Hot Test Results

In the following tables we can see that the BitFenix Formula Gold 650W delivers fair power quality. The maximum voltage ripple on the 12V line is 34 mV, which may be an unimpressive figure nowadays but it is just 28% of the ATX design guide's recommended 120 mV limit. The filtration on the minor 3.3V/5V lines is a little worse, with the ripple reaching 22 mv (44% of the 50 mV ATX design guide recommended limit). On the contract, voltage regulation on the 3.3V/5V lines is at 0.5%, an exceptional figure, whereas the regulation of the 12V line reaches 1.4%. Keeping in mind the class of the product, these figures are excellent.

Main Output
Load (Watts) 131.08 W 326.57 W 487.81 W 647.02 W
Load (Percent) 20.17% 50.24% 75.05% 99.54%
  Amperes Volts Amperes Volts Amperes Volts Amperes Volts
3.3 V 1.78 3.35 4.45 3.35 6.67 3.33 8.89 3.33
5 V 1.78 5.03 4.45 5.02 6.67 5.02 8.89 5.01
12 V 9.6 12.1 24.01 12.05 36.01 12 48.02 11.93

 

Line Regulation
(20% to 100% load)
Voltage Ripple (mV)
20% Load 50% Load 75% Load 100% Load CL1
12V
CL2
3.3V + 5V
3.3V 0.5% 12 16 20 22 12 20
5V 0.5% 14 16 18 22 14 22
12V 1.4% 12 18 26 34 30 28

Operation in high ambient temperatures affects the performance of the Formula Gold 650W PSU by a measurable, yet reasonable amount. The average nominal load range (20%-100%) efficiency drops by 0.8% regardless of the input voltage, going down to an average of 90.1% (230 VAC) / 88.2% (115 VAC). The efficiency drop is miniscule at lower loads but is more significant at high loads, suggesting that the components of the PSU are getting slightly thermally stressed under these conditions.

A look at our temperature graphs comes to back up our previous assumption, as the internal temperatures of the Formula Gold 650W PSU are quite high for a unit with this kind of power output. The thermal control does make an effort to keep the temperatures within reasonable levels, forcing the fan to spin a lot faster than before. Although the fan reaches uncomfortable noise levels with a load of just 300 Watts, it is unable to halt the ever-rising temperature. The Formula Gold 650W PSU barely works at maximum load under these conditions, as the temperatures are right before the point where its thermal protection would shutdown the unit.

Unlike operation in room temperature, the thermal control circuit will continuously increase the fan's speed under these conditions. The increase rate is almost linear, meaning that every increase on the unit's load will also increase the unit's noise output. The noise coming from the PSU's fan reached uncomfortable figures with a load of 300 Watts, with the sound pressure level going above 50 dB(A) at maximum load.

Cold Test Results (Room Ambient Temperature) Final Words & Conclusion
POST A COMMENT

12 Comments

View All Comments

  • DanNeely - Wednesday, June 13, 2018 - link

    "Modern power supplies work on both systems, however the nature of switching means that they are slightly more efficient with higher voltage inputs. "

    Is there something specific to switch mode design that makes lower voltage hurt? I'd always assumed it the difference was due to the 4x higher I^2*R resistive losses on the input end.
    Reply
  • E.Fyll - Thursday, June 14, 2018 - link

    That is generally correct. Actually, a lower input voltage will decrease the switching-related frequency losses as well (the switching cycle is longer to maintain the same energy output). That's why this PSU displayed almost identical efficiency at lower loads regardless of the input voltage -the resistive losses are higher with a 115V input but the switching losses at low loads are lower, so they virtually negate each other. At higher loads however, the switching losses are minimized regardless of the input voltage, whereas the resistive losses are (theoretically) always four times higher. They actually tend to be even higher than that, because higher component temperatures increase the resistive losses, and the higher resistive losses increase the component temperatures even further. It's a vicious circle. Reply
  • benedict - Wednesday, June 13, 2018 - link

    Sell it for 60-65$ and it will be a hit. At 80$ it's just one of many. Reply
  • Orange14 - Wednesday, June 13, 2018 - link

    For $10 more I could purchase an equivalent wattage Seasonic that is fully modular and has a 10 year warranty, At this price point, there is no compelling reason for this PSU to exist. Reply
  • AdrianB1 - Wednesday, June 13, 2018 - link

    You are totally right. For this price it should be at least semi-modular and have better features, like passive mode and/or higher efficiency. Reply
  • qlum - Thursday, June 14, 2018 - link

    I personally prefer non modular over modular as I prefer not ot use a shroud and non modular just looks better on the front of the case. Tje few extra cables I can easily hide on the back anyway. Reply
  • AdrianB1 - Thursday, June 14, 2018 - link

    Semi-modular is the best of the 2 worlds, you have the ATX 24 pin cable built in like in a non-modular PSU and the other cables are connected only if you need them, so you don't have to hide anything on the back, leave it in a drawer. Reply
  • Death666Angel - Friday, June 15, 2018 - link

    I don't follow. How do you see your PSU from the front of your case? What kind of shroud do you use with your PSU and why are more cables better in that case? Do you use some very specific cases?
    My last non (semi) modular PSU was a Silverstone one I used in a Lian Li mATX box (V351b) with a few short cables that fit my need. At that point, modular was still a good bit more expensive. After that though, I had at least semi-modular or fully modular PSU and like the ease of use of installing them, as well as the less clutter. But I also use mATX or smaller cases.
    Reply
  • boozed - Wednesday, June 13, 2018 - link

    Not to be confused with Bitfinex.

    I did once see the Bitfenix logo used by accident at a fraudcoin conference.
    Reply
  • crashtech - Saturday, June 16, 2018 - link

    At this very moment, the SeaSonic SSR-650RM is a few dollars less, so there's no reason to buy this PSU, which ought to be in the $65to be at all interesting. Reply

Log in

Don't have an account? Sign up now