An assortment of data and observations regarding electronics metrology and precision analog electronics.
Click on images to view full size.
LMC662 integrator with AVX MLO
3458A input charge injection
Low noise input protection
ADR1399 output voltage deviation vs. heater voltage
ADR1399 output voltage noise vs. reference current
ADR1399 unheated tempco
VRE102CA voltage reference noise in batch of 8
LTZ1000 unheated TC
Solartron 7081 warmup with 10V input, comparison of 2 firmwares
Solartron 7081 linearity
Keithley DMM6500 linearity
Keithley DMM6500 leakage
MOSFET Idss leakage with 0V gate at 21°C
OPA140 CMRR
LT3042 & LT3093 low frequency noise
G3VM-41GR6 SSR off-leakage
NSD
Keithley DMM6500 linearity #2
Nanovoltmeter
Noise measurements
The ADR1399 datasheet discusses the improved output dynamic impedance for reducing the change of output voltage vs. zener supply current. What about the heater supply voltage?
With heater+ at 15V, and heater- swept from 0V to -15V (to give 15-30V heater voltage range), output voltage coefficient is 0.37 ppm/V. Heater+ is fixed at 15V as this also supplies the zener current (through 2.35k Rshunt).
It is known that the internal schematic of LM399 means that as long as a minimum current condition is met, there should not be any change in output voltage noise for higher currents as the actual zener current is fixed. This is confirmation that the ADR1399 is the same. 1mA is below the allowable range of reference current, it is shown here as a point of interest.
7.8 PPM/degC, linear for 15 to 29 degC range.
Data from eevblog shows that for the LM399, the unheated tempco does vary significantly from device to device so this measurement of one ADR1399 is mainly for curiosity as it's likely to have similar behaviour.
Powered from +/-14V supply, with positive reference output going into 0.1 - 10Hz LNA.
Uncorrected TC of LTZ1000 = 33-37 PPM/C. Corrected TC with 17 ohm resistor = around 0 PPM/C.
Top line in each series is the heat up (using internal heater), bottom line is the cool down.
MickleT has provided a modified Solartron 7081 firmware which resolves an issue with post-mux switching dwell time, to allow internal circuits to settle before taking zero reading.
Using Marco Reps' Fluke calibrator and 3458A to test INL of S7081.
DMM6500 set to 1NPLC, high impedance input, autozero enabled. Per step; 80 seconds to allow the PWM DAC output to settle, approximately 1 minute of samples acquired and averaged to give a single value for the chart.
Connected a lead from the PE screw on the rear panel, to the current input, digitize mode with 100uA range. 140nApp. Lead is unplugged halfway through the trace.
Using Keithley 617 electrometer, with +/-100V DAC output, to sweep the drain pin of MOSFETs from 0.1V to maxV, gate pin and source pin shorted together and wired into electrometer input. FOM is (10V/(datasheet RDS))/(leakage at 10V) to provide a quick comparison point.
OPA140 configured as unity gain buffer. DMM measuring input-to-output. Measured CMRR = 142dB.
22uF Cset. Datasheet hides the 0.1-10 Hz region for good reason...
Using Keithley 617 electrometer, with +/-100V DAC output, to sweep a pin of the SSR from 0.1V to 40V. Other pin wired to electrometer input.
Low frequency NSD estimation of various op-amps and ADCs.
Linearity sweep repeated using ultra linear DAC & 3458A confirmation
Improves CMRR and other similar effects.
OPA140
OPA2140
OPA202
OPA2202
OPA210
OPA2210
OPA189
LTC2057
LMC662
JFE2140
THP210
TMUX6136
TMUX7234
TMUX1134
DG419
NOMCT16031002
2N7002H
MMBT3904
MMBT3906
NSVJ3910
BAV199
SN74LV4053A (TI)