EEVblog Electronics Community Forum
Products => Test Equipment => Topic started by: FrenchRoast on June 12, 2015, 06:08:02 pm
-
Hi everyone. I hope this is the right sub forum. If not, i am sorry.
My question is about DC power supplies and constant current mode. I fear that i have been fundamentally misunderstanding the meaning of CC mode in every supply i have used. Quite embarrassing.
I recently purchased the Rigol dp832 and while playing with my shiny new toy i found that i could not figure out how to manually set the device into constant current mode. Every supply i had used had a switch or knob to toggle between CV and CC mode.
The Rigol would only jump to it when my set current was less than what my load was requesting. It would then, as expected, clamp the current to the limit and remain in CC mode until the current limit exceeded what the load would draw. This confused me a little.
I thought Constant Current meant literally "Constant Current". As in, the supply would FORCE the set current through the circuit and the voltage would rise to value determined by Ohms law. It then occurred to me, i never actually wanted this behavior and never actually used constant current mode. So i had never tested my assumptions.
My initial though was that Rigol had cut a corner and only implemented a current limiting design. I did some more reading, and found that Rigol's behavior seems to be the standard on each of the other supplies i looked at. I was not able to quickly find a supply that seemed to behave in the way that i had previously "Constant Current" to mean.
So i ask, is this "current limiting" behavior the usual and customary? Have i not understood this common feature all these years?
Thanks for your time everyone.
-
Look to it the other way round...
You set your desired current and limit the max. voltage.
If you want a source as ideal as possible (with this equipment) than just max out the voltage...usualy that's not what you want...
73
-
So i ask, is this "current limiting" behavior the usual and customary? Have i not understood this common feature all these years?
I say "yes". I've never seen a bench supply where the normal behavior is to raise the voltage automatically until it reaches the programmed current.
(you can raise it manually, of course...or some may even have a special mode for it, but not by default)
-
It's nothing more than an application of Ohm's Law (voltage, current and resistance are all interrelated). Any load connected to a power supply will have a certain resistance. The only way to limit the current through a given resistance is by adjusting the voltage.
-
The information i had originally posted here was not correct and was a product of my own confusion. I removed it in case anyone else in my shoes happens to read this thread.
-
This used to be called "automatic crossover CV/CC".
With one setting on the voltage, and another setting on the current, for a given load, the output voltage is automatically set by the servo to achieve whichever setting is more limiting.
For example, set to 10 V and 2 A.
With a 10 ohm resistor, the supply goes to 10 V and 1 A (CV).
With a 2 ohm resistor, the supply goes to 4 V and 2 A (CC).
-
Actually i still didn't understand it.
With a CC setting of 1A. I am telling my supply that i want 1A to flow through my circuit. To achieve this goal, the supply raises the output voltage until either it hits the max output voltage, or 1A flows. That's why when my circuit is open the output voltage goes directly to my maximum allowed setting, in this case, the max voltage.
With a CV setting of 28V, I'm saying i want a 28V differential. To a achieve this goal, the supply will source however much current it needs to in order to maintain my desired differential. Until it hits the max current it can source, or the limit i set.
What confused me is why the display would switch between CV and CC. As TimFox said, it's only telling me what limit the supply hit first.
I think i have got it now. Thanks for explaining this folks.
-
Standard bench type power supplies actually implement what is better called current limiting.
If you set the CC to 1 amp, you are telling the power supply to allow no more than 1 amp through the load. The load can draw whatever it needs - up to that 1 amp limit. If it tries to draw more, then the power supply will reduce the voltage as required to maintain 1 amp maximum. As long as the load draws less current than the limit you have set, the power supply will maintain the voltage at whatever you have set for.
-
With a CC setting of 1A. I am telling my supply that i want 1A to flow through my circuit. To achieve this goal, the supply raises the output voltage until either it hits the max output voltage, or 1A flows. That's why when my circuit is open the output voltage goes directly to my maximum allowed setting, in this case, the max voltage.
That's correct, yes.
BUT ... bench power supplies don't do "CC" they do "CL" - Current Limiting, ie. you set a maximum current, not an enforced current.
CC is usually done by LED drivers and stuff like that.
-
It's current and voltage limiting. The voltage setting sets a voltage limit. The current setting sets a current limit.
Imagine a graph, one axis is voltage, the other is current. you can draw a line representing the set voltage, and another line representing the set current. A simple load like a resistor will have a certain ratio of voltage to current. Starting at the origin, this is a line with some slope. Depending on whether the line first intersects the voltage or the current limit line, that's the one that will be active.
That's for simple current limiting. Some power supplies (usually not bench power supplies) use foldback limiting, which is more complicated.
