TROUBLESHOOTING

 

When trouble arises with a furnace control system, it is important to establish where the problem is located; the probe, signal transmission lines, the control instrument, or the furnace itself. Several simple tests can help to isolate the problem quickly. It is most important to first understand the nature of the fault. Aside from erratic behavior like cycling, or failure to stabilize at the set point, the most common symptom is non-conformity of the work pieces to quality assurance specifications.

To evaluate most faults, the recommended tools are:

 

  1. a good 3 ½ digit millivolt meter with at least 10 megW input impedance and 0 to 1999 mV range,

  2. a temperature calibrator and,

  3. a simulator to output 0 to 1300 millivolts at less than 50 megohms output impedance.

 

 

Probe troubleshooting:

 

In order to establish the source of problems in your installation, first resist the temptation to remove the Gold Probe from the furnace. All of the following meaningful questions must be answered while your Gold Probe (or any other carbon sensor) is in the furnace, at temperature, and exposed to a normal atmosphere under manual control:

 

  1. Does an Alnor dew point reading (or shim stock analysis) verify the indicated value from the probe? If there is reasonable correlation, the problem is NOT the probe.

  2. Are the connections from the T/C extension wire and sensor cable clean and firmly attached at the correct probe and control instrument terminals? Note that the shield wire in the sensor cable should be connected to ground at the control instrument end only! 

  3. Is the control instrument CO or H2 factor set to the appropriate value? This “factor” is referred to by various manufacturers as Zone Factor, Process Factor, Gas, Furnace Factor, CO Factor, Calibration Factor, etc. This factor may require adjustment in order to make the calculated %C or dew point agree with other measurements.

  1. Do the actual Gold Probe temperature and O2 mV signal, as measure by the temperature calibrator, and digital voltmeter, agree with the displayed values on the control instrument? If not, an instrument calibration problem is likely.

  1. Is the probe impedance less than 25 kilohms at temperatures above 1550ºF (843°C)? Conduct the test shown in Fig. 3 using a shunt resistor greater than 100 kilohms. Measure the voltage EC before shunting, then EM with the shunt in place. Calculate RP. If it exceeds 25 kilohms, proceed to step 8, below.

  1. How quickly does the probe react to a change in O2 concentration? Read the probe millivolts with the controller or the digital meter. Short the probe for 5 seconds, remove the short and measure the time required to return to within 1% of the original reading. If it exceeds 30 seconds, proceed to step 8, below.

  1. Is there a leak in the zirconia substrate? To test this property, turn off the reference air for one minute. Measure the probe mV as indicated by the controller or a digital voltmeter. Turn the air back on and measure the mV again. If there is a difference greater than 25 mV, replace the probe.

  2. If probe resistance or response times are questionable as indicated in steps 5 and 6, we recommend that the probe be burned off. Introduce 10 to 15 CFH of air to the burnoff fitting for about 90 to 120 seconds, and then repeat the tests. Should problems persist, it may be necessary to conduct a thorough furnace burnout so that all potentially contributing contamination is removed from all parts of the furnace, including the Gold Probe. Burning off the probe will not harm this product provided the probe temperature does not exceed 2000ºF (1093°C) during the burnoff procedure.

  1. Should it be necessary to remove your Gold Probe from a hot furnace, do so carefully. UNDER NO CIRCUMSTANCES should it be removed faster than 2" (51mm) per minute.               

  2. Failing resolution of your atmosphere control problems our technical support staff is available Mon. - Fri. 7:00 a.m. to 7:00 p.m. to assist and serve our heat treating customers. Call us at (513) 772-0060.

    SYMPTOM POSSIBLE CAUSE
    High % C

    Low reading due to:

    • High probe resistance
    • Cracked zirconia
    • Dirty reference air
    • Faulty cable insulation
    • Instrument calib/calc
    • Air leak to burnoff fitting
    • Furnace air leak at probe

    Oily parts or sooted furnace

    Wrong recipe time/temp
    Low % C

    High reading due to:

    • Probe plugged with soot
    • Instrument calib/calc

    Wrong recipe time/temp
    Erratic

    Faulty signal due to:

    • Bad sensor connections
    • Electrical noise source
    • Radiant tube leak
    • Bad Endo
    • Mixing valve setting
    • Instrument setting
    Sooted furnace Endo not cracked (temp too low in generator or catalyst inactive)