Teledyne 9060 - Zirconium oxide flue gas oxygen analyzer - Manual (Page 42)

9060-Zirconia Analyser
Note: If two probes was selected in set-up 1, ‘Cal Gas 2’ must be connected to probe 2.
3.24 DUST IN THE FLUE GAS
For unheated probes with no filter, entrained solids or dust in the flue gas does not present
a problem unless the dust, when settled, is not porous. Allow the dust in the process to
build up on the probe. It will form a porous layer slowing the response time. To avoid
mechanical abrasion of the electrode material in installations with unheated oxygen
probes, pack ‘SAFFIL’ or equivalent alumina based ceramic fibre in the sensing holes to
protect the electrode. Do not use silica based ceramic fibres such as ‘KAOWOOL’, which
can attack the electrode at high temperatures. Once the dust has built up the response
time of the probe will be slower.
For heated probes the preferred method of mounting for dust-laden applications is facing
vertically downwards with the filter removed. Probes can also be mounted horizontally
with no filter with some dusts. An occasional automatic back purge is helpful in this case.
Normally heated probes are supplied with filters for applications with particulates in the flue
gas. The probe response time should be tested when the probe is first installed, and then
regularly until it remains constant for a significant period. Filter purging should be set up
on the time periods determined by these tests. To test the probe response time, use a
stopwatch to obtain the time for a probe to achieve a 63 % change from one reading to
another. If a probe filter blocks completely in a short period of time, then there is no option
but to use the probe without the filter. A trial probe with filter can be installed to test
whether a filter blockage is likely to occur.
3.25 STRATIFICATION
If the transmitter and probe have been fully tested and the oxygen readings in the flue gas
are incorrect, gas stratification may be occurring. The phenomena cannot be anticipated
for any particular installation. Generally, large flues have oxygen differences of
approximately one percent across the flue. Occasionally an oxygen error of several
percent may occur in a flue of any size. Moving the probe to a new location normally
solves this problem.
The effects of stratification can be reduced by using two probes and averaging the two
oxygen readings. This can be achieved within a Teledyne 9060 Analyser controlling two
probes.