Simple, Fully Adjustable CV/CC Regulator Design

Thread Starter

crutschow

Joined Mar 14, 2008
38,945
While thinking about a problem from another post, I had a small "Eureka" moment, and came up with a relatively simple design for a bench-type, constant-voltage/constant-current power supply circuit that allows independent adjustment of both voltage and current from zero to max.
This makes it particularly useful for tasks like charging various battery types, sizes, and voltages, most of which require a specific CC/CV charging profile.
Not sure how unique it is, but generally circuits with both fully adjustable voltage and current tend to be more complex, and I don't see how this could be made much simpler, so I thought I'd post it for anyone interested (below).

It uses a Sziklai-pair buffered Howland current-pump circuit for the CC function (U1-A), and an op amp (U1-B) driving the Sziklai-pair as emitter-follower circuit for the CV function.
D1 allows the CV circuit to take control when the current is below the current setpoint.
The Howland circuit allows the use of a low value, high-side shunt resistor for current sensing (R5 below).
It has a low drop-out voltage of about 1V at maximum current.

The voltage and current control are both referenced to ground for the TL431, 2.5V Ref voltage.
The ground reference means they both could also be readily controlled from the DAC output of a micro in place of the pots.

The circuit below is configured for a maximum output of 15V and 2A current from a 16V supply.
The sim shows the output voltage and current for current settings of 0, 1A, and 2A, with a voltage setting of 7.5V.
Note how it goes from CC mode to CV mode as the output load goes from 1 to 11 ohms.

The op amps do need to be rail-rail types.

For more detail on the design, I posted a description here.

1794930731600.png
 
Last edited:
i like it. very simple and since Iset and Vset are 0-2.5V, they are easy to control by DAC. it is worth noting that TLV9152 is the part on the edge of its specs if the goal is to scale this up to some higher voltage consider different chip.
OPA2991 is virtually identical (exact match) but goes up to 40V. there is also budget match TLV9352 with only marginally worse offset/drift.
 

Thread Starter

crutschow

Joined Mar 14, 2008
38,945
OPA2991 is virtually identical (exact match) but goes up to 40V. there is also budget match TLV9352 with only marginally worse offset/drift.
Good point.
The TLV9152 has max rating of 20V, but I was looking for a low cost RR op amp with a higher voltage rating and the TLV9352 looks good.
 

panic mode

Joined Oct 10, 2011
5,459
ah, taking requests? all right, make it 5A, 30V and with buck as tracking preregulator so transistor needs only tiny heatsink.
 
Last edited:

Thread Starter

crutschow

Joined Mar 14, 2008
38,945
Question. How sensitive is this circuit to changes in temperature?
Should be relatively insensitive.
The main change should be the temperature coefficient of the TL431's reference voltage that controls the output voltage and current settings.

Why the concern for that?
 
Top