Are you a PSU Expert? Prove it!

If you believe that you are an expert on power supplies, you should give this quiz ago. This is one of the quizzes that the Cybenetics personnel has to deal with and pass with a 75% score, at least.

1. The ATX connector mainly consists of AWG20 gauges in a PSU. Is this the suggested gauge thickness according to the ATX v3.0 spec?

 
 
 

2. The EPS and PCIe gauges in PSU A have 16AWG thickness, and in PSU B have 18AWG. Which of the following applies, given that all the rest parts in the PSUs are precisely the same?

 
 
 

3. A PSU has both EPS connectors on the same cable. Is it safe to fully load both simultaneously?

 
 
 

4. The PSU is tested with 115V and 230V input voltages. Which of the following should you expect?

 
 
 
 

5. In PSU A, load regulation at +12V is 2%, and in PSU B, load regulation on the same rail is 1%.

 
 
 
 

6. In PSU A, ripple is 100mV at +12V, while in PSU B, ripple is 40mV on the same rail.

 
 
 

7. In PSU A, efficiency with a 2% load exceeds 60%, while in PSU B, efficiency with the same load is lower than 60%.

 
 
 

8. When you measure T1 (power-on time) and T3 (PWR_OK delay) timings, what is recommended for a PSU to be ALPM (Alternative Low Power Modes) ready?

 
 
 
 
 
 

9. Which rail’s timings are measured to define ALPM compatibility?

 
 
 
 
 

10. A PSU’s rails should increase to 95% of their nominal voltages (T2, rise time) within the following period.

 
 
 
 

11. According to the ATX v3.0 spec, PSUs without 12VHPWR connectors or with 450W and lower capacity should be tested with transient loads up to:

 
 
 
 
 

12. According to the ATX v3.0 spec >450W PSUs with 12VHPWR connectors should be tested with transient loads up to:

 
 
 
 
 

13. According to the ATX v3.0 spec, the shortest transient load apply period is:

 
 
 
 
 

14. We have a 1000W PSU. Which peak transient load will we apply if it is ATX v3.0 ready and doesn’t have a 12VHPWR connector?

 
 
 
 
 

15. We have a 1000W PSU. Which peak transient load will we apply if it is ATX v3.0 ready and has a 12VHPWR connector?

 
 
 
 
 

16. If a 12VHPWR connector lacks the Sense0/1 signals, the maximum GPU power draw will be the following:

 
 
 
 
 

17. If a 12VHPWR connector has both Sense0/1 signals grounded, the GPU knows that it can draw up to:

 
 
 
 
 

18. If a 12VHPWR connector has both Sense0/1 signals open, the GPU knows that it can draw up to:

 
 
 
 
 

19. If you apply 90V instead of 115V input in a PSU (line regulation), what of the following should you expect?

 
 
 
 
 

20. In a PSU A, vampire power is 0.25W, while in PSU B vampire power is 0.1W. Which will have higher efficiency at 5VSB with a 0.1A load?

 
 
 

21. What do you make of the following Fan RPM/Noise graph?

 
 
 
 

22. What do you make out of the following Fan Speed Map?

 
 
 
 

23. What is the lowest voltage for the 12V rail on any connector according to the ATX v3.0?

 
 
 
 

24. What is the lowest voltage for the 5V rail on any connector according to the ATX v3.0?

 
 
 
 

25. What is the lowest voltage for the 3.3V rail on any connector according to the ATX v3.0?

 
 
 
 

26. What is the highest voltage for the 12V rail on any connector according to the ATX v3.0?

 
 
 
 

27. What is the highest voltage for the 5V rail on any connector according to the ATX v3.0?

 
 
 
 

28. What is the highest voltage for the 3.3V rail on any connector according to the ATX v3.0?

 
 
 
 

29. Which is the maximum slew rate for >450W PSUs with 12VHPWR connectors at +12V, according to the ATX v3.0 spec?

 
 
 
 
 
 

30. Which is the maximum slew rate for <=450W PSUs or without 12VHPWR connectors at +12V, according to the ATX v3.0 spec?

