all means, purchase whatever kit that fits your cooling philosophy.

I'm sorry but I disagree.
High flow systems are just a way of implementing a watercooling solution, not the objective itself.
Head on to this website and read this review:http://www.cooling-masters.com/articles-26-0.html

Here you can see that the MCW-6000 reaches the same "performance" as you call it as the 1A-Cooling HV3 block, at higher flow with 3/8 tubing.

The thing that separates these two blocks is the fact that the MCW-6000 is a low restriction block meaning you can use a modest pump with it and still obtain a great performance, and when coupled with a MCR220 rad and quiet fans have a silent loop, whereas, in the 8/6 department you still need a higher head pump (ideally a Laing DDC 10W) to push the water through, altough users typically use a Eheim 1046.

Generally speaking "high flow" systems are costly, and if you are "extreme" even more so.
Dual pumps (or a high head pump, say a Iwaki), dual loops, dual radiators, only to extract a single degree, to me it's not an option.

There's no "throne" here. You as a user when you make your mind to use watercooling as the cooling solution for you PC, must set some objectives, otherwise you may find youself very frustrated, if your only objective is pure performance, or you have to escalate your system=more cost and eventually more noise.

As for bling, it is the thing that promotes sells, and it was shown with the Storm vs. the MCW-6000, DangerDen and others.
Swiftech understood well that lesson and replicated it with the Apogee GTX and the Stealth vs. the standard Apogee, only in this case there's not much of a difference between the two, so to me the price asked for the GTX must come from somewhere else.

In the recent past in another forum, some pro-users also gave the false impression that 1/2or equivalent was the way to go. Now some even changed to aircooling.

People want something to keep them in a daily basis satisfied.That's what they get with aircooling, that they still don't see with watercooling because of these "family wars".

 

[Jag]

dual post
*deleted*

 

[buffbiff21]

Get over it.

High flow = better performance.

In a loop that minimizes tight bends, restrictive blocks, and tube length, high flow will ALWAYS beat low flow.

 

[Jag]

Get over it.

High flow = better performance.

In a loop that minimizes tight bends, restrictive blocks, and tube length, high flow will ALWAYS beat low flow.

Sorry to say, but this is just a typical forum answer.
Do you have anything substancial to add to the topic?

 

[buffbiff21]

Sorry to say, but this is just a typical forum answer.
Do you have anything substancial to add to the topic?

No I'm not going to waste my time arguing something that I know as fact.



You who think otherwise are out to lunch.

 

[Arcygenical]

TN, While I don't agree with the use of aluminum in newer swiftech blocks, they're plated to a millitary grade specification.

No anodization is millitary grade, for the purposes of corrosion protection.

Anyone else have some GOOD viable reviews for any of the low-flow stuff?

 

[Shoggy]

Swiftech uses military spec plating which is more stringent than the techniques used by any other manufacturer. It will never become an issue with corrosion. Even in the harshest of environments, the plating will not deteriorate to expose the aluminum. The aluminum will never come into contact with the water resulting in the possibility of corrosion.

Wow... *yawn*

And other companys are using voodoo magic...?!

That "military spec plating" is nothing else than a alumina (Al2O3) ceramic which is also used by Aqua Computer

The parts are processed by the company Rohde AG which is well known for their high quality.

The surface treatment of one GPU block costs around 5 Euro ($6.70). Other companys would do that for a tenth part but not at that level of quality.

edit: I'm using a mix of aluminium and copper since three years now and I have absolutely no problems.

 

[madmat]

Wow... *yawn*

And other companys are using voodoo magic...?!

That "military spec plating" is nothing else than a alumina (Al2O3) ceramic which is also used by Aqua Computer

The parts are processed by the company Rohde AG which is well known for their high quality.

The surface treatment of one GPU block costs around 5 Euro ($6.70). Other companys would do that for a tenth part but not at that level of quality.

edit: I'm using a mix of aluminium and copper since three years now and I have absolutely no problems.

