Maze 3-1 or Swiftech MCW462-UHT for use with tec?

[Wayne Harrow]

Which of the two blocks will give me the best performance the maze 3 would have a 220w tec and the swiftech would be using the supplied 226w tec. So what one will give the best performance?

[pHaestus]

I am not sure either one will be a substantial upgrade from the Maze2-1 you already are running. My instinct would be to just make sure your pump and rad are giving you the most out of the Maze2 you are using now.

[Wayne Harrow]

The thing is the maze 2's threads all mucked up for the cold plate bolts so i cannot clamp it properly noting to do with my pump and rad there well up to the job i usally have a watertemp of about 30 degrees and im using a eheim 1250 pump.

[pHaestus]

ah. Well I haven't used the Maze3, but the Swiftech MCW462U is a very well-built block, and their TEC addition seems very solidly put together. I would probably go with that, and I would additionally suggest that you skip the UH and just get a piece of copper pipe to put into the -U block as a fitting. You can use 1/2" ID tubing that way, and you maintain the high velocity at the inlet (over the die) that helps heat transfer.

[decodeddiesel]

another vote for the swifty, also the pre-cut gaskets would make your condensation proofing much easier, take my word for it.

[Les]

Dunno, but my guess would be that the Maze3 may potentially give lower temps .
Suspect that the labrynth design of the Maze3 will give more uniform and better overall cooling of the Peltier's hot-side.

[bigben2k]

My vote's for the swifty, but only by the narrowest of margins. The Maze3 has fat corners, the swifty doesn't. In other words, the Maze is better designed for a CPU than a Pelt (ever so slightly though).

[Les]

Originally posted by bigben2k
My vote's for the swifty, but only by the narrowest of margins. The Maze3 has fat corners, the swifty doesn't. In other words, the Maze is better designed for a CPU than a Pelt (ever so slightly though).

My assessment differs.
The "fat corners" are are outside the 50 x50mm Peltier area and possibly irrelevant. .
Unfortunately there is no reliable data for cooling a 50x50mm heat source.(Billa's work is for a 10x10mm source) so predictions depend on the heat-spreading versus convection arguments..
Cooling a 50x50mm Peltier possibly requires zero heat-spreading by the baseplate..Definitely considerably less than a 10x10mm (ish) CPU.
Exclusion of spreading resistance allows much simpler calculations .For example reduction of thickness of a 50x50mm from 8mm to 3mm will reduce the oveall block resistance by 0.005c/w or 1.5c for a Peltier producing 300w.The thinner base(2-3mm)* of the Maze3 may offer this improvement the thicker(8mm) Swiftech.

The convection situation is possibly the question of the Switech's ability to "carry its direct impingement flow instabilility through to the perimeter of the block"verus the Maze3's "channel size determined flow charactaristics ". If the turbulence at the perimeter of the Swiftech decreased to that characteristic of the channel dimensions (10x50mm ?) then the flow could even become laminar.(giving 1/10th the convection coeff) below ~ 1gpm.


* http://forum.oc-forums.com/vb/showth...ighlight=maze3

[webmedic]

I agree with the swift also with the 1/2 inch pipe. Here is why.

You looking at 1/2 barb on the left, 1/2 pipe in the center, and 5/8 inch barb on the right. I modded this top from the original with the push in fittings for samll tubing. It was majorly lacking in its previuos state.

[Wayne Harrow]

The swiftech MCW-462-UHT allready has 1/2" barbs it doents use them quick connect fittings thats why it has the H so would i still have to do what you are saying?

[webmedic]

I already new that. The point is how much more flow the 1/2 pipe has. The 1/2 pipe will fit nicely in the new top and you don't have to mod it like I did with my top. You are free to use whatever barbs you like. I was just showing the difference in flow you will get.

[Wayne Harrow]

Ok cool thanks alot for the help, looks like swiftech it is