Hot Test Results (~45°C Ambient)

The Toughpower GF A3 750W PSU is atypical because Thermaltake claims its performance is rated at 45°C. Typically, PC PSUs are rated at either 40°C for low-end and entry-level mainstream units or 50°C for quality mid-tier and high-performance products. This alone would suggest that Thermaltake is somehow trying to wedge its way in between these two worlds.

Nonetheless, the performance of every PC PSU will degrade once the ambient temperature rises, with the magnitude of the drop primarily affected by the design and quality of the unit itself. While the average energy conversion efficiency drop of the Toughpower GF A3 750W PSU is 0.75%, a relatively high figure for a quality 80Plus Gold unit, it is uniform across the entire load range, with no significant change when it is heavily loaded. These results suggest that the thermal stress on the active components is mild to low.

Despite the high ambient temperature, the Toughpower GF A3 750W PSU still managed to keep its fan off while the load was low. The high temperature forces the fan to start earlier than before, with the fan constantly increasing its speed to keep up with the unit's cooling demands. It reaches its maximum speed when the load is just over 500 Watts, after which point the fan cannot do anything more to help with the unit's rising cooling demands, resulting in climbing temperatures. We had the OTP protection kick in when the unit was operating at maximum load for about three minutes, which is good to see that the OTP works normally and protects the unit if necessary. It's troubling because, clearly, the unit could use a more powerful fan. Nonetheless, the Toughpower GF A3 750W did operate without electrical issues even at maximum load, where the temperature of the secondary side went unnervingly high. The PSU is undoubtedly not designed to be stressed like this for prolonged periods, but it is good to find out that it can withstand it for short periods if it has to.

Cold Test Results (~25°C Ambient) Power Supply Quality & Conclusion
Comments Locked

10 Comments

View All Comments

  • meacupla - Monday, October 23, 2023 - link

    I find it interesting that TT doesn't use their toughfan in this. The 120mm version has good static pressure, which seems ideal for a PSU. Especially when it has a restrictive stamped finger guard.
  • Samus - Monday, October 23, 2023 - link

    Probably just money. I have been constantly let down by mid-range PSU fans and I get it, they need to cut corners somewhere, but the fan is arguable the most important part because its naturally, being mechanical, should be the only component that will eventually fail. And they constantly do.

    I recently replaced a 120mm fan in a Seasonic Focus. You would think anything Seasonic would be of exceptional quality but the fan started chattering, and at closer inspection the axial shaft in the motor disconnected from the fan housing and would be very difficult to repair (and not worth repairing for such a junk FDB fan)

    Disturbingly these fans are used in their top-dollar Prime PSU's and upon research, people do have failures. While Seasonic has a 10 year warranty, even if you can provide proof of purchase and are willing to send it in and wait a week or two for RMA, is it worth it to get a PSU with the same crappy fan back when you can upgrade it to a superior fan for $10.

    Obviously I chose to replace it with a better fan, which revealed another problem. The fan controller of the PSU is calibrated for the specific power draw of the fan, in my case it was 0.58A. I installed a Noctua 120mm fan that draws only 0.05A, and it runs at full speed all the time making more noise than any other component in my PC because the PSU is expecting a 0.58A fan. The Prime PSU's have a silence button that doesn't run the PSU fan until it's needed (power draw\temp at a certain threshold) but the Focus lack this button, so it's just whizzing away at full speed all the time.

    All of this could have been avoided if they just used a good fan, but again, it's a $100 PSU, I realize compromises had to be made. But when you consider this same crappy fan is used in their $300 PSU's, I'd be pretty damn pissed if this fan failed in one of those because you expect a quality ball bearing fan or something when spending top dollar.

    My PC Power & Cooling 750 Quad has an 80mm ADDA Ball Bearing fan and is subjectively silent compared to most other comparable modern PSU's. And it's 15 years old.
  • back2future - Monday, October 23, 2023 - link

    [ on a 50mA current You could possibly install series resistors being a voltage divider to the fan's supply voltage, what might give You some control for adjustments on rpms(?) ]
  • Samus - Tuesday, October 24, 2023 - link

    I thought about adding a resistor or a rheostat, but since this isn't a modular PSU it's a pain in the ass to remove and I've decided to just live with the fan noise. It isn't loud enough to be a bother (this is a 1600rpm Noctua fan) but it's annoying in doing so I made the quietest component of my PC the loudest at idle.

    It's my fault for lapsing on the fact I was dealing with a 2-pin fan that would obviously be miscalibrated to a control circuit designed for an entirely different fan.
  • back2future - Tuesday, October 24, 2023 - link

    [ You even need to cut the org. fan cable, since for a $1-2 there are extension cords 'https://ae01.alicdn.com/kf/S4616494b80094bcb816bcb... or connect that power supply fan with a mainboard pwm controller for load dependent rpm control, if 12V, therefore having a configurable, spare fan port or a (split) extension cable 'https://images-eu.ssl-images-amazon.com/images/I/7... ]
  • back2future - Tuesday, October 24, 2023 - link

    [ meant: 'even need not to cut the org. fan cable' ]
  • Samus - Monday, October 23, 2023 - link

    With the 4070Ti being a 285w TDP card and the 12vHPWR connector rated at 300w, does the aggressive OCP pose a threat to premature power cutoff if the card happens to spike above 300w? I feel that the 4070Ti is a realistic pairing with this PSU.
  • love2fly59 - Tuesday, October 24, 2023 - link

    Please excuse my newbie question but how can this be possible:

    "The efficiency curves would easily allow the PSU to receive an 80Plus Gold certification for an input voltage of 115 VAC. Still, it would not be able to receive the same certification for an input voltage of 230 VAC as it could not reach an efficiency greater than 92% at half load"

    when the efficiency curves in this test show that 230 VAC efficiency measurements are consistently better than those for 115 VAC?
  • love2fly59 - Tuesday, October 24, 2023 - link

    Can it be that 115 VAC has lower efficiency requirements than 230 VAC?
  • MachineLearning - Wednesday, October 25, 2023 - link

    Power supplies operating on 230v are generally more efficient than when they operate on 115v / 100v. Therefore 80plus and Cybenetics standards for efficiency are higher for 230v than 115v, since a lower quality power supply could theoretically be 80plus Gold on 230v but only Silver or Bronze on 115v. The seemingly stricter requirements are to account for this.

    Hope this helps.

Log in

Don't have an account? Sign up now