YSI ADV6600 - Manual (Page 118)

Section 9. Principles of Operation
ADV6600
Y S I
Environmental
Page 108
which is necessary to obtain relatively flow independent measurements. The second important
aspect of the Rapid Pulse technology is the integration (summing of the current) over the total pulse
(on and off). Because the charging current of the electrodes is subtracted in this process, the net
signal is due only to the reduction of oxygen. From a practical point of view, this means that when
there is zero oxygen partial pressure outside the membrane, the Rapid Pulse signal will also be zero;
this in turn allows the system to be calibrated with a single medium (air or water) of known oxygen
pressure.
9-9.2. Effect of Temperature
The DO readings of steady state oxygen systems are greatly affected by temperature (approximately
3% per degree Celsius) due to the effect of temperature on the diffusion of oxygen through the
membrane. The Rapid Pulse system exhibits a greatly reduced effect of temperature (approximately
1% per degree Celsius), but this factor still must be accounted for if DO readings acquired at
temperatures different from that at calibration are to be accurate. The sonde software automatically
carries out this compensation.
In addition, the relationship between the measured partial pressure of oxygen (percent saturation)
and the solubility of oxygen in mg/L is very temperature dependent. For example, air saturated
water (100 percent saturated) contains 9.09 mg/L at 20
o
C, but only 7.65 mg/L at 30
o
C. The sonde
software compensates for both of these temperature-related factors after instrument calibration. The
temperature compensation for the percent saturation reading is empirically derived, while the
conversion from percent saturation and temperature to a solubility in mg/L is carried out using
formulae available in Standard Methods for the Examination of Water and Wastewater (ed. 1989).
9-9.3. Flow Dependence
As noted above, oxygen readings acquired using the Rapid Pulse technology are much less affected
by sample flow than steady state probes. However, there is a finite stirring dependence exhibited by
the Rapid Pulse system if measurements are taken when the probe is being pulsed continuously.
Our tests indicate that, under these sampling conditions, observed dissolved oxygen readings can be
2-3 percent lower than the true readings in very still water. Minimal movement of the water (which
occurs during most environmental measurements) removes this effect.
This small flow dependence of the sensor is greatly reduced in longer term monitoring deployments
where the sampling interval is longer, e.g. 15 minutes. Under these conditions, the sensor is pulsed
for only approximately 40 seconds every 15 minutes, and normal diffusion of oxygen in the medium
re-establishes the oxygen which has been depleted in the previous warm-up/read sequence.