- Discharge voltage
- 0.1 to 20 V on CH906A; 0.1 to 10 V on CH906D
- Discharge current
- 0 to 10 A on CH906A; 0 to 2 A on CH906D
- Charge voltage
- 0.2 to 20 V on CH906A; 0.2 to 10 V on CH906D
- Charge current
- Up to 10 A below 10 V and 5 A above on CH906A; up to 2 A on CH906D
- Setting resolution
- 1 mV up to 5 V and 5 mV above on CH906A; 1 mV on CH906D; 1 mA on current
- Current readback resolution
- 0.0001 mA on the low range; 0.1 mA on the high range
- Voltage accuracy
- 0.05% + 2 mV up to 5 V; 0.05% + 10 mV above 5 V on CH906A
- Ripple
- 60 mV maximum
- Display
- 4.3-inch TFT touchscreen with keypad and knob
- Body
- 230 mm deep × 350 mm wide × 88 mm high
A product that charges and discharges is tested from both ends, and the change from one to the other is where the interesting behaviour sits. A simulator that has to be switched between the two modes cannot show what happens at the crossing. This range is built so that the crossing happens by itself.
Product overview
Two models share one control system and one body. CH906A reaches 20 V and 10 A on discharge; CH906D reaches 10 V and 2 A. Both present the same behaviour at the terminals: they hold a set voltage and pass current in either direction, and the changeover between charging and discharging is automatic.
The output stage is linear, which is the reason the ripple figure is as low as it is and also the reason the larger model weighs about seven kilograms. The front panel carries a 4.3-inch touchscreen, a numeric keypad, a knob and function keys, so a value can be typed or dialled. The rear carries the remote sense terminals, the serial port and the voltage selector.
How the battery simulator works
The instrument regulates its output to a set voltage and measures the current that flows. When the device under test draws current, the instrument supplies it and the device is being discharged. When the device returns current, the instrument absorbs it and the device is being charged. There is no mode to select, and no discontinuity at the point where the current passes through zero.
That matters for a product that manages its own charging. A phone, an earphone or a power bank does not simply stop drawing current at a fixed point; it negotiates, pulses and backs off. An instrument that holds a voltage and follows the current in both directions reproduces what a cell would do, without the cell’s own state of charge interfering with the test.
The remote sense connection is what keeps the delivered voltage honest. The sense leads measure at the product, and the regulator corrects for the drop in the current leads, which on a 10 A output is not negligible.
How to select the right battery simulator
The voltage window decides the model, and the window has two ends. A 12 V product needs CH906A because its charge voltage sits above 10 V; a single-cell product is covered by either, and CH906D is the smaller and lighter choice.
The current decides the rest. A product that draws 3 A in operation and returns 1 A while charging needs headroom on the discharge side, and the charge current limit matters if the test is meant to reproduce a fast charge rather than a slow one.
The third question is whether the test is driven from a computer. Both models carry an isolated RS232C port with SCPI commands, which allows the panel to be bypassed entirely, and that is the difference between a bench instrument and a component in a test system. If the sequence is written in software, the command set should be reviewed before the order rather than after.
The fourth question is the mains supply at the site. The instrument is ordered for one voltage or the other, not for a wide range, and the marking beside the inlet should be read before it is connected. Send the voltage window, the current in both directions and the intended control method, and a model can be proposed.
Measurement limitations and selection considerations
The voltage resolution and accuracy are specified in two bands, and the boundary sits at 5 V. On CH906A the resolution relaxes from 1 mV to 5 mV above that point, and the accuracy term grows with it. A test that depends on millivolt-level resolution at 12 V is therefore asking for something the printed specification does not offer.
The charge current limits are printed in more than one form, and they do not agree at the boundary. Until the supplier confirms which applies, the conservative reading is the one that limits current above 10 V, and a test plan should not be built around the higher figure.
Ripple is quoted as a maximum without a test condition or a bandwidth. That is enough to compare this range with another instrument quoted the same way, and not enough to predict the noise at a particular load; a bench that needs that figure should measure it in place.
The environmental limits are absent from the material. Operating temperature and humidity are not stated for either model, so a bench in an uncontrolled space should have those confirmed before the instrument is installed rather than assumed from the laboratory conditions the accuracy figures were probably taken in.
Every value below is taken from source documentation. Where a parameter is not stated in the available material it is marked as information required rather than estimated.
