[Kaitain]

Originally posted by Claudius and Random Nonsense
[b]

I think actually both you and Random are focusing on single effects rather than the system-wide effects of both. In fact both of you are correct in many ways...

bigger wider channels = less surface area, and also lower velocity

so actually the rate of heat transfer into the water is slower. Although the residence time of the water is greater, if it has slowed down sufficiently for the flow pattern to be laminar, then you'll get orders of magnitude less heat into the water than with the narrow-bore channels.

I would think that the water in the block with the smaller channels would get warmer.

I disagree. Heat moves into the water faster. The water leaves the block sooner. The result is negligible increase in water temperature (remember it takes a lot of Joules to heat a gramme of water). The difference between the exit temperatures of blocks is unlikely to be measurable within reasonable accuracy.

Since there is more resistance, fewer gph is going to pass through the small-channeled block than the large-channeled one.

Very much depends on the pump. Centrifugal pumps are very forgiving of back-pressures, producing maximum flow-rate up to almost their full design spec. Provided you don't exceed that, developing a back pressure is NOT going to reduce the number of gph through the system.

Also, since more water is in contact with the small-channeled block, the water will heat up faster.

Which is actually what you want!

If water is be pumped too slowly through smaller channels it will reach its heat capacity and won't remove heat from the end of the channel, lowering the efficiency of the system. In this case you would want larger channels so that there is more water to heat up.

The "heat capacity" of a system like this is only reached when it boils. You really don't want that, now. If the water is pumped too slowly through a channel such that it's flow pattern ceases to be turbulent, you'll get crap heat transfer. Doesn't matter whether the channels are large or small.

You don't want water moving too fast through large channels though because the water in the center of the channel will not heat up and will prove useless.

That's actually totally wrong. You can NEVER flow the water too fast. You can reach a point of diminishing returns, but there's no threshold where the increase in speed disimproves the system.

Depending on how your pump moves water through the block, there must be an ideal channel size/surface area which proves to be most efficient.

YES!!!!

Just going with a block with smaller channels and more surface area may not be a good idea if your pump can't force water through it fast enough.

NO!!!! It's easier to force water "fast enough" through small channels. All you're after is turbulent flow, pref with Re > 10,000.

I have written a simple way of calculating pressure drops in tubes in this very topic (do a site search for "kplonk" and "bends" and you should have it). You'll find that, yeah, the largest pressure drops are due to the inlet and outlet points in the system. If you follow through this and the relevant calculations on heat transfer rate (here's another guide: http://www.wlv.com/products/databook/ch2_4.pdf ) you'll be able to determine the trade-off between the flow velocity (that's in ft/s, gal/h is a flow rate) and pressure drop that will give you the most efficient cooling, at a rate your pump can handle.