0
\$\begingroup\$

I have a bunch of cheap buck (I think) converters that do 6-20V input and 5V output; and I want to use one powered by a 24V PSU (this is for a 3D printer).

The current draw of the 5V system should be comparatively low, and constant, but the main supply will vary quite a bit in draw. as the regulator can deal with 6-20 volts I wonder if I can 'scale down' or 'clamp' the supplied power via a simple voltage divider?

I know that usually this is a solid 'No' as voltage drop varies according to current draw, but in my case:

  1. I need only target a range of voltages between 6-20V versus a single stable voltage
  2. The input to this is a regulated power supply, which should stabilize the voltage (which transitively stabilizes the current, right?), and I expect this should help.

What about some arrangement with Zener diodes? As I look this up mid-post, it seems along the lines of what I am looking for (i.e. something I can throw together with simple passives I am likely to have or can find in scraps): https://en.wikipedia.org/wiki/Zener_diode#Voltage_shifter

But power supplies and analog electronics always surprise me with their subtlety and I don't trust myself enough to know, so would something like the above work, or should I just buy different DC-DC converters and call it a day?

-- new edits as of 2022-12-17 --

Also, I have a bag of these 6-20v -> 5v converters, could I hook two up in series (or otherwise passively arranged) to get effectively 10v per input, but hook their outputs up in parallel?

I will have to dig up the specs on these converters and the power supply, I was more curious if there were generally applicable approaches to this problem given the circumstances (more or less regulated power on each side of the converter).

\$\endgroup\$
7
  • \$\begingroup\$ I order to work out the values for a voltage divider or zener (or even if it's practical) you'd need to know what the min/max range of current draw into your DC-DC converter would be. \$\endgroup\$
    – brhans
    Dec 15, 2022 at 18:23
  • 1
    \$\begingroup\$ You can use an npn to drop the voltage, I've used this with a DC to DC once \$\endgroup\$
    – Voltage Spike
    Dec 15, 2022 at 18:28
  • \$\begingroup\$ What is the quiescent current draw of the buck converter you have? \$\endgroup\$
    – Andy aka
    Dec 15, 2022 at 18:56
  • \$\begingroup\$ Consult the PSU's documentation: adjustable output voltage is not unheard of, 22V would not be untypical for a nominal 24 V supply. Zener diodes don't have much of an advantage compared to a simple voltage divider from stabilised 24 V (or 22 V…). Just buffer it properly, I'd try and first simulate a three transistor "complementary feedback triple" as an emitter follower. \$\endgroup\$
    – greybeard
    Dec 15, 2022 at 21:10
  • \$\begingroup\$ hook two up in series [to] get effectively 10v per input 24/2 ≠ 10. \$\endgroup\$
    – greybeard
    Dec 17, 2022 at 17:43

2 Answers 2

1
\$\begingroup\$

Anything with a voltage divider or a shunt (Zener) regulator will be very inefficient. Further, the voltage divider output will vary with your 5V loading. That may be ok or you, but not knowing your system current demand particulars it’s impossible to say.

If your overall current is low enough you could use a series regulator like an LM317 to make your 20-ish V pre-rail for the 5V switchers.

\$\endgroup\$
4
  • \$\begingroup\$ You can simply use a stack of at least 6 diodes, just be sure they will handle the current. Check the forward voltage to be sure it is enough. \$\endgroup\$
    – Gil
    Dec 15, 2022 at 19:54
  • \$\begingroup\$ LM317s are cheap - quite possibly cheaper than a Zener or 6 large-ish 1N4001 type diodes. And, 1 device vs. 6. \$\endgroup\$ Dec 15, 2022 at 20:06
  • \$\begingroup\$ 3A diodes are less then $0.05. Nice thing about using diodes there is no additional parts or caps needed. What is the total load current. \$\endgroup\$
    – Gil
    Dec 15, 2022 at 20:18
  • \$\begingroup\$ I’m seeing LM317 for less than 20 cents in Asia. \$\endgroup\$ Dec 15, 2022 at 21:07
1
\$\begingroup\$

If the operation conditions are really constant, you can get away with inserting an appropriately sized resistor (or several to get the power rating), to burn off the excess voltage.

If a wider load current range is possible, I second the comment of @VoltageSpike to use a power transistor to burn off the excess heat. An NPN can be used for lower currents, but for higher currents and power handling capabilities, I think MOSFETs might be easier to obtain with the required power. And MOSFETs don't load the resistor R1.

Good thing with this circuit as compared to Zener droppers, is that you don't need a special valued high power Zener around but can parts that you may have in your BOM already.

schematic

simulate this circuit – Schematic created using CircuitLab

\$\endgroup\$

Your Answer

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

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