Pumping faster won't cool it faster, especially since you're recirculating. At some point you may be pumping water faster than the heat can flow from the wort to the coolant. That speed depends on chiller length as much as flow rate.
Seriously question this. I'm currently a senior chemical engineering major, and have recently had mass, heat, and momentum transfer, as well as unit ops, and am currently doing labs applying and analyzing the design aspects of several pieces of chemical engineering equipment, including heat exchangers. (disclaimer: senior, havent graduated YET, so if I'm wrong, please correct me, I'd love to know)
A wort chiller is a simple liquid-liquid heat exchanger relying on convective heat transfer. Convective heat transfer is much more efficient than conductive heat transfer, which is why when it comes to IC's people recommend either whirlpooling/stirring the wort, or moving around the chiller inside the kettle. The more turbulence you have in the liquid, the more convection, and thus, the more heat transfer.
Looking at the flow of cooling water within the tube, more flow means a higher Reynold's number (a dimensionless number defined as density * velocity of the fluid * characteristic length, in this case the diameter of the tube, all divided by the viscosity of the fluid). A higher Reynold's number literally means more turbulence, as anything under Re = 2300 is considered laminar, over 4000 is considered turbulent. The higher the number, the more turbulence. So higher flow = more turbulence = more mixing = more convective heat transfer.
You can also look at it this way. If your cooling water is coming out extremely hot, a certain length of the chiller before the outlet is going to be extremely hot as well. It's actually going to be a gradient where the inlet is equal to the cold water temp coming in, and the outlet is equal to the hot water temp coming out. Everything in between is going to be somewhere between the two temperatures.
Now we also know that heat transfer not only relies on the amount of turbulence in the fluid, but also the temperature gradient (i.e. 33F cooling water works a helluva lot faster than 70F water). What this means is that if your cooling water is coming out at the same temp as your wort, heat transfer on the surface of the chiller for a certain distance back from the outlet is next to none. You cant transfer heat from a hot thing to another hot thing. On the other hand, if you had the flow up so damn high that its coming out still cool, the entire surface of the chiller for our intents can be approximated as being at the same cooler temperature, and thus there will be a larger temperature gradient for a larger surface area, and thus more heat transfer. So while it feels like its coming out cooler, it is impossible for it to come out at the same temperature as what it was when it went in (as long as the wort is hotter), and even though it doesn't feel like its moving much heat, it will actually be moving more heat. There will just be less heat transferred for a given volume of water, but you're moving much more water.
Yes, you still have the same number of BTU's for cooling, and yes, the temperature of the bulk cooling water tank will still probably come up to the same temperature, but it's going to happen faster.
Now if we're talking water efficiency in terms of gallons used, clearly you don't want to pump it as fast as you possibly can. But in a recirculation, it doesn't matter. It's just more electricity cost in running the pump faster, but then again, you'll be running it a shorter period of time, so I don't know what the difference will be.
Also, you WOULD want the water to be coming out scalding hot if you happen to run a major brewery, where your cooling water is then piped over to your mash tun. This will make cooling take longer, but it will save a lot of energy, as you won't have to heat it up for your next mash.
Hope that all makes sense, and again, if I'm missing something, please let me know. Oh, and sorry about the wall of text. This has to be the longest "well actually..." in HBT history.
tl;dr: if your goal is to chill as fast as possible, flowing more cooling water in the chiller is better.
.