Hmmm... this:

The housing is CNC machined out of billet aluminum and receives two plating's for a lifetime protection against corrosion: electroless nickel plating (MIL-C-26074E grade B) and Zinc Cobalt plating (ASTM B 840-99 grade 6). A black die is applied after plating strictly for cosmetic appeal.

Doesn't sound like a ceramic coating to me, instead it sounds eerily like a metal plating. Ceramics can chip whereas metal plating is considerably more robust when done properly.

 

[Erasmus354]

This thread is flamebait. It shows the ignorance and willingness to spread FUD of many members of this community. A simple search would show that people are not pleased about the use of aluminum in the block. However it will also show that Swiftech has gone above and beyond to ensure that the use of aluminum does not become a problem.

Is it ideal? No. Does it mean others were right? Absolutely not.

 

[Shoggy]

Madmat, I can only tell you that Aqua Computer is using a alumina ceramic.
It is much more corrosion-resistant than stainless steel for example.

It also doesn't mean that it is ceramic as you think of. Aluminia is only a ceramic raw material which is also used to produce extremely shatter-proof glass
And lately it is also used for the plating of vehicles. It reaches the double protective capacity of tank steel at a similar grammage.

 

[madmat]

Madmat, I can only tell you that Aqua Computer is using a alumina ceramic.
It is much more corrosion-resistant than stainless steel for example.

It also doesn't mean that it is ceramic as you think of. Aluminia is only a ceramic raw material which is also used to produce extremely shatter-proof glass
And lately it is also used for the plating of vehicles. It reaches the double protective capacity of tank steel at a similar grammage.

Yeah I know what it is, they also coat headers with it. Actually, they've been coating headers with it for nearly 25 years IIRC. It doesn't surprise me that it's more corrosion resistant than stainless since pure non-magnetic stainless is very susceptible to acids. Household vinegar will eat through the best stainless in a few days time. Still, a ceramic is a ceramic and a metal plating is just that, a metal plating. That was my whole point.

 

[madmat]

Arg! Double post... bad madmat! Bad, bad madmat!

 

[ohms]

I recently put a Swiftech H20-120 Premium kit into my PC, is this something i'm going to have to worry about?

Are the Apogee GT and MCR-120 QP made before this change by Swiftech?

Also it makes me wonder, I purchased a Thermochill PA-160. I saw it is made of copper/brass, is that going to be ok?

 

[madmat]

The GT is topped by a molded delrin top, not plated aluminum. That's the GTX.

 

[ohms]

Ack! I just realized that my Aqua Computer aquagfx 8800 GTX waterblock is aluminum. Is this really gonna f**k my setup? I'm using Swiftech HydrX coolant which is anti-corrosive but i'm still paranoid about the whole thing.

Another thing is I was thinking about was the D-Tek Fusion block but I was told because of the flow restriction that it's not the most ideal choice with my current pump (MCP350). What I was wondering is if the D-Tek GFX UNI-Sink will work with the Swiftech MCW60 w/ G80 adapter. I don't like the looks of using plain old copper heatsinks for the vcore mosfets, memory and the i/o chipset.

 

[Ranker]

Ack! I just realized that my Aqua Computer aquagfx 8800 GTX waterblock is aluminum. Is this really gonna f**k my setup? I'm using Swiftech HydrX coolant which is anti-corrosive but i'm still paranoid about the whole thing.

Another thing is I was thinking about was the D-Tek Fusion block but I was told because of the flow restriction that it's not the most ideal choice with my current pump (MCP350). What I was wondering is if the D-Tek GFX UNI-Sink will work with the Swiftech MCW60 w/ G80 adapter. I don't like the looks of using plain old copper heatsinks for the vcore mosfets, memory and the i/o chipset.

I'd be wary of ANYTHING that contains aluminum regardless of what it is secured with. Even Swiftech's military plating which is better above all others in the industry in its quality and stringent demands, doesn't give me too much peace of mind. So yes, I'd sincerely suggest you invest in some pentosin although hydrax might work since it's supposed to be a cheap "all in one" type of additive.

