question about radiator (computers) vs car engine radiators

[Ja.KooLit]

a quick question

How come most of the radiators used in computers does not have same construction compared to engine radiators used in cards?

From the radiators used in computers, if inlet and outlet are on same side, would it be that there is more tendency that hot water will just "pass" wherein, if the inlet and outlet is the opposite side of the fins, the hot water will be more cooled instead?

car radiators:

 

[FX-GMC]

a quick question

How come most of the radiators used in computers does not have same construction compared to engine radiators used in cards?

From the radiators used in computers, if inlet and outlet are on same side, would it be that there is more tendency that hot water will just "pass" wherein, if the inlet and outlet is the opposite side of the fins, the hot water will be more cooled instead?

car radiators:

This is an interesting question.

Makes for a cleaner install and reduces space requirement possibly?

 

[Ja.KooLit]

probably and possibly.

interestingly, none of the computer water cooling companies made it like that.

leads me again to speculation that by making a rad same design as the engine rads, you need a more stronger pump cause passing thru those fins needs some pressure to pass thru.

But, if a 360 or 240 rad or even 120 rads can effectively cool the water passing thru, we dont really need lots of rads and so on.

 

[erocker]

Many car radiators are single pass. Inlet on top, outlet on the bottom. The coolant passes through the radiator one pass before exiting. On many computer radiators, they are dual pass meaning the water goes in, circulates twice (down then up for example) and exits next to the inlet.

It's shown here pretty well: http://www.bryke.com/blog/single-pass-radiator-vs-double-pass-radiator-vs-triple-pass-radiator/

 

[FX-GMC]

Many car radiators are single pass. Inlet on top, outlet on the bottom. The coolant passes through the radiator one pass before exiting. On many computer radiators, they are dual pass meaning the water goes in, circulates twice (down then up for example) and exits next to the inlet.

It's shown here pretty well: http://www.bryke.com/blog/single-pass-radiator-vs-double-pass-radiator-vs-triple-pass-radiator/

You beat me to it. Was just about to post this (of course I imagine the one in the OP just goes down then up and out.):

 

[Jetster]

The way PC rads are its a longer run and more time to cool and less water. With a car its more volume shorter run. Car fans are a little louder lol

 

[lilhasselhoffer]

The engineer inside is killing me when I see some of these explanations. It isn't that there is a difference in location, it's a difference in thermal output that separates a car and a computer.

A radiator is designed to convey energy from one fluid to another. In the case of a car, you've got a very hot fluid inside the system (212 F or 100 C), and a much colder environment (up to 120 F or 49 C). The vast difference in temperature means even a modest flow of the cool fluid over the radiator will transfer a good deal of energy. As such, the radiator can have a very small quantity of fins, in relation to the fluid bearing tube. This is why car radiators generally have very separate inlets and outlet areas (I've seen plenty that output at the bottom and the top, it's a design choice made by the manufacturer), as convection is very good at transferring heat.

Computers are very different. A CPU might only put out a fraction of the heat of a combustion reaction (140 F or 60 C), and thus it has a very small difference with environmental temperature. This means that the radiator must have significantly greater surface area from which to cool the fluid. This is performed by vastly increasing the volume of fluid in the radiator, which means doubling or tripling the fluid carrier tube back upon itself. With this extra fluid carrier length, you could put the inlet and outlet anywhere, with little ill effect. What really matters is that the system is easy to install and use. As such, one side of the radiator generally has the inlet/outlet ports. A segregation is made in the inlet/outlet area, and the small difference between input fluid temp and outlet fluid temp means they can be very close with minimal heat transfer.

Boiling all of this down into two sentences, I can offer the following. The difference between car and computer radiators is the difference in fluid temperatures. Neither is required to have the input and output anywhere in relation to one another, so they are located based upon useability considerations.

 

[Ja.KooLit]

The engineer inside is killing me when I see some of these explanations. It isn't that there is a difference in location, it's a difference in thermal output that separates a car and a computer.

A radiator is designed to convey energy from one fluid to another. In the case of a car, you've got a very hot fluid inside the system (212 F or 100 C), and a much colder environment (up to 120 F or 49 C). The vast difference in temperature means even a modest flow of the cool fluid over the radiator will transfer a good deal of energy. As such, the radiator can have a very small quantity of fins, in relation to the fluid bearing tube. This is why car radiators generally have very separate inlets and outlet areas (I've seen plenty that output at the bottom and the top, it's a design choice made by the manufacturer), as convection is very good at transferring heat.

Computers are very different. A CPU might only put out a fraction of the heat of a combustion reaction (140 F or 60 C), and thus it has a very small difference with environmental temperature. This means that the radiator must have significantly greater surface area from which to cool the fluid. This is performed by vastly increasing the volume of fluid in the radiator, which means doubling or tripling the fluid carrier tube back upon itself. With this extra fluid carrier length, you could put the inlet and outlet anywhere, with little ill effect. What really matters is that the system is easy to install and use. As such, one side of the radiator generally has the inlet/outlet ports. A segregation is made in the inlet/outlet area, and the small difference between input fluid temp and outlet fluid temp means they can be very close with minimal heat transfer.

Boiling all of this down into two sentences, I can offer the following. The difference between car and computer radiators is the difference in fluid temperatures. Neither is required to have the input and output anywhere in relation to one another, so they are located based upon useability considerations.

True, but think about this, car engine radiators where not designed like this.

instead, rads on cars has a very small tubing with very fine fins installed on each side instead. With that kind of construction, I think it will be more efficient if the construction of such computer rads is same as car rads. Maybe, i am not engineer so I wouldnt really know

 

[lilhasselhoffer]

Let's look at this again. Car radiators generally have less fins than a computer radiator. I'm basing this conclusion off of people who might know, at hot rod magazine: http://www.hotrod.com/techarticles/engine/cooling_system_tech/

Breaking out an H100, I count at least 50% more, with a corrugated pattern between horizontal fins. You're looking at, roughly, a 75-100% increase in density of fins. Does this agree with my earlier statements; yep.

Less requisite fin area, for a greater difference in fluid temperatures is expected. It's also beneficial, as the fouling layer for a car will be much easier to clean with a hose than a computer radiator. I had to spend three months designing and testing a heat transfer system back in college. After about 5 iterations you discover that the easiest way to increase cooling is by adding fins. Adding passes through the fins is less effective than the fins, but it has a very real benefit. Finally, changing the fluids has the greatest cost, and least benefit. As air is always the fluid outside the radiator, you'd have to change the water to something better suited to heat transfer in the expected conditions. This kind of change would require a fundamental reworking of hardware utilized in the system, so it is effectively cost impossible.

As a side note, the most effective way to cool your system would be a phase change loop. The energy required to phase shift a material (liquid to gas in this case) is magnitudes of order higher than changing temperature within one phase. This still requires a radiator, but it artificially increases the temperature differential (via the compressor) of the fluid in the loop. In short; that AC unit outside your house, inside your car, the radiator attached to the engine, and you computer are all the exact same thing. They alter their designs to account for spatial restrictions, fluid types, and the most important factor of cost to manufacture. If all you are doing is asking why one is slightly different, you'd best serve your curiosity by breaking out 2-10 temperature probes (2 for fluids, 10 to see heat dissipation in radiator fins) and seeing why things are designed the way they are.

 

[Norton]

Design of rads all comes down to heat transfer requirements, footprint/space available, air/fluid flow, and materials used. There are multiple ways of doing this so cooling systems may look different while accomplishing the same thing.

Also, as lilhasselhoffer discussed, phase change is a great way to transfer heat away from one part of a system to another.