 
 
 
 
 
 

31. What is the maximum slew rate for the minor rails (5V & 3.3V) on all PSUs, according to the ATX v3.0 spec?

 
 
 
 
 
 

32. With a full load, the 5VSB rail’s efficiency should be equal to or higher than:

 
 
 
 
 
 
 

33. With 1.5A load, the 5VSB rail’s efficiency should be equal to or higher than:

 
 
 
 
 
 
 

34. With 0.55A load, the 5VSB rail’s efficiency should be equal to or higher than:

 
 
 
 
 
 
 

35. With a 45 mA load, the 5VSB rail’s efficiency should be equal to or higher than:

 
 
 
 
 
 
 

36. The maximum allowed ripple at +12V according to the ATX spec is:

 
 
 
 
 
 
 

37. The maximum allowed ripple at 5V according to the ATX spec is:

 
 
 
 
 
 
 

38. The maximum allowed ripple at 3.3V according to the ATX spec is:

 
 
 
 
 
 
 

39. The maximum allowed ripple at 5VSB according to the ATX spec is:

 
 
 
 
 
 
 

40. Which transient response results must we consider according to the ATX spec?

 
 
 

41. Check the following scope screenshot carefully and check the right answer. Blue is +12V, red 5V, and orange 3.3V

 
 
 

42. A PSU has a 17ms hold-up time, and its power ok signal’s hold-up time is 15.5 ms. Does this PSU follow the ATX spec regarding hold-up timing requirements?

 
 
 

43. A PSU has a 17ms hold-up time, and its power ok signal’s hold-up time is 19 ms. Is the power ok signal accurate?

 
 
 

44. What can an inaccurate power ok signal cause:

 
 
 
 

45. What can a short hold-up time cause:

 
 
 
 

46. I heat the secondary side of the PSU till it reaches 200°C and doesn’t shut down. What does this mean?

 
 
 
 
 

47. Which fans are better for high operating temperatures?

 

 
 
 
 

48. Which is the average noise output value that Cybenetics has set for its best noise certification?

 
 
 
 

49. The lower the leakage current, the better?

 
 

50. A PSU has an OCP triggering point high enough that allows the +12V to have 200mV ripple. Is this correct?

 
 
 
 

51. What is essential for a good transient response on all rails?

 
 
 
 
 

52. Which bulk cap from these options is better to use to achieve higher reliability (we assume both have the same build quality)?

 
 

53. Which is the best noise output level (LAMBDA) in the Cybenetics scheme?

 
 
 
 
 
 
 

54. Check the following +12V oscilloscope screenshot and choose the correct answer.

 
 
 
 

55. A short hold-up time is the cause of the following:

 
 
 
 
 

56. Which are the most crucial factors we must consider when evaluating a PSU?

 
 
 
 

57. Which is the highest efficiency level (ETA) in the Cybenetics scheme?

 
 
 
 
 

58. When should you expect higher inrush currents.

 
 
 
 

59. I measure high EMI spurs during the EMC pre-compliance test with both the average and peak EMI detectors. What does this mean?

 
 
 
 
 

60. Which fans produce lower noise output under the same operating conditions?

 

 
 

61. What is the appropriate Over Current Protection (OCP) triggering point for all rails?

 
 
 
 
 

62. A non-accurate power ok signal is because:

 
 
 
 
 

Related Posts

8 thoughts on “Are you a PSU Expert? Prove it!

  1. Do not take the test if you’re in a hurry.

    Some questions are worded weirdly, with lots of unnecessary words. Also, special cases are not considered (ie: PSUs with alternative inrush limiter approach like the Silverstone PTS series).

    Just read carefully and consider that a non native English speaker wrote this.

    Also, do not use the older ATX revisions as reference. Make sure you have the latest ATX v3 revision.

    Good luck

    1. Non-native English speaker: My pronunciation might not be the best, but my writing skills are in a whole different level so I don’t think that this applies.

      1. It does, but it is minor. There are just some idiosyncrasies in English and in Greek that do not play very well with each other.

        Note that I had to take the test twice to top it, cause of weird past experiences and not paying attention on some tricky questions, resulting in couple of errors on the first try.

        Apart from my personal opinion with the approach of the test going overboard on the technical terms, the test looks pretty much fine.

        Good work Aris!

    2. I did the test in a hurry, without Googling anything. I’m experienced 😀 So it can be done. 😉

      But I believe there was a mistake. Shouldn’t question two be PSU A with the 16AWG wires.

        1. Exacly!

          And question 21 is more like an opinion and not really a fact.

          In my opinion it’s agressive because the lowest fan speed is above 1000 RPM and it’s at 2000 RPM before full load.

          1. It is aggressive, the fan speed profile for the following reasons:
            It doesn’t increase linearly, from zero RPM goes to well over 1000 RPM without any intermediate stages
            At high loads, it is steady at full speed

Leave a Reply

Your email address will not be published. Required fields are marked *