A frequently encountered category of analog system component is the precision current source. Many good designs are available, but concise and simple arithmetic for choosing the component values necessary to tailor them to specific applications isn’t always provided. I guess some designers feel such tedious details are just too trivially obvious to merit mentioning. But I sometimes don’t feel that.
Wow the engineering world with your unique design: Design Ideas Submission Guide
Here are some examples I think some folks might find useful. I hope they won’t feel too terribly obvious, trivial, or tedious.
The circuit in Figure 1 is versatile and capable of high performance.
Figure 1 A simple high-accuracy current source that can source current with better than 1% accuracy.
With suitable component choices, this circuit can: source current with better than 1% accuracy and have Q1 drain currents ranging from < 1mA to > 10 A, while working with power supply voltages (Vps) from < 5V to > 100 V.
Here are some helpful hints for resistor values, resistor wattages, and safety zener D1. First note
- Vps = power supply voltage
- R1(W), Q1(W), and R2(W) = respective component power dissipation
- Id = Q1 drain current in amps
Adequate heat sinking for Q1(W). Another thing assumed is:
Vps > Q1 (Vgs ON voltage) + 1.24 + R1*100µA
The design equations are as follows:
- R1 = (Vps – 1.24)/1mA
- R1(W) = R1/1E6
- Q1(W) = (Vps – Vload – 1.24)*Id
- R2 = 1.24/Id
- R2(W) = 1.24 Id
- R2 precision 1% or better at the temperature produced by #5 heat dissipation
- D1 is needed only if Vps > 15V
Figure 2 substitutes an N-channel MOSFET for Figure 1’s Q1 and an anode-referenced 431 regulator chip in place of the cathode-referenced 4041 to produce a very similar current sink. Its design equations are identical.
Figure 2 A simple, high-accuracy current sink uses identical design math.
Okay, okay, I can almost hear the (very reasonable) objection that, for these simple circuits, the design math really was pretty much tedious, trivial, and obvious.
So I’ll finish with a very less obvious and more creative example from frequent contributor Christopher Paul’s DI “Precision, voltage-compliant current source.”
Taking parts parameters from Christopher Paul’s Figure 3, we can define:
- Vs = chosen voltage across the R3R4 divider
- V5 = voltage across R5
- Id = chosen application-specific M1 drain current
Then:
- Vs = 5V
- V5 = 5V – 0.65V = 4.35V
- R5 = 4.35V/150µA = 30kΩ
- I4 = Id – 290µA
- R3 = 1.24/I4
- R4 = (Vs – 1.24)/I4 = 3.76/I4
- R3(W) = 1.24 I4
- R4(W) = 3.76 I4
- M1(W) = Id(Vs – Vd)
For example, if Id = 50 mA and Vps = 15 V, then:
- I4 = 49.7 mA
- R5 = 30 kΩ
- R4 = 75.7 Ω
- R3 = 25.2 Ω
- R3(W) = 1.24 I4 = 100 mW
- R4(W) = 3.76 I4 = 200 mW
- M1(W) = 500 mW
Stephen Woodward’s relationship with EDN’s DI column goes back quite a long way. Over 100 submissions have been accepted since his first contribution back in 1974.
Related Content