Looking for power supply with fast current limit regulation

[michaeliv]

Hello,

Does anyone have any good suggestions for a power supply with fast current limit enforcement.
Let's say 1-10 us to settle to +- 20% when limits are set to 10V/100mA and connecting a 10 Ohm load(output needs to drop from 10v to 1v).
The supplies I've tested so far seem to all have a large output capacitor after the current regulator or just have slow regulation.
Cheap China power supplies seem to have a super capacitor on the output and take hundreds of milliseconds to settle.
An HP E3632A takes about 35ms and an Agilent 66319B takes 3.5ms.
The only thing to do well is a Keithley 220 current source which settles to +-20% in under 1us, but it's not exactly a power supply.
Is there any particular reason that it's hard to design current regulation without a capacitor following it ?

Thanks!

[PedroDaGr8]

Hello,

Does anyone have any good suggestions for a power supply with fast current limit enforcement.
Let's say 1-10 us to settle to +- 20% when limits are set to 10V/100mA and connecting a 10 Ohm load(output needs to drop from 10v to 1v).
The supplies I've tested so far seem to all have a large output capacitor after the current regulator or just have slow regulation.
Cheap China power supplies seem to have a super capacitor on the output and take hundreds of milliseconds to settle.
An HP E3632A takes about 35ms and an Agilent 66319B takes 3.5ms.
The only thing to do well is a Keithley 220 current source which settles to +-20% in under 1us, but it's not exactly a power supply.
Is there any particular reason that it's hard to design current regulation without a capacitor following it ?

Thanks!

Not quite your limits but the old HP 6114A nd 6115A takes a maximum 50us to settle to within 10mV of any voltage as long as the current change is within the abilities of the supply. At 10V it should be easily able to hit what you mention.

[kwass]

I'm interested in this too.

Here's a quick read from Keysight on this issue: 

part 1:  http://powersupply.blogs.keysight.com/2014/11/why-does-response-time-of-ocp-vary-on.html
part 2:  http://powersupply.blogs.keysight.com/2014/12/why-does-response-time-of-ocp-vary-on.html


Take a look at their new, low power, E36100 supplies.  Overcurent protect is claimed to be less than 1.5ms response time and transient response less than 50us.   My guess is that it's much faster than this in practice, I'd be very interested to read a review of one of these supplies.

[rodpp]

Maybe the Keithley 2300 series can do that, it's main purpose is to simulate batteries. See below the specs of the 2304A model:


TRANSIENT RESPONSE TO 1000% LOAD CHANGE:


NORMAL MODE:
Transient Recovery Time:
<50µs to within 100mV of previous level.
<100µs to within 20mV of previous level.

ENHANCED MODE:
Transient Recovery Time:
<40µs to within 100mV of previous level.
<80µs to within 20mV of previous level.


Transient Voltage Drop:
<100mV, typical.
<200mV, typical.

Enviado de meu SM-P905M usando Tapatalk

[rodpp]

And here is an application note talking about the load impedance effect in the PS output stability and techniques to avoid oscillations: http://www.tek.com/dl/_2870_Stabilizing_AN.pdf

Enviado de meu SM-P905M usando Tapatalk

[michaeliv]

And here is an application note talking about the load impedance effect in the PS output stability and techniques to avoid oscillations: http://www.tek.com/dl/_2870_Stabilizing_AN.pdf

Thanks, very interesting article.