1
\$\begingroup\$

I am using a computer PSU as a lab power supply. In the current setup, I am drawing about 1ma on the 5VDC rail. And still the multimeter indicates 5.18VDC. I can upload a scope show if needed, but this seems too large of an error for a multimeter.

Is this normal? Why would a regulated power supply output a very incorrect voltage value?

\$\endgroup\$
4
  • 6
    \$\begingroup\$ That value is within +- 5%. That is completely acceptable. \$\endgroup\$
    – vini_i
    Nov 21, 2015 at 14:45
  • 1
    \$\begingroup\$ What's the power supply rated output? If it's many amps and you have it pretty well unloaded it may have difficulty regulating accurately. You might also find that having a fraction of a volt extra will guarantee 5 V on the computer board despite voltage drop along the wires. \$\endgroup\$
    – Transistor
    Nov 21, 2015 at 14:53
  • 1
    \$\begingroup\$ Of course I am drawing about 0.004% of the rated 23A. \$\endgroup\$
    – Vorac
    Nov 21, 2015 at 15:00
  • 1
    \$\begingroup\$ @transistor - maybe paste your comment into an answer; it seems like a resonably complete answer. \$\endgroup\$
    – gbulmer
    Nov 21, 2015 at 15:14

4 Answers 4

2
\$\begingroup\$

As others have mentioned, 1) you are drawing only a tiny fraction of the current usually drawn by the computer; and 2) the 5.18V is probably within spec for a 5V rail (although they are usually closer than that).

If you need the voltage to be nearer to 5v, you could try putting a 10Ω 5W resistor across the output terminals, which will put a load of 500 mA across them, consuming 2.5W. See if that helps. RadioShack sells a 10Ω 10W resistor, 2 for $2.49, which will run cooler than a 5W one anyway.

enter image description here

\$\endgroup\$
7
\$\begingroup\$

If the power supply is rated at several amps and you have it pretty well unloaded it may have difficulty regulating accurately. You might also find that having a fraction of a volt extra will guarantee 5 V on the computer board despite voltage drop along the wires.

\$\endgroup\$
1
  • 2
    \$\begingroup\$ Cheap USB cables drop a lot of voltage, and cheap USB devices might not work at the low end of the allowed range (4.35 V), so many machines deliberately increase their USB power to near the upper end of the allowed range (5.25 V). \$\endgroup\$
    – CL.
    Nov 21, 2015 at 16:33
2
\$\begingroup\$

Low cost power supplies only regulate the voltage at the (internal) terminals of the supply. The drop across the resistance of the leads between the internal terminals and your load is not controlled by the supply. in order to ensure that there is 5 V at the load at the max. current, a supply generally has a regulation set point at a slightly higher output voltage.

If you draw maximum load current, you will find the voltage will be much closer to 5 V.

\$\endgroup\$
0
\$\begingroup\$

5.2V is well within USB spec. Especially knowing that this is a no load voltage. The USB standard voltage is 5V, but can range from 5.25 to 4.45 while remaining in spec. If you needed tighter regulation, you could use a boost switching regulator to get the voltage up past 6V, then use a 1% 5V LDO. However, this is not necessary if you are using the 5V rail as a USB charger, or for CMOS logic such as the 74AHC family or related.

\$\endgroup\$
1
  • \$\begingroup\$ Good to know, but I am using a bare ATX PSU. \$\endgroup\$
    – Vorac
    Nov 24, 2015 at 7:02

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.