Printable View
:shakes:
Come across this thread at another site: http://forums.bit-tech.net/showthread.php?t=209025
Straight away tested this on my spare Feser Triple and guess what....
Its leaking through...
This was not a cheap piece of hardware, I guess more people should perform this test and feedback to wake those companies up.
hafiqb: i might be wrong but IIRC for some rads it was made such by design. As in that doesn't drop too much their cooling thermal efficiency, but reduces flow pressure drop/flow resistance, if i recall right reading about it at some old reviews or threads. Hmm, wasn't it about thermochill PA-s? Or about Fesers? IIRC conclusion was that done right it was more positive then bad thing.
hmm...need to check
Flow restriction versus thermal efficiency by bypass, could make sense, let's dig
Tried with my MCR320-QP some time ago and it leaks water the same way as most do in those videos.
Oh my... and to think I just had my loop torn apart last week. Two of my rads are on the list (magicool and thermochill).
Maybe they do this because some radiators dont have enough flow channels in the radiator to match the amount of water coming into the radiator, so they add a small bypass to keep the pressure buildup low? That's my wild guess.
Vinas && others: Don't take it as some megaproblem. I'm guessing that original whistleblowers just didn't have a clue about bypass design, it's pros and cons and passed it all as failures of manufacturing and QC process as they knew better [tm] then manufacturer how rad should be designed. Mass of LC users have such seemingly "faulty" rads, who cares? They still show particular flow restriction and cooling efficiency numbers. Who cares how they manage to reach it? These "tests" showing "faults", thermal images showing water inner flow distribution .. so what? One cools liquid in LC loop by rad itself, not by images. If rad does the job dissipating heat as it should, it wouldn't suddenly cool worse because someone says that he thinks that rad shouldn't be this way.
Simple question - do rads made "right way" cool THAT much better then "faulty" ones? I doubt it but it's all that matters at the end.
Very very old news around here. :yepp:
http://www.xtremesystems.org/forums/...d.php?t=183897
The ironic thing is, they (bit-tech) use a rad from a company I don't like very much and it was one of that companies known shills that posted the original article. Maybe that's how they knew to look for it, from their own design. :rofl: It's nice how some other sites regurgitate old info like this as if it were some sort of earth shattering revelation.
*face palm*
im sorry.. this was so hashed out a long time ago, i cant believe someone brought it up again.
If it was as bad as you say it is, the PA would not hold onto the king title for so long, and still be considered one of the best rads out there.
It still better then the MCR, probably scales better then the XSPC, and Feser because of bulk size and mass.
So rad bleedthough from inlet and outlet is moot, and in some cases on the rad design has shown to be an improvement.
:shakes:
Proper test for checking chamber leaks...
Place a powered LED in one side of the tank, if you see the light coming through, you have a hole between the chambers (and should RMA, well a pinhole is no big deal and won't mean squat performance wise). If you don't see light... loop it.
Flow bypasses have been by design (i.e., intentional) for a while now. With some radiator cores, performance doesn't scale much with flow. Because of that, they add a bypass lowering overall restriction (at the expense of lowering flowrate through the core).
The 'downside' of a bypass is that air can get trapped in the initial bleeding process, other than that, I'm not sure why a bypass is bad (if the core can handle the lower flowrates without sacrificing performance). A bypass doesn't = cheap.
How a radiator performs should be part of the buying decision, not the internal design decisions, IMO. It's like demanding you won't buy a radiator with an FPI of greater than 8.2. It's just a design spec :shrug:
If there are inconsistencies within a product line, that's another thing...