Where do we discuss the ZM600-HP review?

[Howard]

Anyway, I was reading the Zalman heat pipe PSU review and was just about to make a fool of myself by posting that they should have flipped the heat pipe around since the fluid wants to go down.

Then I realized that the PSU would be upside down in a case.

That is all.

[CAD4466HK]

Seems you found the spot:p

[GalvanizedYankee]

As far as heat pipe operation is concerned, I think it would not matter if the fan was on top or bottom orientation as long as it got unresticted air.

Post here more often Howard :)

[Howard]

As far as heat pipe operation is concerned, I think it would not matter if the fan was on top or bottom orientation as long as it got unresticted air.

Post here more often Howard :)
No, not the fan, I'm talking about how the heat pipe is routed.

See, in the pics, the heat pipe goes through the top of the aluminum block and also through the top of the radiator. At that end, it bends and goes into the radiator again, but at the bottom.

Heat pipe theory would dictate that for optimal performance, the lowest point should be where the temperature is highest and vice versa. This does NOT happen if you run the PSU in the orientation that it's photographed in, but it works just fine if you install it in a case.

Still, I do not like the fin spacing of the radiator. It looks way too tight for any appreciable airflow. It's a good start to effective passively-cooled PSUs, though.

[GalvanizedYankee]

You are so right Howard...

[Bbq]

if it's a wickless, it only works one way. if it's a partial wick, it only works one way. If, like almost ALL of them, it's a full wick, it'll work in either direction.

[GalvanizedYankee]

According to ThermalRight, even the fully wicked work at thier best with the heat source lower than the finned area.
Yes, I know heat pipes are used in gravity free enviroments.

Howard, the two vertical sinks funnel the air to the fins of the heat pipe. I don't think the fins need wider spacing. ;)
You should be glad this Epsilon based unit stays within ripple spec at full load.

[Howard]

Well, I can kind of see the funnel effect, but it doesn't work. The aluminum sinks don't go all the way up to the radiator, and the plates don't go to the edges of the radiator (and forget about alignment). To deal with the air going around the radiator, the radiator would have to be a lot bigger.

But, since the PSU works, I guess the radiator provides sufficient cooling. It irks me to see such crudity, though. They might even have reduced fan speed if the fin spacing was wider.

EDIT: Yeah, the ripple is much lower than the Epsilon's. I wonder what they changed. Hopefully it isn't just a matter of increasing capacitance.

And what the heck is the hot glue for?

[Oklahoma Wolf]

Yeah, the ripple is much lower than the Epsilon's. I wonder what they changed.

The output pi filters changed, which is the only thing they could have changed to affect ripple. I'm almost certain it's due almost entirely to replacing one of the OST 2200uf output caps on the 12v with four parallel Teapo 470uf caps instead. The four small ones would have better ripple suppresion than the one big one. So, instead of two parallel 2200uf caps on the 12v output, there is now five in parallel - one 2200 and four 470.

The capacitance on the other rails was apparently increased as well.

[Bbq]

Hmm, I was wondering, how is the ripple on the OCZ GXS 700 when it is loaded up to 600w?

It might just be that the zally is just a 700w with some mods, and labeled as 600?

[Howard]

Hmm, I was wondering, how is the ripple on the OCZ GXS 700 when it is loaded up to 600w?

It might just be that the zally is just a 700w with some mods, and labeled as 600?
http://www.jonnyguru.com/review_details.php?id=35