Changes

The most basic method of measuring the power usage of a device is doing so by measuring the single-ended voltage across a resistor. Single-ended means we have a probe connected to one side of the resistor, and a common ground connection. But besides just measuring the drop across the resistor, you will also measure any ''variation'' in the power supply. This could be due to the regulator, environmental noise, or voltage variations from other areas of the circuit switching.

Rather than use a single-ended probe, a differential probe ignores that variation in voltage that is common to both sides of the resistor, also called ''common mode'' voltage. The measurement is now ''only'' the voltage drop across the resistor.

The following diagram shows the key connections for the differential probe:

[[File:diffprobe-top-out.jpg|image]]

* Can power from single or double-ended supply.

* 10x gain.

The supply voltages of the differential probe must be at least 2V above/below (for +V & -V respectively) the common mode voltage. The Probe Power Supply (CW503) provides a +/- 8V supply, allowing you to use the probe on any reasonable shunt inserted into VCC and GND.

[[File:pinout.jpg|image]]

Using a single-ended supply means connecting the -V input to the GND pad. This is done via a jumper mounted as shown on the 6-pin connector:

Note again the common-mode limits still apply. This means the common-mode voltage must be at least +2V since the -V supply is 0V. Thus you can '''only use the single-ended power supply for a VCC shunt'''.

The common-mode reference point can be used to connect your system ground to the probe ground. Note you need to ensure the common-mode input voltage is with-in the allowed range (e.g. at least 2V away from either supply voltage). If you are powering the differential probe with another power supply, you must make sure they are referenced to each other by connecting the 0V/GND points together. Using a resistor reduces a small differential from causing a large current to flow if you accidently have ground loops (multiple connections of system grounds).

When using the probe, there will often be a constant voltage we wish to null out. If measuring the current signature of a system, there will be a constant current usage we don't care about. Without a null adjustment, this constant current would cause the output of the differential probe to stick at a rail or otherwise clip the output.

When the null adjustment is set, the constant output voltage from the output of the differential probe should be half-way between the -V and +V supply. For example if using symmetrical supplies it should be around 0V, if using single-ended supplies it should be half of the +V voltage.

Due to the null-offset adjustment, there is a lower frequency response of around 20 kHz. There is a gain roll-off below that point, meaning your gain at DC will be different (around 3-5x) than your gain at frequencies above 20 kHz. For applications such as measuring USB inrush current across a shunt you need the DC gain to be the same as the higher-frequency gain.

# Solder SJ1 to GND (which should short capacitor C5).

A number of resistor pads are present around the input. These can be mounted for a variety of options, most noteably if you mount resistor R6 it can serve as a shunt resistor.

[[File:diffinputs.png|image]]

Note there is an older version of the hardware shown below. This version does not have the feedback LEDs for nulling the offset, or the shunt resistor mounting options. This is otherwise equivalent to the newer revision.

[[File:Cw501_schematic.png|image]]

{{Template:Hardware}}