ataraxey

@metal god
i really hope i dont need to do that. Would suck massive dongs to have to drain my entire loop, buy more coolant and then reassemble everything =/

Ashimon

What you are hearing are the inductors inside your graphics card or power supply's power regulation circuit. Many people incorrectly diagnose electronics noise, and capacitors do not have any moving parts nor can they induce an EM field necessary to cause motion and, ultimately, audible noise. The copper coils are what you're hearing.

That said - it very well could be your power supply power regulation inductors making noise, or the GPU inductors reacting to the power supply. This comes from the nature of the power supply changing Alternating Current (AC) to Direct Current (DC). Unfortunately, sometimes PSU power output can be 'dirty' while still being within factory specifications. That is to say that your PSUs output is causing those inductors to oscillate. These oscillations will cause noises ranging from whining to buzzing. It isn't an indication of failing equipment. It is an indication of inappropriate power conditioning.

maniacvvv

I ran 2 470's (see my MODS RIGS)
My TX850 could run them at stock, but I had to upgrade my PSU to overclock them properly...
Just because something "works" does not mean your not stressing it...

Simple test, pull the card out and go over to your friends house, remove all his cards and stick yours in it... Either that or swap PSU's with him.
--> if it does not make noise, your next stop should be the store for a new PSU

Dual 470's can eat 300w+ EACH when overclocked

I have no idea where you came up with 550w for an entire system and 2 470's 
Thats just completly incorrect 

I fail to understand why you would buy a $700 videocard, then use a "used" powersupply.

You could certainly have a defective card, but much more likely is that the powersupply your using is failing and you are risking both your build -and- your videocard. 

maniacvvv

Swap PSU's with him or someone else? seems like its that or RMA time on the videocard..
-or- buy a new PSU and rule that out 100% (would be my personal choice) 

I hear you bro, sorry your having problems.

Ashimon

The phenomenon of the inductor oscillation can vary between parts. it takes a special combination of factors to induce audible inductor oscillation. The values of the capacitors and inductors vs. the load presented to them, and the 'cleanliness' of the power being supplied.

The power that eventually gets to your CPU and GPU goes through several 'conditioning' stages. Your PSUs job is simply to convert AC to DC current as efficiently as possible (less heat). There are parameters that have been set as 'acceptable' output current pertaining to amperage, voltage, EMI, temperature, humidity, altitude, thermal shock, etc. etc. These are parameters - and every PSU sold will meet these parameters with unique values.

We all know that the graphics cards, CPUs, RAM, and other components differ in ways that allow for wide variations in overclockability. In that same vein, PSUs share a similar unpredictability. The way a particular PSU will interact with the components in your system will vary. In the case of the noisy inductors, the culprit can be found in the way the rectifier output is conditioned (filtered).

Inappropriate power conditioning is causing the coils on your GPU, Mobo, or PSU to oscillate. It can be any single one, a combination, or all three. Why does this happen? It really boils down to cost. This cost can be found in four main areas that manufacturers do their able best to minimize, while maximizing reliability, efficiency, and profit.

1) Compatibility research. There are so many combinations of parts that it becomes impractical to test the interactions between them all. Most PSU manufacturers I assume design a power supply to the specifications laid out by the form factor of their target market. This means that while their product performs to compatibility standards, it might not perform as well in your particular system as a competing product.

2) Design. This is a big one. Most of the issues described by noisy power supplies could be solved with better filtering. However this impacts the performance in three very significant ways: 1) It increases cost, 2) It can lower efficiency, and 3) It introduces additional points of failure. So manufacturers meet these challenges as well as they can. With power demands of individual machines going stratoshperic, these challenges put ever more pressure on the design and, perhaps inevitably, some designs will fall short of consumers' expectations.

3) Parts. The transformers, capacitors, resistors, inductors, diodes, transistors, and even the solder that make up today's electronics are mass produced on scales we cannot even fathom. I mean, it's one thing to sit back and read about Apple selling 6 million iPads. It's another to consider what 6 million even looks like. What mass production means is minimizing variance. Take two capacitors off the same production line and you will get two different values. These parts are tested and binned according to their performance and values. In the same way eVGA cherry picks the very best GPUs to go on their SSC-line of cards, so do the parts supply manufacturers. Tolerances drive costs; the tighter the tolerance, the higher the cost. So you've got a balancing act between creating the perfect product and cost. While this might seem elementary, it's absolutely crucial to the issue we're discussing. When building the PSU, a manufacturer simply pulls parts out of a bin that fall within a certain range of values (tolerance). While the final product is tested for tolerance and conformity, the product description tells us nothing about how much EMI that particular PSU emits, or how well the AC is filtered. How could it? It is a symptom of mass production which provides the best possible product at the best possible price (I'm assuming blind business morality here).

4) Competition. Take all the challenges presented already and throw competition into the mix. Not only are there concerns about market pliability (can I effectively sell this high-end, quadruple-filtered, super-silent, highly-efficient 1000W PSU when my competitor is selling a single-filtered, not-so-efficient 1000W PSU for half the cost?), there is pressure to provide power for an ever-hungrier consumer in a very limited amount of space and cost. Competition is a good thing for everyone involved. However, in a price-driven market competition can drive impractical, unsustainable, even unsafe design. Each company does its best to survive in an unpredictable market. But I am certain we can all relate to the notion that we’d like as much of a good thing for as little cost to ourselves as possible.


I hope no one’s fallen asleep. Class dismissed.

ataraxey UGHHHH. I guess i'll go to Fry's when im home for Spring Break from  Uni and grab one of their PSU's, see if the same problem persists, and  if it does RMA the card. Otherwise i'll return the PSU and buy another  one off newegg.

pretty awful problem on a $700 card =/

ps, sorry about my earlier language. frustration can do terrible things!

Afterburner

I think it is the Turbo Whine...