The D-Tek GFX block is a flow killer. It's best used in a separate loop. You've got a great pump, but I'd switch to a MCW60 instead if you're going to be using one loop for both CPU/GPU. The uni-sink does work with the MCW60 according to prelim reports. However, just call up Petra and he should be able to give you the definitive answer before you pull the trigger.

 

[ohms]

Thanks Ranker.

I actually just called Petra and bugged him so more (he probably wants to choke me by now) but he's pretty sure the MCW60 will work with the uni-sink but not definite because has has no more MCW60s in stock to test it with

Also thanks again to Petra for answering all my annoying wc noob questions.

 

[Erasmus354]

What is your cpu block? Depending upon your CPU block you might not see much of a difference between the two gfx blocks (in terms of impact on cooling of other components). Your being a bit paranoid Ranker if you think that two blocks can't be in the same loop...even if they both are highly restrictive. Your biggest concern with those two blocks is more heat than the combined restriction.

 

[Ranker]

What is your cpu block? Depending upon your CPU block you might not see much of a difference between the two gfx blocks (in terms of impact on cooling of other components). Your being a bit paranoid Ranker if you think that two blocks can't be in the same loop...even if they both are highly restrictive. Your biggest concern with those two blocks is more heat than the combined restriction.

If he had a block like the D-Tek Fuzion where flow doesn't matter, I'd have told him not to worry about it. According to specs and Petra's own testing, the D-Tek GFX has the same restriction as a Storm. I use the blocks, but I wouldn't recommend them in a loop with any other block that might be sensitive towards flow. But you're right, that's personal preference. Take my suggestion with a grain of salt. Switching back to EK FC's as speak.

 

[ohms]

I have the Apogee GT cpu block, Aqua Computer Aquagfx 8800 GTX block, Thermochill PA-160 rad, Swiftech MCR120 rad all in the same loop being powered by a Swiftech MCP350 (soon to have the Petra top).

Loop is : pump > gpu > mcr120 > pa160 > cpu > res

 

[Halk]

Who in their right mind would put WATER inside a computer?

Well, it turns out water is fine in a computer, as long as you take the appropriate precautions. Make sure it doesn't leak, break, snap and the water escapes.

Who in their right mind would put aluminium inside a cooling loop?

Well, it turns out aluminium is fine in a cooling loop, as long as you take the appropriate precautions. Make sure it doesn't corrode.

As for pipe width? Of course big fat pipes cool better. But does a higher pressure pump. So does silver waterblocks. So does a peltier, or cascade, or LN. You don't need to use any of those to adequately cool a computer. And at the end of the day the difference in a few degrees does not make or break an overclock, except in very rare circumstances. Low flow is elegant, more easy to control, requires a quieter pump, which means it can be quieter.

 

[Erasmus354]

Low flow is elegant, more easy to control, requires a quieter pump, which means it can be quieter.

1) Elegance is opinion not fact, there are plenty incredibly clean and elegant high flow, large bore systems. There are also plenty of nasty spaghetti mess low flow small bore systems.

2) Controls have nothing at all to do with the flow of a system or the size of the tubing.

3) The DDC is one of the more quiet pumps around. Not all low flow pumps are quiet either.

Any other falsities and half truths you would like to spread?


To the OP : With your CPU block dont worry about running a GPU block in series with it. The Apogee is quite indifferent to flow rates, and performs almost as well at low flows as it does at higher flows.

 

[Justintoxicated]

I guess I will be stocking with my COPPER TDX (high flow or low flow I will jsut change out the nozzel). And my Acetal (Derlin) Maze 4 GPU block corrosion is one less thing to worry about when water cooling, which is also why I use Pentosin (or whatever the audi dealership sold me as equivilant)

 

[Halk]

1) Elegance is opinion not fact, there are plenty incredibly clean and elegant high flow, large bore systems. There are also plenty of nasty spaghetti mess low flow small bore systems.

2) Controls have nothing at all to do with the flow of a system or the size of the tubing.

3) The DDC is one of the more quiet pumps around. Not all low flow pumps are quiet either.

