For large installations of 5V LEDs, this PixLite page suggests using "power injection" from supplemental power supplies. Their wiring sample says:

Note that the DC ground of all power supplies used should be connected together to provide a single common ground reference. In addition to this, the positive output from each power supply should be isolated in order to avoid differing voltage potentials occurring across multiple power supplies, which could result in damage.

Multiple 5V power supplies injecting same circuit

I am trying to imagine a way in which "damaging" voltage (which I assume would mean significant overvoltage) could be developed by injecting from identical power supplies without isolation on the positive side if they share the same ground reference. Is that possible, and if so then how?

  • \$\begingroup\$ Yes, you can do this if you link the datasheet or website for the LEDs, the intended string configurations, the PS datasheet and measure the error voltage before connecting to avoid circulating currents. Can you do this and indicate what you have and are planning to get? \$\endgroup\$ – Tony Stewart Sunnyskyguy EE75 Oct 7 '18 at 17:21
  • \$\begingroup\$ @TonyEErocketscientist my question pertains to whether and why the instructions provided are correct. I.e., a great answer can start with the example shown and explain exactly what negative/damaging effects could result from not isolating the positive line from each DC supply. \$\endgroup\$ – feetwet Oct 7 '18 at 18:47
  • \$\begingroup\$ It is assumed all outputs are floating. But I expect all your specs. \$\endgroup\$ – Tony Stewart Sunnyskyguy EE75 Oct 8 '18 at 0:33

As the power supply is a low impedance source: even small differences in voltage output can result in large currents flowing between them if the supplies are connected in parallel. Depending on the design of the supply this may cause damage.

Rather than connecting the supplies as shown it would be better to connect the power and ground at the strip for each auxiliary supply. This will keep the ground current in the common ground within sensible limits and as the power ground is also used as a signal ground will keep the ground noise down for signalling.

| improve this answer | |
  • \$\begingroup\$ So if I understand what you're saying: The risk is not overvoltage, but rather overcurrent? I.e., if the voltage varies even small amounts between the supplies (which in practice is inevitable) then you can get unexpected current flows within the circuit? Or are you talking about something internal to the power supplies that would only pose a risk to the supplies themselves? \$\endgroup\$ – feetwet Oct 7 '18 at 16:12
  • \$\begingroup\$ If you parallel the supplies you can get large currents flowing between them. \$\endgroup\$ – RoyC Oct 7 '18 at 16:13
  • \$\begingroup\$ I'm not understanding the difference you've described between "parallel supplies" and wiring the whole thing with a "common ground." Could you sketch an "incorrect" circuit of the former, and then show the "correct" version with common ground? \$\endgroup\$ – feetwet Oct 20 '18 at 17:13
  • \$\begingroup\$ Supplies are never exactly identical there will be small differences in the voltage set point. Your diagram is the less ideal one. The better one has the horizontal ground line just below the supplies removed and the ground from each extra supply connected directly to the strip it is driving as well as the ground from the driver board. \$\endgroup\$ – RoyC Oct 21 '18 at 8:52
  • \$\begingroup\$ The common ground is simply the ground connecting the strips and the driver board. Try to think about where current is flowing in your circuit. \$\endgroup\$ – RoyC Oct 21 '18 at 8:58

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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