The circuit shown in Fig was devised as a cheap alternative to a moving
coil meter when I wanted to monitor the output current of a power
supply. Its operation is as follows: As the output current increases
from zero, it flows through l.e.d. D1 and resistors R1 and R2 until the
voltage across R2 is about 0·6V. At this point transistor TR1 starts to
conduct, shunting current round D1 and through D2/R3. When this current
produces 0·6V across R3, TR2 begins to conduct, shunting the extra
current via D3/R4. This continues until finally D5 is illuminated,
showing “full scale deflection”. The purpose of resistor R1 is to
provide sufficient voltage across the circuit to cater for:
- 0·6V across the 33ohms resistor
- The voltage across the l.e.d.
- 0·3V across the transistor in saturation (this is also the voltage across R1)
Bargraph using LED's |
The
minimum voltage required across the circuit is about 3V, and this must
be taken into account when considering the power supply voltage, which
must be connected before any voltage regulator.
Using the resistor values shown, the maximum current for each l.e.d. is
about 17mA, so “full scale deflection” is about 85mA, though the
display is not perfectly linear – the second l.e.d. starts to illuminate
above about 15mA, and the other l.e.d.s start in approximately20mA
increments.
I tried building a display using 10 l.e.d.s but the circuit was unstable; as soon as the 6th l.e.d. started to illuminate, the circuit oscillated in the Megahertz region, and the l.e.d.s lit before they were supposed to (this situation was worse when I left the decoupling capacitors out), so five l.e.d.s. seems to be the limit.
I tried building a display using 10 l.e.d.s but the circuit was unstable; as soon as the 6th l.e.d. started to illuminate, the circuit oscillated in the Megahertz region, and the l.e.d.s lit before they were supposed to (this situation was worse when I left the decoupling capacitors out), so five l.e.d.s. seems to be the limit.
No comments:
Post a Comment