Four Channel Strain Condtitioner
The 5078 is a four channel programmable-gain differential-input
amplifier that provides signal conditioning for bridge type
transducers. In addition to analog signal amplification each
independent channel includes bridge completion, input offset
bias compensation, voltage excitation with remote sense capability,
local or remote shunt calibration, digital signal processing
with programmable filtering and output buffering.
The 5078 utilizes onboard DSPs (one per channel) to configure
the input circuitry and handle amplifier gain and offset compensation.
High performance front end analog components are combined with
digital signal processing techniques and ultra stable calibration
reference sources to maintain system accuracy, bandwidth and
gain tolerances. Front-panel LEDs indicate signal presence and
activity level, and warn of module operational problems. All
circuitry is housed in a shielded enclosure for improved reliability
and noise reduction.
On-board bridge-completion resistors are provided for full, ½, or ¼-bridge completion
to accommodate a variety of sensor configurations. Bridge configuration is completely
programmable, including the selection of 120 or 350 ohm completion resistors. By setting
the configuration to the Full Bridge mode, the 5078 can also be used as a standard differential
When CAL is activated, a pre-selected shunt cal resistor is electrically connected to
the sensor to simulate a known amount of strain. Selecting Local shunt cal makes this
connection inside the amplifier for simpler sensor wiring. Selecting Remote shunt cal
makes this connection at the sensor (using separate shunt cal lines). Using Remote shunt
cal also allows a user-supplied external shunt cal resistor to be used if desired.
DSP - Programmable Gain
The differential bridge output is fed to an instrumentation grade x10 pre-amp avoiding
the use of switched resistors in the most noise and temperature-sensitive portion of the
circuit. A programmable offset voltage is injected following this preamp for input offset
compensation and automated bridge balancing. After offset correction, the signal is fed to
a variable gain amplifier controlled by an onboard DSP prior to digitization and subsequent
processing. The Digital Signal Processor uses stored offset and gain calibration factors to
correct the digitized data values and generate a digitally filtered output that is ready for
conversion back into an error-corrected analog output voltage.
The processed digital output is converted back to an analog voltage by a high-speed 16-bit
Digital-to-Analog Converter. A four-pole low-pass filter/buffer-amp removes the digitizing
steps in the reconstructed signal, along with any high-frequency noise.Temperature-stable
components and ultra stable references are used to ensure that system calibration holds
over a wide temperature range.
DSP – AutoBalance/AutoGain
Using a programmable offset voltage allows the DSP to automatically servo the amplifier
output to zero adjusting for bridge imbalance. In a like manner, the DSP can adjust the
amplifier gain to a targeted value upon application of a shunt resistance or other calibration
Voltage Excitation with Remote Sense Capability
The 5078 provides four independent programmable excitation voltage sources for use in powering
input sensors. Each source is controlled by a 16 bit DAC with an output range from 0 to 10 Volts
and 100 mA output buffering. Remote sense can also be selected to provide feedback accounting for
IR losses over conductor lines between amplifier and sensor.
High accuracy is obtained during the conversion process by implementing a unique end-to-end
calibration scheme within the 5078 conditioner. A precision programmable voltage source is
connected to the input, and two calibration voltages (0v and 80% of full-scale input) are fed in,
amplified by the input stages, converted by the A/D, processed by the DSP, converted back to analog
and then measured by a high-accuracy 24-bit A/D converter. The measured input and output voltages
are then used to compute offset and gain correction values. During operation, their correction factors
are applied to each data point in real time resulting in system accuracy better than ±0.05 of full