Author Topic: A question about output pulldown of DC power supply.  (Read 1145 times)

0 Members and 1 Guest are viewing this topic.

Offline desertTopic starter

  • Contributor
  • Posts: 18
  • Country: cn
A question about output pulldown of DC power supply.
« on: March 02, 2023, 12:42:27 pm »
I heard that DC power supplies have a pulldown circuit at their output. When the output is disconnected (turn-off), the output port is internally pull-down to ground. And it’s not a simple short to ground nor connecting to ground through a resistor, but an active pulldown with a constant current.
To verify this, I ganged two DC power supplies together, set output voltage to 10V for them. I turned on one power supply, I can see 20mA current; then I turn on the other one, the current are 0s for both. After that, I turned off the first one, and the other power supply shows 20mA current. At last I increased the output voltage to 20V, the current is still 20mA.
Seems there is an active pulldown circuit.
Does active pulldown circuit exist on all kinds of power supplies, including SMU (Source Measurement Unit) and ATE VI (Automatic Test Equipment Voltage and Current Source)?
What's the purpose of the pulldown function?
Is there any example of active pulldown circuit diagram?
Thanks.
 

Offline Stray Electron

  • Super Contributor
  • ***
  • Posts: 2253
Re: A question about output pulldown of DC power supply.
« Reply #1 on: March 02, 2023, 01:01:07 pm »
  A lot of older power supplies worked this way. But I don't think that they were intentionally designed that way. However some PSUs were/are intended to work like this and were often referred to as "four quadrant power supplies" since they can sink as well as source current. And as you point out, it wasn't a dead short and you could control the amount of current flowing back into the power supply. HP makes a line of modern ones with digital displays that are intended to act similar to batteries in test circuits in that they can supply current (like a battery) or absorb current (like a battery charging). I have one but I don't recall the model number.

  You have to read the manual for the PSU in question and see what they say about the subject. A lot of the old PSUs would absorb a certain amount of power safely but it was usually considerably less than the amount of power that they were capable of supplying.

  It took a while to find something but this manual for the HP 6111A Power Supply talks about "Reverse Loading" in sections 3-48 and 3-50.  Long story short, it's not allowed with this power supply; and I suspect that it's not allowed in any power supply that is capable of being connected in parallel to other power supplies (see section 1-5) since one PSU would try to drive the other since their outputs could never be exactly equal.   https://xdevs.com/doc/HP_Agilent_Keysight/HP%206111A%20Operation%20%26%20Service.pdf

  I know that I have a couple of old (linear PSUs made HP I think) power supplies that do allow reverse power to be absorbed but I don't recall which ones so I will have to do some looking around.

    A couple of links  https://www.accelinstruments.com/Applications/TS200/Four-Quadrant.html  http://powersupply.blogs.keysight.com/2012/10/what-is-bipolar-four-quadrant-power.html


« Last Edit: March 02, 2023, 01:33:20 pm by Stray Electron »
 

Online thm_w

  • Super Contributor
  • ***
  • Posts: 7526
  • Country: ca
  • Non-expert
Re: A question about output pulldown of DC power supply.
« Reply #2 on: March 02, 2023, 10:05:56 pm »
Does active pulldown circuit exist on all kinds of power supplies, including SMU (Source Measurement Unit) and ATE VI (Automatic Test Equipment Voltage and Current Source)?
What's the purpose of the pulldown function?
Is there any example of active pulldown circuit diagram?
Thanks.

This is normal for any sort of modern lab bench power supply.
The main purpose is to discharge the output capacitors, so if you have the output voltage set to 20V, then set the voltage to 10V and there is no load on the output, the output won't just sit at 20V forever.

If you want a schematic, search for "constant current load", something like this with one transistor: https://www.ourpcb.com/constant-current-circuit.html

I don't think a SMU would have it, as it can already drive current in both directions, so its not needed.
Profile -> Modify profile -> Look and Layout ->  Don't show users' signatures
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf