Products > Test Equipment
Lab Power Supply Turn ON and OFF Characteristics
Martin72:
Something else that occurs to me when I see all these pictures:
At work, so-called discharge circuits are tested from time to time.
The principle is simple: if the auxiliary voltage is removed (the device has been switched off), an FET conducts and the electrolytic capacitors in the DC link can discharge quickly via load resistors.
And very quickly at that...
So, now colleagues had the “problem” when testing this board that the discharge curve, which is normally very steep, drops rather “limp”.
What's the problem? At the power supply unit that provides the auxiliary voltage for testing the board.
We have dozens of laboratory power supplies with these “enable” buttons for the outputs.
If you press this to turn it off, the voltage is not immediately gone, but decreases slowly - as some pictures here show.
Do you know what I'm getting at...
It is possible that some power supplies have a capacitance at the output after the switch.
That is why the output voltage is not immediately gone after switching off.
nctnico:
--- Quote from: Martin72 on October 06, 2024, 09:01:17 pm ---Something else that occurs to me when I see all these pictures:
At work, so-called discharge circuits are tested from time to time.
The principle is simple: if the auxiliary voltage is removed (the device has been switched off), an FET conducts and the electrolytic capacitors in the DC link can discharge quickly via load resistors.
And very quickly at that...
So, now colleagues had the “problem” when testing this board that the discharge curve, which is normally very steep, drops rather “limp”.
What's the problem? At the power supply unit that provides the auxiliary voltage for testing the board.
We have dozens of laboratory power supplies with these “enable” buttons for the outputs.
If you press this to turn it off, the voltage is not immediately gone, but decreases slowly - as some pictures here show.
Do you know what I'm getting at...
It is possible that some power supplies have a capacitance at the output after the switch.
--- End quote ---
1) The on/off button isn't driving a relay but simply forces the output voltage setting to 0 at the point where the setpoint is fed into the output stage. I have modified quite a few power supplies to add this feature. Simply short the voltage control potmeter to ground to disable the output using a toggle switch. But in a digitally controlled PSU it is a simple matter of having the DAC which drives the output stage output 0V. Same principle though.
2) The rapid discharge effect you are describing is a parallel power sinking circuit (Keysight calls this a downprogrammer https://docs.keysight.com/kkbopen/66xxa-down-programming-589741416.html) which kicks in when the output voltage is higher than then setpoint voltage. So when an output is switched off, this circuit will kick in because the output voltage is higher than the setpoint. But it will also kick in when power is fed into the power supply. Attach a fan and turn the output on / off. Chances are you see the current becoming negative due to the sink circuit absorbing the power from the fan while it is spinning down and working as a generator. However, this feature is not available in all power supplies so you need to check the specs. A way to tell whether a PSU has a sink circuit is when the voltage fall settling times are specified as well.
sonpul:
The RD6006P has a 10uF output. When turned off, the output is shunted to ground by the circuit, accelerating the discharge of this capacitor. In my screenshot of the shutdown, you can see the beginning of the discharge first and then a sharp decline.
ArdWar:
Please note that my resistor load is 45Ω
R&S NGP804
PS001: Turn ON without any slew control. Stays at 8V for approximately 16ms before exponential decay and settle into ~2V. Which means slightly more than 40mA setpoint. I don't have external DMM, but internal readback is 41mA.Scope reading hints even more than that (~44mA)
PS002: Turn ON with 20ms slew setting. Peaked at 7.5V before decay. If you extrapolate out the initial jump, the slew is indeed approx 20ms, into 8V... (i.e. ~400V/s)
PS003: Turn ON with 100ms slew setting. Peaked ar 4.7V before decay. Slew rate is approx 80V/s
PS004: Turn ON with 200ms slew setting. More or less the same behavior. Peaked at 4.2V. Slew rate is indeed ~40V/s. SO far I still can't quite figure out what dictates the peak voltage w.r.t slew settings. It neither stays at constant voltage nor constant time.
PS005: Turn OFF. No drama here. Interesting that the slope is linear instead of exponential.
Keithley 2281S
PS011: Turn ON with max (1000V/s) slew. Peaked all the way at 8V before decay. Slew rate is actually ~600V/s. Settling bang on at 1.80V, 400mA. Verified by DMM6500.
PS012: Turn ON with 100V/s slew setting. Peaked at 3.7V. Slew rate is actually ~107V/s. Just like the R&S, peak points are neither at constant voltage nor time. Also note how it settles into 1.7V for a while before "snapping" into 1.8V
PS013: Turn OFF with max (1000V/s) slew. Nothing weird other than the initial drop by ~100mV for 6.2ms before the distinctly stairstep slew down. Not sure how's the slew rate here, the stairstep section is ~1000V/s.
PS014: Turn OFF with 100V/s slew. Similar behavior as before, only scaled up in time. Slew is indeed approx 100V/s in the linear section.
ArdWar:
Like before, load is 45Ω
GWInstek, I think it is PPE-3323 although not really sure.
PS021: Turn ON. (Almost) typical linear PSU behavior. Interesting that CV-CC switchover apparently happen at 200mV. It also sink ~0.5mA when output disabled, not unusual with certain design.
PS022: Turn OFF. Your typical exponential output cap decay. Very slightly accelerated by 0.5mA sink.
Interesting ripple, ~5.6s period :o
Keithley 2461
PS031: Turn ON. Even a SMU have its quirks. Distinct three steps output: initial step into 50mV and stay there for a whooping 125ms, jumps into ~1.3V, before finally jumps into 1.8V. Output is bang on at 40mA.
PS032: Zooming in at the weird bump. Not sure what it is.
PS033: Turn OFF. Clean step, no drama here.
Navigation
[0] Message Index
[#] Next page
[*] Previous page
Go to full version