Oxygen sensor test

A full test experiment to access instrumental performance is described in the section Quality control of the OroboPOS (O2k-SOP) in MiPNet06.03 POS-Calibration-SOP.

Such a quality test should be done

• whenever a problem is encountered
• when the performance of the instrument is doubtful
• from time to time for Quality Assurance
• after a sensor service or after applying new membranes
• during the troubleshooting procedure by switching components a sensor test is done after each switching step.

For the purpose of quickly accessing instrumental performance during trouble shooting (e.g. when several test experiments have to be run in sequence with intermediate switching of components) the determination of the instrumental background parameters may be omitted. For communication with Oroboros Instruments during trouble shooting is is helpful to perform the test experiment with very well defined parameters (temperature, medium,...) to allow us to asses the performance parameters in a comparable way.

After a sensor service or after changing membranes the oxygraph should run (i.e. the oxygraph, stirring, etc is switched on) over night (chambers with EtOH/water or just water) before a sensor test is performed or other experiments are started. This ensures proper stabilization of the signal.

How to do an abbreviated sensor test for the purpose of trouble shooting (omitting the instrumental background):

Note: If for one of the described phases the required stability can not be reached, continue the sensor test with the next step.

• select "Z Troubleshooting" from the Layout menu to observe the raw signal
• fill O2k-Chamber with water
• in the 'Oxygraph' / 'O2k Control' window
  • set T = 37 °C
  • set gain = 2
  • set stirring = 750 rpm
• in the 'Experiment' / 'Edit' window
  • set 'Flux derivation N' to 40 points for both chambers
• wait for thermal equilibration (watch for Peltier power % to get stable)
• with open chamber wait for stable signal
• do an air calibration (open chamber, stable signal = zero flux)
• close the chambers: stable, non-zero flux
• do a stirrer test (switch off stirrer for aprox. 30sec, wait until the oxygen concentration signal flattens out, switch on stirrer, wait)
• do a zero calibration (dissolve a few crumbs of dithionite (= zero solution powder) in a drop of water, inject immediately 50 µl into the closed chamber, wait until most of the drop in oxygen concentration or raw voltage has occurred and then inject another portion to make sure that zero oxygen has really been reached. No stepwise drop should occur after the second injection otherwise oxygen has not been zero after the first injection. A second important criterion is to observe the (red) flux signal. At zero oxygen plus an excess of dithionite that consumes any oxygen diffusing into the solution the flux should be zero. However, especially with a rather slow sensor it may take some time to reach zero flux. The flux will in this case be positive and decreasing. It is not necessary in this case to wait for a complete zero flux for a simple sensor test. A negative flux on the other side strongly indicates that zero oxygen has not bee reached jet (all the ditionite has been consumed by oxygen, therefore the oxygen back diffusion is visible as a negative flux). In this case another injection of dithionite is necessary.

A DatLab demo file showing a typical sensor test (from a working system) can be downloaded with this link from the OROBOROS Homepage.

Evaluation of a Sensor Test

Most parameters are best viewed by observing the Raw Signal. It will be helpful to compare the obtained test with the test of a functioning system, e.g. from the DatLab demo file.

Informations gained by the sensor test include:

Raw Signal at air saturation

Signal stability at air saturation

Flux at open chamber

Response time of the POS

Flux at closed chamber near to air saturation

Raw signal at zero oxygen (zero current)

Signal stability at zero oxygen

Consequences of a Sensor test

Hardware problems are detected by assessing the parameters listed above in 'Evaluation of a Sensor test'.

If a hardware problem was observed in a 'stand alone' sensor test, please follow the link for the problematic parameter. Frequently it will be necessary to find the faulty component by locating a problem.

If a sensor test was already performed as part of locating a problem the information in which chamber of the oxygraph the problem appeared will be used in the further process of locating a problem.