most watercooling radiators can take 2 bar (almost 30 PSI) .. i'm not sure about automotive, but since car radiators are designed to work under pressure (because of liquids in excess of 100°C, hence the expansion vat they have) I guess that you would not have to worry about that all
Not ALL watercooling rads can take 2 bar and most blocks & fitting wouldn't like that very much either.
But unless you are planning to use a couple of Iwaki RD30
in series, you'ld be hard pressed (see what I did there?)
to get anything above 5 PSI, not with a normal D5
. You're well within specs here.
Of course if you plan to introduce some crappy glued PVC joints that fall apart when you look at it, well yeah, THEN you should worry mixed materials:
Acetal is not an issue.. copper + nickel + alu.. better make sure to add about 30% automotive to the distilled water
Choose an automotive additive with lotsa anti-corrosion properties, not just any "antifreeze". You don't have to worry about freezing here.build:
just visualise as each PC has its own standalone internal loop. The only difference, the hose which goes from your "last" block back to the pump inlet actually has a QDC sitting in the middle. When connected to itself, you can move the PC freely around, or even "plug-in" a normal 240/360 rad using the QDCs.
When connected to the BIG loop, that hose gets "split" and the QDCs connected to their counterparts which are attached which a (short) length of hose to the manifolds.manifolds
Either build something yourself , like naja did in his car rad build
OR use some "professional Central/Floor heating ones build redux:
A mere suggestion or two:
Most automotive rads come with several & different fittings/holes. At least two of them will be rather large, in most configurations the is one "big" one at the top and one "big" one at the bottom. Probably something like 1 or 1.2 or maybe even 2 inch dia. Use these
Get some (white) PVC (drainpipe) in the same diameter, it allows you to easy couple this pipe to the radiator using a few short ends of (rubber) automotive radiator hose coupling tying them in place with wormdrives.. just like you see it done under the hood of a car/truck
Mount the bottom of the automotive rad about 3 feet off the floor - depending if the bottom outlet is vertical or horizontal. In most cases its horizontal.
run the PVC pipe horizontal (and preferably level or with a slight 2° downward angle) through the wall to where it needs to go.. add the double, triple, quadruple manifold (if gluing, preferably 24 hrs in advance) This is now your feeder pipe FROM the rad.
Mount the second PVC pipe & manifolds about one foot higher as the top of the radiator - horizontally (and preferably level or with a slight 2° downward angle towards the RAD) This is now your return pipe TO the rad.
Before connecting up with the top of the rad (you might need an S bend for that); add in a PVC "T" pointing upwards and have one foot of the same pipe pointing upwards. This will become you fill & balancing pipe & reservoir.You rad is now ready for gravity feed operation.
Start filling up the radiator slowly as you would any. At some point most of the air will be out of the system and you might even start seeing some ware coming in through the return pipe. Time to put the radiator cap on and continue filling slowly through the vertical balancing pipe.
Continue till it (almost) overflows then its time to start the pumps (not the PCs, only the pumps, use a separate PSU to temporarily plug in the pump if needed or if more convenient)
Since there is more as enough air stil trapped in nooks & crannies, you'll get some spluttering till most of that air is replaced by water.
You may need to top up during this process.
Now let run for 24 hrs, sporadically checking for leaks or if the system needs to be topped up.
The vertical pipe now acts as a reservoir/ air trap. The liquid in this vertical pipe should NOT be overflowing, it should be no more as a few inched above the horizontal pipe. Maybe halfway up.
Now its time to convert this pipe to a balancer : just slide a rubber balloon (or even a condom) over the pipe. This is mainly to prevent dust or nasties from choosing to start a new life in the pipe. At the same time... if the liquid level rises because of thermal expansion.. there is enough room for it. And should the pressure rise a lot (which it wont, but IF), then this is where it goes without being lost (iow it keeps your loop pressurised). And when the level drops because PCs being idle on a cold night (thermal shrinking) there is enough reserve in the balancer to avoid air getting sucked into the loop.
Of course, this is only one possible setup, there ARE other ways of doing it, but this is frankly one of the easiest ways of doing.
PS: yes, Linus did it wrong
NOTE: For those that are going to argue that the D5 pump does have "no suction" and thus it wont work.. it doesn't matter.. we have + 1 gallon of liquid "pushing" at 1 G at the "suction" end. And since the D5 has ~10 feet "head", unless the car rad is 9 feet tall, there should be no issue on that "lifting" front either
Extra note: Need to do some calculations: because of the "gravity feed" nature of this setup, there might be a limit on the number of ports (PCs) on the manifold... but need more exact numbers for that ... ie total volume of water held in the Rad , but from the top of my head, the "weight" of one gallon of water above the bottom manifold should be enough for at least a 5 port manifold....
Especially when using "big-bore" pipe
Where is Archimedes when you need him?