Any other falsities and half truths you would like to spread?


To the OP : With your CPU block dont worry about running a GPU block in series with it. The Apogee is quite indifferent to flow rates, and performs almost as well at low flows as it does at higher flows.

Oh don't talk nonsense. The larger a pump is, the more water it pushes the louder it will be, the more vibration it will cause. Just because there's good and bad pumps out there, that cannot possibly translate to higher volume pumps being quieter than a lower volume pumps.

Look up elegance in a dictionary. The applicable definitions I found were -
a. Restraint and grace of style.
b. Scientific exactness and precision.

So my small tubes, which are, what, 1/4 of the volume of the great big huge ones you can get, do (in the same ballpark) the same job. They're more precise, restrained, and exact. Oh look, they're elegant.

And I probably was unclear when I said "more easy to control". Low flow is easier to fit and route in the case. I wasn't meaning the AquaComputer controls - they can be used for high flow as well.

Fact is that low flow has it's place. If you want to get your components the absolute coldest you can, and a few degrees really does matter, then low flow probably isn't the way to go.

But when it does the same job, is easier to work with, is more elegant then really, it does have a place.

 

[Arcygenical]

Howabout we agree on dainty then?

If you wanna get all dictionary on us that is.

 

[Halk]

Howabout we agree on dainty then?

If you wanna get all dictionary on us that is.

Heh, that word fits too I'm going to paint flowers on on mine, and have a pink radiator!

 

[Ranker]

Let's get back on to the important stuff. Alu has no place in a well built water cooling loop. To do otherwise screws the consumer.

 

[Halk]

Let's get back on to the important stuff. Alu has no place in a well built water cooling loop. To do otherwise screws the consumer.

I don't agree. As I said you could say the same about water in a computer. As long as it's controlled then it's fine.

 

[Justintoxicated]

I don't agree. As I said you could say the same about water in a computer. As long as it's controlled then it's fine.

Water in a computer has it's advantages.

Aluminum in the loop does not.

 

[Digger]

Shoggy, a quick question. What is the thermal characteristics of Alumina Ceramic?Is there an additional performance hit with the aluminum block? I'm not flaming or being rude just curious. thanx

 

[madmat]

Shoggy, a quick question. What is the thermal characteristics of Alumina Ceramic?Is there an additional performance hit with the aluminum block? I'm not flaming or being rude just curious. thanx

There wouldn't be a performance hit since the aluminum is not the point of heat transference. The base of the blocks are copper.

 

[Digger]

my bad , thanx Madmat

 

[R1ckCa1n]

This is what happens when you mix copper and aluminum: http://www.xtremesystems.org/forums/showpost.php?p=2215997&postcount=1

Forget to mention this dumbshit didn't use an additive and the thread ended up getting locked?

Everyone just needs to get over it, when used RIGHT, alu in a loop is fine. When done right you have a better chance of a leak that those FUD pictures posted by Runker.

It is funny to watch some defend Swiftech though .... blah blah blah our technic is better blah blah blah

 

[Top Nurse]

No I'm not sorry. Swiftech uses military spec plating which is more stringent than the techniques used by any other manufacturer. It will never become an issue with corrosion. Even in the harshest of environments, the plating will not deteriorate to expose the aluminum. The aluminum will never come into contact with the water resulting in the possibility of corrosion.

No one involved with Aqua Computer, till this day, has ever commented on this subject that I know about. Everyone just ASSUMED that the only thing protecting their aluminum reservoirs and copper blocks was a hard anodizing. Turns out you bashers were all wrong one more time...

So how do YOU know that the mil spec process (mil spec don't mean shit) isn't or hasn't been used by other manufacturers? Seems to me that you are just spouting off more FUD on a subject that you have little direct knowledge of or are you just mouthing off again about what you heard somewhere else? And you were heading in such a good direction over the last few hours on the D-Tek/AC thread...

