Connect the diode in series with the negative terminal of the power supply to simplify current monitoring

As defined by the diode formula IF?I0×exp (eVF/kT), the voltage on the diode rises continuously with the logarithmic current flowing through it. Where IF is the forward current, IO is the reverse saturation current, e is the charge (1.602×10-19 C), VF is the forward voltage, T is the temperature (K), and k is the Boltzmann constant (1.380×10-23) J/K).
According to our purpose, the following formula can be extracted from it:
VF∝logIF (at a given temperature)
shunt diode
Now, let us look at a diode with a measuring instrument. When the current is very low, it will indicate the milliampere (mA) level current flowing through the meter instead of the diode; and when the current is large, it will display the voltage on the diode and the logarithm of the current derived from it (the diode Think of it as a self-adjusting shunt). Therefore, the bottom of the meter scale is quite linear, the top has sufficient logarithmic nature, and the middle is the transition phase, so the entire range is very useful.
As shown in Figure 1, the use of a Schottky rectifier, a 100μA/1.7kΩ meter and a suitable series resistance can monitor currents from 10μA to more than 100mA in a single range, and the indicated speed is limited to the pendulum of the meter. speed.




Connecting a suitable diode in series with the negative terminal of the power supply can achieve the purpose of simplifying current monitoring
Figure 1 A Schottky rectifier, a 100μA/1.7kΩ meter and a suitable series resistor.
This kind of simple circuit often has more problems than the number of components! In addition to the need for a high-precision calibration process, this circuit has two major drawbacks: series voltage drop and temperature stability. The voltage drop of the diode is as high as 400mV, so use a new or fully charged battery when monitoring, otherwise your UUT may show that the battery is low. Or think of this circuit as a convenient low-voltage detection test circuit, so you might want to add a short-circuit switch.
Add additional diodes
At the bottom of the scale, almost all current flows through the meter, which is limited by the mechanical and magnetic temperature coefficient of the meter's measuring mechanism, and the measured temperature coefficient is very low. But at higher currents, we will see a voltage drop across the diode. Of course, as the diode formula predicts, this voltage drop will drop at a rate of about 2mV/K. This not only affects the slope of the law of logarithm (low of logarithm), but also affects the transition point from linear to logarithm. In addition, the instrument winding accounts for a large part of the total series resistance, and the TCR of copper at room temperature is 3930ppm/K. Figure 2 shows the deviation vs. current curves of 1N5817 at 0℃, 25℃ and 50℃ respectively. These curves take into account the TCR of the measuring circuit and the temperature coefficient of the diode, but ignore any self-heating effects of the latter, making it more stable There is no problem under the temperature conditions.

Connecting a suitable diode in series with the negative terminal of the power supply can achieve the purpose of simplifying current monitoring

Figure 2 Deviation and current curve.
The self-heating mainly present in D1 is actually not a problem. Assuming that the current flowing is 100mA, the voltage drop of D1 is 400mV: that is 40mW. According to the data manual, the basic thermal resistance of D0-41 1N5815 with a slightly longer pin and a large number of heat sink copper fins is 50K/W. Taking these data into account together, the temperature rise of the junction is only 2° at 100mA, which is equivalent to a VF drop of about 4mV, or an error of about 1% at full scale. Try to keep the diode with short pins and high thermal mass. Note that there may be high transient currents during turn-on, as these will cause errors until the junction temperature cools down again.

Figure 3 adds an extra diode in series with the meter measurement circuit, which is an improved version to offset the temperature coefficient; Figure 4 shows the curve of this circuit. Note that most of the curve is now in logarithmic form, and that additional diode effectively suppresses the initial linear region. However, the choice of this diode is quite critical, because the forward voltage of D2 should be slightly lower than that of D1, but other characteristics should match, which is a bit confusing.
Connecting a suitable diode in series with the negative terminal of the power supply can achieve the purpose of simplifying current monitoring
Figure 3 An improved version after adding a diode.

Connecting a suitable diode in series with the negative terminal of the power supply can achieve the purpose of simplifying current monitoring
Figure 4 The deviation and current curve after adding a diode.
The role played by LTspice
LTspice is here to save us! I was fortunate enough to come across the combination of D1 using 10MQ060N and D2 using BAT54-this is the component I simulated. Both are cheap, modeled by LTspice, and are therefore recommended components. A pair of 10MQ060N works almost the same (but a pair of BAT54 is not consistent). Combinations with other components will show worse temperature changes and strange indications most of the time, so modeling is required before building the circuit. If the sensitivity and resistance of the meter are appropriate, R1 can be omitted. The thermal performance of D1 and D2 must be the same so that they can track changes in temperature with each other.
Silicon P-N junction diodes generally have a very straight line (log IF)/VF relationship, while Schottky’s straight line is not straight. This is because their structure itself has a higher series resistance. At very low currents, the relationship between the two is closer to linear rather than logarithmic, and there is also a guard ring to control the formation of a PN diode in parallel with the Schottky junction. Therefore, in actual use, the logarithm law will change with the current and component type.
Although for this version, a used diode may be enough, but in view of the inevitable inadequacies of this circuit, the dual diode design still needs to be carefully selected. And Schottky diodes can provide more background.
Because I have a box of cheap 100μA/1700Ω indicators from the past, which fits the 35mm×14mm aperture, I use them. Such indicators are very common, very tightly connected, and very practical, and their construction, linearity and consistency between units are also good.
The calibration points used in Figure 5 are generated by arranging the monitor, battery, fixed and variable resistors, and a series combination of DMM. The existing test scales are marked at appropriate points, then eliminated and scanned, and the scan is used as a template for the final layout. The simulation results are used to generate the reference points in Figure 5 (left), and the results reflect the implementation very well, even though the instrumentation is poor. These scales can save time, but unlike the new ones (obviously these measurement structures require different scales), you can fine-tune the calibration by adjusting R1 (the meter is specified as ±20%), and both scales take into account the non-linearity of the meter structure .
Connecting a suitable diode in series with the negative terminal of the power supply can achieve the purpose of simplifying current monitoring
Calibration points derived from the combination of monitor, battery, fixed and variable resistors, and DMM (right). Although the gauges are not very good, the reference point (left) reflects the actual situation well.
Note, I call this "monitor" instead of "meter", the latter term should have better accuracy for me. In any case, I now embed these circuits into most of my development projects and even production test devices. They are very effective for finding various faults and problems, including short circuits on power lines to incorrectly coded pull-up pins.
In order to facilitate the monitoring of the current, just connect a suitable diode in series with the negative terminal of the power supply and monitor its forward voltage drop. After some simple calibrations, you can monitor the supply current in complete synchronization with other parameters you want to detect.