Learning About the UCC283 range of LDO voltage regulators
: Description : Notes on how to use the UCC283-5 and UCC283-3 low drop out (LDO) voltage regulators
: Status : Research—don't take my word for it.
Product Info
Introduction
The UCC283-X range of voltage regulators from Texas Instruments is a selection of fixed and adjustable regulators featuring high-current with low drop out operation. The input voltage can be as high as 9V and down to under—possibly well under—only 0.6V above the output voltage—the exact minimum input voltage depends on the current needed.
Other features mentioned in the datasheet include:
- "reverse voltage sensing that prevents current in the reverse direction"
- "short-circuit current is internally limited"
According the the datasheet, the difference between the UCC283-X and UCC383-X parts is that the "UCC283 series is specified for operation over the industrial range of −40°C to 85°C, and the UCC383 series is specified from 0°C to 70°C."
The parts are available in the through-hole, protoboard and breadboard friendly TO−220 package style.
Connections (UCC283-5)
VOUT : Regulated output voltage. A bypass capacitor is not required at VOUT, but may be desired for good transient response. The bypass capacitor must not exceed a maximum value in order to insure the regulator can start. GND : Reference ground. VIN : Input voltage, This pin must be bypassed with a low ESL/ESR 1-μF or larger capacitor to GND. VIN can range from (VOUT + VDROPOUT) to 9 V.
Apparently ceramic and tantalum capacitors have low ESR, suitable for the input capacitor. Aluminium electrolytic capacitors have a high ESR and are not suitable.
While the datasheet doesn't seem to specify it, the product page states the output capacitor should be tantalum.
There is a calculation for the maximum value of the bypass capacitor—good luck. :-)
Here's a typical application circuit from the datasheet:
Other information
Even though the UCC383 family of linear regulators are not optimized for fast transient applications (Refer to the UC182 Fast LDO Linear Regulator), they do offer significant power supply rejection at lower frequencies. ... The performance can be improved with additional filtering.
Progress
- I've successfully created a test circuit and driven the 5V regulator with both a 9V battery and 4 x AA batteries (not at the same time!).
- Successfully created a test circuit with a 5V and a 3.3V regulator in a chain to power two LEDs—one off 5V and the other off 3.3V. I have also powered an ATMega8 circuit off the 5V supply at the same time.
- The above circuits used 10uF tantalum capacitors for the input and output capacitors. The first circuit used two of the capacitors, the second circuit used three of the capacitors.
- The above circuits were all created on a standard breadboard.
- Powered the WIZnet WIZ810MJ module from the 3.3V supply while module was plugged into an Ethernet switch—communication untested however.
- Powered Arduino Diecimila from 5V output, WIZnet WIZ810MJ module from 3.3V and it all worked enough to ping the module. Has a fairly negative impact on the charge of the (non-alkaline) AA batteries I was using to test it with though. :)
@@ TODO : Document (breadboard layout and schematic) circuits used.
Dual voltage breadboard test circuit
I created a breadboard test circuit utilising an UCC283-5 (5 volt) voltage regulator and an UCC283-3 (3.3 volt) voltage regulator to power two LEDs.
Here is an overview of the complete circuit showing the regulators, filtering capacitors and the LEDs with their associated current limiting resistors (note the two different values):
Unfortunately the point at which power enters the circuit isn't very clear as I didn't have power connected when I took these photos—but on the breadboard, in the lower right corner, the positive source voltage (up to a maximum of 9 volts) is connected to point (j-63) and the source ground connects to the (obscured) point (j-62).
The left power rail (with the brown jumper wire) is +3.3V and the right power rail (with the red jumper wire) is +5V.
The 3.3V voltage regulator connects to the output of the 5V regulator in an attempt to reduce the wasted power—it could be connected to the source supply directly but my understanding is this would result in a waste of (for a 9V source):
Waste = (9-5) * 2 + (5-3.3) = 9.7V * I
rather than:
Waste = (9-5) + (5-3.3) = 5.7V * I
(Okay, so I really need to include current (I) in there for total power dissipated, but we'll ignore that currently. Heh. @@ TODO : Do this properly.) ( Note : It's been pointed out to me that because the 5V regulator will now need to pass the current required by the 3.3V regulator all I've done is moved the power dissipation from the 3.3V regulator to the 5V regulator which could create an issue. There's a handful of posts about using two, dual, series, or cascading voltage regulators: cascade voltage regulators?, cascaded voltage regulators?, Cascade two voltage regs.)
Here's a closer view of the regulators from the front—the source input points are badly marked with the red circle (+5V) and the black circle (Gnd): (Yeah, okay, so I should've just reshot the image with the wire in it...)
Note that the other two capacitors are obscured in the above view—but the visible capacitor has its positive lead leftmost.
Another angle shows the three capacitors and the polarity on the right-most one: