Need to understand what is wrong with Aerocool

[jordan12]

Guys,

I have read a few places, even in the sticky, where Aerocool is on the Do not buy list.

Why is that exactly?

 

[Ice Czar]

for starters its not an "official" list
that would require qualative and definative testing of each supply
information thats simply lacking

thus its computerpro3's assembly of ancedotal and empirical evidence
I dont have any hard data on the aerocools
which is itself a concern, they arent farming them out for competent review
another is that they supply just rudimentary spec
are an obvious rebrander and cater to the bling bling crowd

none are "good" signs
Id guess you should expect basic compliance
and Id derate it a full third on each of the rails and total wattage for temperature
(Its likely at "industry norm" read intel cert temp of 21C)

they arent highly considered but I have no horror stories or sucess stories to guide you
generally speaking (and as you might see in a sig or two)
"this isnt a court of law, supplies are considered guilty until proven innocent"
a prudent course for the one device that can fry all the rest

 

[ashmedai]

Summary: Because it sucks. Sorry there's no detailed test results for it, but if a PSU is bad enough that we think it sucks it's not worth the risk of frying all our pretty electronics over.

 

[Ronbo]

Send one to me, I'll dust off the oscilliscope, rewire a variac and run some real tests on it. I need a good excuse to put my workbench back up anyway anyone like purple CRT Traces cause my ancient HP O-scope has purple phosphor

 

[Ice Czar]

test one of these instead and let us know
http://www.directron.com/hp585d.html

 

[SJetski71]

test one of these instead and let us know
http://www.directron.com/hp585d.html

I'm especially curious to see that one professionally tested

As opposed to my non-professional test: http://www.hardforum.com/showthread.php?t=891640

 

[Ice Czar]

Id like jonnyGURU, Lee, Mike or Oleg to test that
point being any static crossload testing being of limited value
thus really requiring more equipment than the average joe would have
but I saw your test and its interesting
would be nice if you worked up its probable load by rails
and a guess at total watts it was drawing max (with temp)
along with the actual config

OH and Ronbo, I like purple, a well known flaw in old farts
any AC Ripple (DMPARD) info would be most welcome
on any supply thats not fully documented and even those that are

 

[Oklahoma Wolf]

Last time I saw the innards of the Aerocool 550w, it was a Topower inside. Probably one of the lower end models though.

 

[Ronbo]

Ice, for 22 bucks I'll definitely get one and test it, I'm gonna need some more PSU's for my folding farm, and I've got enough stuff laying around the workbench from my on again off again car stereo amp PSU designing / building to make a decent PSU tester, even crossloading and possibly PC control and monitoring. I swear one day I'll finish designing/building that +/- 36v 50 A SMPS to power a pair of leach amps in my explorer, really I will. (please don't hold your breath till then though )

 

[Ice Czar]

http://www.xbitlabs.com/articles/other/display/psu-methodology.html

specifically > http://www.xbitlabs.com/articles/other/display/psu-methodology_13.html
crossload charts
man I want crossload charts in the worse way
Oleg has a damn nice proceedure there, but very few actual supplies in the database
and it can make a huge difference in real life (and is easily seen)

example > http://hardforum.com/showthread.php?t=889008
(heavily weighted +5V system) goin by specs I inadvertantly recommend a Powerstream
well the one Im thinking of is great but not all of them, jonnyGURU points that out to me in a different thread and bam

bakward compatible to ATX12V V1.2
red crossload profile below
(actual ATX12V 1.3 hybrid)

bakward compatible to ATX12V V1.3
green crossload profile below
(actual ATX12V 1.3 hybrid)

Charts & Graphs by Xbit \ Oleg Artamonov

(assumed similar crossload profile for the higher wattage class)