I have a dual output PSU that take 9-36VDC in, and outputs +/-12V. This PSU is powered by my DC bench supply. This PSU is powering 2 amplifiers for signal processing, and a HV supply. The HV supply takes in 0-12V, and outputs 0-1000V. I need 900V, so I put a linear voltage regulator in to step down the 12V to 9V.LVR circuit

When powered separately, the LVR/HV supply function properly, and I get the 900V output. Same with the 2 amplifiers. However, when all are powered together, the PSU enters its short-protection mode, and will not power on. I noticed that if I start my DC bench supply at 0V, and slowly go up to the 24V I need, the PSU powers on and everything works properly. If I power off the bench supply while it is set at 24V, then turn it back on, the short-protection of the PSU engages.

The data sheet for this PSU states it has a max load of 298 mA before the short-protection engages. However, when all components are powered on and functioning properly, my bench supply reads only 150mA at 24V. More interesting is if I ramp up the bench supply from 0V to ~10V, it reads 400mA of current! Shouldn't the PSU shut off?

I would like to simply turn the bench supply on/off while set at 24V. Is there some way to automate the slow 0V to 24V ramp up via some circuit? Any advice would be greatly appreciated.


What you are looking for is called a "soft-start" circuit.

It can be as simple as an NTC thermistor or "inrush current limiter". It is just a resistor with a large temperature coefficient. This means it will have good resistance at first, but quickly go to near zero resistance as the circuit comes up.

  • 1
    \$\begingroup\$ I searched for NTC thermistor softstart, and it does seem to be what I need. product.tdk.com/en/techlibrary/applicationnote/… Looking at the DC-DC converter section, all I would need to do is add this NTC between my bench supply and the PSU? Seems awfully simple. \$\endgroup\$ – Tristen Lee Mar 30 at 19:29
  • \$\begingroup\$ Yes, just put it in series on the positive leg between the bench supply and PSU and away you go. This is a pretty common problem, and the thermistor is a common solution. \$\endgroup\$ – evildemonic Mar 30 at 19:33
  • \$\begingroup\$ What considerations should I take when choosing a thermistor? \$\endgroup\$ – Tristen Lee Mar 30 at 23:25
  • \$\begingroup\$ You use the typical current you expect through it to size it. You also probably want one with as low a resistance as possible. \$\endgroup\$ – evildemonic Mar 31 at 2:06

Another posibility should be to soft the output voltage of LM317. Minimum voltage is 1.2 V, and you have ajusted output to 9 V. If you increase resistors R1 and R2 with the same relationship to keep 9 V output (to reduce control current), and connect a capacitor between pin 1 (ADJ) and ground, the output voltage will be 1.2 V at the start, and will increase to 9 V in ramp.

  • \$\begingroup\$ Interesting. So the time to get from 1.2V to 9V will be associated with this new RC circuit (R1 and R2 in parallel along with the capacitor on pin 1)? Since R2 is variable, could I just increase R1 (which is fixed), and tune the output back to 9V via R2? \$\endgroup\$ – Tristen Lee Mar 31 at 21:58
  • \$\begingroup\$ If R1=2k2 ohms, R2 should be about 12 kohms ajustable (take in acount I_ADJ from LM317 max 100 uA). You can check with a capacitor of 1000 or 2200 uF, may be higher, if inrush current is below the bench supply limit.A diode to discharge this capacitor is also convenient (anode to capacitor, cathode to output voltage), that way it is quickly prepared for next start when you shut down. \$\endgroup\$ – Bravale Apr 1 at 8:50

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.