I ran down the link Shoggy provided and after looking at the equipment I think they are not plating their parts per se. I suspect that they are using a gas coating process as that is the only way to get alumina ceramic to bond properly to the metal. What you get out of this kind of process is an extremely dense layer of alumina ceramic that completely covers the part in question. IIRC, this process deposits the material right into the pores of the metal. Perhaps Shoggy might enlighten us on this if he knows.

So absolutely a big flat no. The use of Aluminum is malicious towards the end user by harming him and his investment. Those who produce such a product need to get their shit together; even if it's the almighty Swiftech for just one experiment in design. This goes doubly so for all of the other manufacturers out there that continue to cut corners and use aluminum. Did I buy a GTX? Yeap. Will I ever buy another Swiftech product again if they continue to use Aluminum? Nope. Even with the strictest standards, military specifications (which basically means you can run it for 2 decades and the plating will never wear (something that can't be said for various companies who use dubious methods))

Since it is so malicious why are you defending Swiftech? Hell you even went out and bought the aluminum and copper block they produced! BTW, why do you continue to spread FUD about manufacturers methods that you don't no jack about? And then defend your vaunted Swiftech when you got no clue about their methods either?

This is what happens when you mix copper and aluminum: http://www.xtremesystems.org/forums/showpost.php?p=2215997&postcount=1

You amaze me! Throw around a few spectacular pictures that are purported to be corrosion. The OP of the thread you linked to said "I used ONLY destilled water without any "admixture" to stop corrosion." Sounds almost like they did it on purpose just for a pic. BTW, how do you think they got that much "corrosion" to gunk up the water path? Looks to me like they threw that part in a bath and added a voltage till they grew enough stuff to take a pic.

Others in that same thread who sound like they know what they are talking about took you too task on your BS and you responded with that same old rant that you have been throwing around here as well. The last thread you were foaming at the mouth over there got locked.

I suggest any one who is interested to read the whole thread that was linked and you will see that even those who believe in extreme water cooling don't have a problem with it except for one other vocal guy like you that bought the Swiftech alu/copper block. .

The way you are wanking off on this subject reminds me of some lawyer trying to flame up a class action lawsuit of some type. Please stick to what you know and if you don't then go straight to the source instead of flaming good manufacturers that you know nothing about their products.

 

[Top Nurse]

No I'm not going to waste my time arguing something that I know as fact.

You who think otherwise are out to lunch.

I love it when people respond this way because it justs makes it so easy to show who is the tweaker in the crowd.

 

[Top Nurse]

TN, While I don't agree with the use of aluminum in newer swiftech blocks, they're plated to a millitary grade specification.

No anodization is millitary grade, for the purposes of corrosion protection.

Please quote the mil spec you are talking about or are you just repeating something that sounds catchy? My dad's home built aircraft has a lot of mil spec parts in it. My dad told me that there really wasn't any difference in the mil spec fasteners he bought from those in the local hardware store. The key issue was that if you crashed your plane, then you wanted to avoid liability and that was why he bought mil spec stuff.

P.S. My dad was a top of the line aerospace engineer involved in the Orbiter project so I think he knew what he was talking about.

 

[Top Nurse]

Nice chart! However, a lot of these guys won't believe it because it wasn't done in the USA and it isn't in English. For some strange reason people of a high flow philosophy seem to be quite ethnocentric in their water cooling practices. Must have something to do with high performance racing as evidenced by the "Edelbrock" look that Swiftech is putting out now.

I'm sorry but I disagree.
High flow systems are just a way of implementing a watercooling solution, not the objective itself.

You can see that the MCW-6000 reaches the same "performance" as you call it as the 1A-Cooling HV3 block, at higher flow with 3/8 tubing.

Generally speaking "high flow" systems are costly, and if you are "extreme" even more so.
Dual pumps (or a high head pump, say a Iwaki), dual loops, dual radiators, only to extract a single degree, to me it's not an option.

 

[Top Nurse]

Ceramics can chip whereas metal plating is considerably more robust when done properly.

That would depend entirely on how the material is applied to the metal. Some coatings are applied by a plating process and others are done by a gas deposition process.