Bang for the buck PSU for +3500

[mybrainisawaffl]

Just built a new system:

+3500 skt 939 130nm A64
512mb PC3200 Patriot XBL
MSI K8N Neo2 plat
6600gt AGP

My case came with a POS Rosewill 400wt PSU. I can't pull an OC better than 11x210 with it. Needless to say, I need a PSU that will let me get to 11x250 or so, without breaking the bank. So, I ask, is it possible to get a PSU round 40-50$ with which I can pull a decent OC?

 

[MrCooper]

Just built a new system:

+3500 skt 939 130nm A64
512mb PC3200 Patriot XBL
MSI K8N Neo2 plat
6600gt AGP

My case came with a POS Rosewill 400wt PSU. I can't pull an OC better than 11x210 with it. Needless to say, I need a PSU that will let me get to 11x250 or so, without breaking the bank. So, I ask, is it possible to get a PSU round 40-50$ with which I can pull a decent OC?

You can pull a 530W Fortron for $76 and if you're patient enough to save a little more you can get the 550W version which will give you enough headroom ti'll your next upgrade...
Fortron 550W

 

[Oklahoma Wolf]

350w, dual 12v rail: http://www.ewiz.com/detail.php?name=PS-FS35TNB

This is about the lowest watt unit I recommend for overclocking. I'd save up an extra $9 for this one myself: http://www.directron.com/fsp40060thnr.html

 

[MrCooper]

Or this TTGI 550W $69. Hopefully the color doesnt bother you, As it's a mixture of red/pink. But a lot of juice for maximum clocking.

 

[d94]

i have a TT 480W psu thats your for $40 shipped
i had a fully loaded 2.6c@ 3.4 on it 24/7 stable rails

 

[Enigma422]

Fortron BlueStorm 500W. IMO best bang for the buck of any PSU out there.

 

[Binman_UK]

Fortron BlueStorm 500w. Nice and quiet and delives the goods when you need it most.

Binny.

 

[mybrainisawaffl]

How much would the Fortron BlueStorm 500w cost me, and where could I get it?

 

[AMD-Me]

Will the Bluestorm 400w work as well? The 12v rails are the same on both, Im not running any crazy raid array, just the basics 1 HDD, 2 CD, 1 CPU, 1 video card

 

[larva]

I noticed the 400W Blue Storm at directron for $59, and considering that my TP430 has always pumped out prodigous heat, the fact that I recently noticed a bulging cap in it, and that I am one seriously cheap bastid, I was all set to buy one. I before punching the submit button I googled the model number and found this review:

http://www.insanetek.com/index.php?page=fortron400bs

The reviewer did a nice job and has great interior pics of the supply. There isn't anything inside. I found pics of the AX500-A, and all those missing components aren't missing. He said it struggled under the load of a 2.4c at 3.48GHz. Scratch that idea. BS/AX500-A=great, BS400=total BS. Very un-Fortron like.

I'd like to retire my TP430 before the bulging cap in it explodes, I'm not sure whether I can bring myself to spend $90. Don't get me wrong, the AX500-A is all it should be and without question the stellar PS value at the $90 mark, but $59 is more the kind of money I was wanting to spend. Where's a newegg refurb when you need one.

I also really like the looks of the new Antec TPII series. They have 120mm fans, the prerequisite ATX2.0 dual 12V rails and 24 pin ATX connecter, and adequate ratings. They cost from $89 for the TPII430 to $109 for the TPII550. While a few bucks more than the AX500-A, I have grown to love the fan speed controller in my TP430 very much. It runs my two Sanyo 120x38mm case fans at 1385rpm at idle, and ~1900rpm at max load--about perfect. It also tames the wild 45mm Delta screamer on my NB1c, achieving 3750rpm at idle, and ~5000rpm at max boogie. Again, just about perfect.

 

[yeha]

that empty space is filled in the ax500 with something, but what? ax500 internals, bs400 internals, otherwise identical. what's that big fat thing in the lower left of the 500w model? it can't be something critical or it wouldn't be completely absent in the 400w model, but its size implies some significance.. i wish i was better with electronics.

oh and if you don't like sparse pcb's, then fortron's upcoming green psus won't be your cup of tea - that's the 400w model from here, the efficiency is staggering but the stock fan was quite bad, unfortunately.