-
It's current and voltage limiting. The voltage setting sets a voltage limit. The current setting sets a current limit.
Precisely. Bench supplies don't have regulators, they have limiters, which is the same thing, except limiters can only influence the output in one direction (i.e. down).
The I-V-curves of a lab supply looks like these:
(https://upload.wikimedia.org/wikipedia/commons/thumb/6/62/Voltage_regulator_foldback.svg/774px-Voltage_regulator_foldback.svg.png)
Blue is your usual CV/CC supply.
Green is a supply with foldback current limiting, usually found in supplies with high output currents or compact supplies with no space for very large heatsinks
Red is supply with no current limiting or protection (the communist star marks the point where the supply disbands and grants freedom to it's individual components' magic smoke)
-
Standard bench type power supplies actually implement what is better called current limiting.
If you set the CC to 1 amp, you are telling the power supply to allow no more than 1 amp through the load. The load can draw whatever it needs - up to that 1 amp limit. If it tries to draw more, then the power supply will reduce the voltage as required to maintain 1 amp maximum. As long as the load draws less current than the limit you have set, the power supply will maintain the voltage at whatever you have set for.
The usual distinction between CC and CL is that CC is a reasonably accurate control of the current, and CL is only for safety, etc. and is not so accurate. CL can also implement foldback limiting as discussed by another post. Usually, one needs to re-start the power supply after a foldback event. This can be a nuisance when tripped by a surge such as capacitance charging.
Laboratory supplies such as my -hp- 6200-series and Kepco ATE linear analog supplies have similar-quality servos for CV and CC modes, with automatic switchover between modes as required. (Note: some -hp- and Kepco supplies are CL instead of CV/CC.)
A power supply designed for CC operation also has a voltage limit, which may be adjustable, referred to as "compliance voltage". Usually, this is not as accurate as the CC mode and may include a "crowbar" to short the output when the voltage is excessive, requiring re-starting to clear the crowbar.
-
CL can also implement foldback limiting as discussed by another post. Usually, one needs to re-start the power supply after a foldback event.
I think you mean something else. Fold-back limiting just makes the current limit dependent on the output voltage (the lower the voltage the lower the limit). Thus you can draw the full current while the voltage limiter is active, and a voltage-dependent (and thus load resistance dependent) current while the current limiter is active.
Did you mean OCP? Those are usually latching like OVP.
-
Yeah, foldback that requires a reset is just wrong. That's just implemented wrong. Foldback is usually done with active components in a linear manner. Remove the problem that causes the foldback .. and it goes back to normal operation.
OCP (over-current protection) and OVP (over-voltage protection) are great features to have in a bench power supply, and these are usually set above your voltage and current limit settings. They don't usually kick in unless an abnormal event occurs, and then these almost always require a manual reset before normal operations can resume.
Also, not directed at TimFox or dom0, but to everyone reading this, don't confuse foldback current limiting with crowbar circuits. Crowbars are an over-voltage protection that short circuits the output when the over-voltage event occurs, and this causes the current-limiting to kick in due to the short-circuit load. The current limiting may be the basic current limit setpoint, or it may also be a foldback limit depending on the design. Crowbar OVP usually requires a manual reset, although there are some crowbars that blow fuses, and some that auto-reset and crowbar again if the fault is not removed.
-
Yeah, foldback that requires a reset is just wrong. That's just implemented wrong. Foldback is usually done with active components in a linear manner. Remove the problem that causes the foldback .. and it goes back to normal operation.
Unfortunately, this is exactly how it's implemented in at least one variant of the cheap crap Chinese bench PSUs. After you trip the foldback limiting, you have to toggle the PSU mains switch to reset it, an unbelievably brain dead bit of design.
-
A caveat: some supplies have an output filter capacitor beyond the current limit stage. If you're not careful, this capacitor can contain enough energy to destroy the load.
For example, say you have your power supply set to 10mA "CC", the peak voltage at 12V and a LED connected to the output. The LED accidentally becomes disconnected momentarily. The output capacitor charges to 12V and when the LED is reconnected it discharges into the LED, destroying it.
On the HP 6200 supplies you can disconnect the 80-400uF output capacitor, which leaves in a small 15-20uF output capacitor in place for loop stability. Dave found the Rigol has a fairly large 1000uF output capacitor in his tear-down vid.
Design idea: an entirely external & independent OV / OC protector so you don't have to know and trust the internal workings of your supply.
-
The foldback on CL that I have seen changes the current setpoint depending on the actual output voltage. Therefore, if you short the output, the current limit is greatly reduced. This minimizes the power dissipation on the pass element when the load is shorted. Since this is a positive feedback, it can latch up. (I have seen that on linear component supplies, not bench supplies.)