That is the term right? 1 inlet - 2 outlet = pressure drop, sOoooo.. why do people put y splitters on afterwards? Surely that stops any would be drop, unlike connecting both outlets straight to a res? Or does it still work, but less so? They need high input flow right too? So would a pump -- gpu -- rbx == res kill the flow a bit too much u recon?
Not quite right. With blocks such as the White water and the RBX, just about the only reason that they have two outlets is because they have a central inlet over the core. The least restrictive method of getting the water through the block is to have an outlet on either side of the central inlet. It would be possible to vahe a channel in the block moving round the fins, joining up at a single outlet but this would be more restrictive due to the bends made. "pressure drop" is a term referring to, as the name suggests, the pressure drop across the restriction. Think of it this way. Much like thermal energy flows down a temperature gradient, the liquid forced by a pump does so down the pressure gradient. The more restrictive the block, the greater the "pressure drop" required to induce the same flow rate of coolant. It is for this reason that watercooling enthusiasts are "starting" to take notice of head (or pressure) ratings of pumps as well as the free flow rating. The free flow rating is rather useless on it's own as it tells you very little about how the pump behaves when a restriction is placed between the outlet and the inlet. The max free flow rating and the max head rating is better, giving you a good indication of the pumps performance, while a full p-q curve is the best option, giving you the pumps characteristics from zero flow at max pressure to max flow at zero pressure. Take the white water for example. This block requires a large pressure drop to generate high flow simply because it is very restrictive. It is restrictive because the fins are packed very close together, not because of the outlet. The dual outlet, as I have stated is there to ensure that the water can get out of the block with as little restriction as possible while still having even flow to the left and the right of the central inlet. Why do you want as little restriction as possible when exiting the block? First, think what is causing the restriction? Small channels and bends for the most part. These things generate turbulence, which saps energy from the coolant flow. This is good in and amongst the fins, as turbulence breaks down the barrier layer adjacent to the copper of the fins, thus decreasing the thermal resistance associated with getting heat from the block into the water. But once this is done, you want as little restriction as possible. Why, because, as I said, this saps the energy from the coolant flow, and this is wasteful if the resulting turbulence is not actually improving the performance of the block. By not wasting the kinetic energy of the coolant in this way, you will actually get a greater flow rate through the block, improving performance. Think of it this way. The pump gives the coolant kinetic energy. It is important to use this energy efficiently where it is needed to create turbulence in the heat exchanger parts of the blocks/rads, and not wasting it where it is not needed, ie tubing/fittings etc. Hence the argument for using large bore tubing and fittings. That is not to say that small bore systems are wrong. In this case, the block internals tend to be more restrictive so that the amount of energy used in the block is still large when compared to the total amount provided by the pump. What is not good, is using small diameter tubing with unrestrictive blocks, since a large proportion of the energy provided by the pump is "wasted" in the tubing and fittings. If larger tubes and fittings were used in this scenario, then more of the pumps energy would be put to good use where it is needed, OR a smaller pump, which adds less heat to the coolant, could be used in it's place, while still providing the same flow through the blocks/rads. Does that make sense. If not, ask away. 8-ball
What i've always been curious about with dual outlet blocks is, in a 3 block system, CPU, NB, GPU, would connecting one of the outlets to the NB block, and one of the outlets to GPU block have any serious side affects on performance? I imagine it would make the tubing a whole lot easier to route and look a lot better without having a big Y-splitter straight after the block. Good idea or not?
interesting idea olv, I'd be fairly interested in that too, although that'd mean the splitter would have to be placed after the northbridge and gpu, it could possibly be done out of sight though which makes for a cleaner setup.
I was thinking it could just be used with one of those dual inlet bay resevoirs. Depends on the order of the loop i guess.
A quote from Cathar himself, when I asked him about this a while back. Mind you, this was in reference to using Danger Den Northbridge and GPU blocks. In other words, you can do it. The closer the resistance of the two paths, the better, as this will lead to balanced flow within the block. 8-ball