 

[larva]

That's a big nasty steel-laminate transformer. It is responsible for filtering ripple from the output of the switching power transformer beside it. This reduction of ripple would both increase the average power of the signal, directly increasing output, and is also likely responsible for the innordinately high efficiency level the AX500-A demonstrates.

This type of transformer would cost roughly 5-10 dollars, and could increase sustainable 12V output by as much as 20%. It would be interesing to know if the ferrite core transformer (next to the laminate-core) in the two units are identical. It is conceivable that the addition of this feature takes what is in the BS400 an innadequate 12V section and elevates its output into the useful realm. And it underscores just how good the engineering is that goes into Fortron supplies.

I'm not concerned with the "sparseness", but rather the output. If you can give me something like the AX500-A's performance with two caps, a coil, and a resistor or two I'm game. But a supply that can't drive a 3.5GHz Northwood with ease won't play to many around here.

ZZF has the TPII-480 for $99 including 2-day FedEX. Mine should be here shortly. Between lack of sales tax and the free shipping it is cheaper into my hands than the AX500-A Fortron, and saves me the trouble of either hacking into the Fortron to access the fan controller's output (and go through another painful fan selection process), or extracting the fan controller circuit from my old TP.

 

[yeha]

looks like you'll have the chance to write the first review of the tp-ii series, at least i can't find any. is there any way you can test efficiency when you get it up and running?

i'm curious about how you came upon using the psu's fan control circuitry to drive the case fans too i don't think i've heard of that being done before. i'm guessing you picked a specific molex line and replaced its 12v source, so all fans run off it for convenience - how much current can you draw that way?

 

[larva]

I don't know if I'll be able to give any more than anectdotal evidence of the supply's properties for a few weeks. This is the last week of the semester, crunch time for sure. I'll be graduating May 7, and afterwards I may well get together with my father (a retired big-time TV power supply engineer) and produce some representative data.

The fan controller idea is Antec's. The "fan-only" molex's on the original True Power series is driven from the same controller that drives the supplies' iffy internal 80 and 92mm Top Motor fans. While I dismissed the value of this feature originally, it has proven to have real merit.

Previously I used Mechatronics 120mm case fans running at 12V. These are 25mm thick units that spin at 1500rpm, move 55cfm, and are (realisically) rated for 25dBA. The fan controller circuit on the Antecs really pulls the fan voltage down, and as such requires powerful fans for the realized fan speeds to be in a useful range.

I also find that the stock fans on the TP's spin too slowly for fully loaded operation with this degree of voltage reduction. The 92mm Top Motor OEM fan in my TP430 started making spurious noises not long after I started using it, and since the air leaving the supply's exit would shrink heat-shrink tubing replacing it with something stiffer would seem a natural step.

After trying every fan in the house (50 or so), the best combination turned out to be an 80mm NMB fan from a FSP350-BN (.35A) and a old Pabst 92mm I had laying about (.5A). I also power, from the same circuit via the original "fan only" connectors, two Sanyo 120x38mm fans (.52A ea), and the 45mm Delta (not much current). So in all, I've got about 2A comming off the TP430's fan controller, and it does not seem strained.

It is not safe to assume that the fan controller in, say, an AX500-A is capable of driving this much load. As the fan-only connector is an Antec exclusive, their fan controller may be capable of more current. I haven't analyzed the circutry, but will at some point.

What really surprised me was that using the power supply temperature as the basis for case fan regulation works very well--better than any other temperature dependant scheme I've seen. The Sanyo 120s are quieter at 1385rpm than the existing Mecha's 1500rpm, and ramp up noislessly (other than airflow) to 1750rpm with a instance of Prime95 running. Running 3DMark 2001 to further stress the cpu and maxmize video card power consumption brings them to 1900rpm-with only a very moderate increase in the woosh of the machine.

Certainly not silent, but capable of dealing with the thermal consequences of a relatively advanced performance level. And really nice at idle. The 38mm thick Sanyo still provide good cooling power at ~1300 rpm, and whole lot more effective at 1750-1900rpm than the 25mm Mechas turning their 1500rpm.