Quote:
|
Originally Posted by Silent Bob
Cathar:
I have read a lot about your experimentations with watercooling and the major threads concerning the development of your waterblocks(which i consider the world's finest), both in OCAU and in Pro/Forums. You always try to optimize the performance of the block in the widest range of flowrates, and the "Storm" is yet another step in that direction.
But you always place the "flowrate vs performance sweet spot" very high in that range. Why haven´t you used your own simulation software to develop a very low flow block other than a micro-channel kind of block?
|
Hmmm. I would argue that the blocks that I make don't exactly need a super-high flow rate to work well. At 0.75gpm and above, Phaestus's results show that the Cascade is at the front of the pack. I don't see things so much in flow-rates as in pressure-drops though. At 0.75gpm the Cascade has a pressure drop of around 0.40mH2O. That's pretty low. That's something that the Eheim 1046 could do in a full setup so long as it is not restricted overly by the rest of the loop (such as would happen with thin tubing and restrictive radiators).
I am still somewhat puzzled as to why there is the persistent reputation that the blocks I make demand a strong pump to provide leading performance. They don't and never have. However, and I do admit, that flow rates much lower than 0.5gpm aren't ideal, but given that no pump I could find here in Australia pumps such low flow rates in an otherwise unrestrictive system, I really didn't see the need to bother optimising for such low flow rates.
That, more than anything, typifies my approach. Even the weakest pumps I could find still gave excellent performance (comparitively) so long as you don't choke the pump, so "Why choke the pump?".
That's probably why I have a problem with the WCP testbed and the "German" approach. It just seems to me that the system is unnecessarily choking the pump, and for what end, I cannot fathom. It creates an artificially low flow rate problem to solve, when really the problem should not exist at all so long as the pump is not being choked down by really small ID tubing and highly flow restrictive radiators.
It's like trying to feed a car engine's fuel demands with a good pump (the Eheim 1046 is still an adequate pump in my books - even though it is weak), but then introducing so many restrictions that it cannot possibly feed the engine with enough fuel.
So to me, what the "Germans" have done is decide to optimise the engine around the amount of fuel that is flowing through the restricted fuel line, instead of simply (and quite obviously I would have thought) removing the restriction.
The WCP testbed, as I said before, is both the best and worst thing to happen to German waterblock development. Sure, its stature as
the premier European testbed has driven waterblock developments that allow for some very impressive waterblock performance at very low flow rates. Don't get me wrong, I do admire what the Germans have achieved given the artificial limitations of the WCP testbed. In that way it is good. It is bad because it myopically limits waterblock development to a single point of view. It is restricted. It doesn't show the best of what's possible with an Eheim 1046 (or any other pump for that matter). It doesn't provide sufficient pressure drop across the block for good jet impingement activity to take a leading role in a block's design.
As long as the WCP testbed holds its pre-eminence within Europe as the be-all/end-all of waterblock rankings, with its one-eyed view of what a waterblock should be fed with in terms of flow rate and pressure-drop, then there will always be the bitter division and confusion between the "Germans" and the "US" and why each side thinks the other has it wrong.
I design the blocks with the assumption that the pumps aren't being choked. I'm not really interested in playing the WCP game of artificially choking the fuel line and optimising the engine to suit.
Do I believe that 1/2" is large and ugly? Actually I do. I believe that 3/8" ID / 1/2" OD is really quite an acceptable solution for most blocks/pumps. However, there are those who use pumps that push flow rates that demand the use of 1/2" ID tubing, and rather than optimise for the lowest common denominator, I'd rather optimise for the "open class" crowd, and let people make up their own minds on whether or not they want to needlessly choke their pumps down.
[Edit: It is good to see people like Pug offering less needlessly restrictive systems though - I really feel it's about time. On a more open-minded front, I would really like to see more 10mm ID systems that help to dispel some of the myths about "needing" 1/2" ID tubing to perform well. I was just arguing with Pug's selection of hardware as not exactly fitting into the mold of the "traditional" German setup. Aside from that, I believe it's really good that he's sending a less restricted setup that more follows the "US" view. I just believe that he could've sent the pump as the AP900 only with 10mm ID tubing, rather than send the pump in AP1500 mode with 8mm ID tubing to force the flow through.]