Warmup tests

The warmup tests are designed for daily use, to catch errors and faults that might compromise your measurements. If everything passes, you can be reasonably sure (but not guaranteed) that the system is working as expected. Common failures are due to worn out desiccant or an empty CO₂ cylinder. Other failures may be simply because a setting or measurement is just outside the expected range, in which case you have to decide if it needs attention or can be ignored. The descriptions below should help you make those decisions.

All of the system tests that rely on flow rate contain a silent test that checks the state of the flow system. ("Silent" means that output is logged only if there is a problem.) If there is problem with flow, the system test is abandoned, and reasons given.

If the issue is a console leak, the following results:

If the issue is unexpectedly low flow, the following results:

Cal coeff test

The calibration coefficient test (Cal coeff test) tests each current user calibration setting to see if it falls within an expected range. If a value falls outside that range, it marked as failed. This does not necessarily mean something is wrong, but it does bear further investigation. For example, if a zeroing parameter failed, it is OK if that value is what it really takes to properly zero that sensor, but make sure that the sensor hasn’t been mis-zeroed instead.

h2oazero, h2obzero, co2azero, co2bzero

Zero values for the IRGAs (a = sample, b = reference) are typically between 0.8 and 1.2. These values are set from the calibration menu, either by flowing zero air through the appropriate cell and pressing a button, or by direct editing. If one of them is out of range, visit the calibration menu and make sure the zero was properly done. If you need to redo the zero, see Setting the IRGA zeros.

Example: suggestion when warning about co2azero:

h2oaspan1, h2obspan1, co2aspan1, co2bspan1

Span 1 values for the IRGAs (a = sample, b = reference) are typically between 0.9 and 1.2, but expect them to be very close to 1.0. Departures from 1.0 may be real, or may be due to spanning at an inappropriate time, or with a gas that is not actually at the assumed concentration (or for H₂O, an issue with the LI-610 dew point generator: lack of water, overpressure, etc.) These values are set from the calibration menu, either by flowing calibrated air through the appropriate cell and pressing a button, or by direct editing. Hint: manually setting these values to 1.0 can sometimes be the best action for span related issues. If you do this, also set the associated span 2 value (next) to 0.

Example: suggestion when warning about co2aspan:

h2oaspan2, h2obspan2, co2aspan2, co2bspan2

Span 2 values for the IRGAs (a = sample, b = reference) are typically between -0.3 and +0.3. A value of 0 means there is no secondary span. Values outside the expected range should be treated with suspicion, and are likely due to improper calibration. These values are set from the calibration menu, either by flowing calibrated air through the appropriate cell and pressing a button, or by direct editing. Hint: manually setting these values to 0 can sometimes be the best action for span related issues.

Example: suggestion when warning about co2aspan2:

tazero, tbzero

These are offsets to correct the IRGA exhaust temperature sensors (a = sample, b = reference) to the IRGA temperature thermistor (T_irga). These are typically between -0.5 and +0.5 °C. The test for an appropriate setting is to compare T_s and T_r - they should be quite close under stable temperature conditions and similar flows going though the cells. These temperatures can be matched in software version 1.4 and above, by using the BP (background program) named ZeroTaTb.py found in /home/licor/apps/tech/.

Example: suggestions when warning about tazero:

ssa_ref, ssb_ref

These values are factory-set reference readings used to compute the IRGA cell signal strength SS_s and SS_r. (ssa_ref is for sample, and ssb_ref used for reference). Expected values are between 22000 and 55000.

Example: suggestion when warning about ssa ref

chamberpressurezero

This value is the signal from the chamber overpressure sensor when overpressure is 0. The expected value is between 2 and 3 V. This can be set in the flow zeroing calibration screen. A value out of that range would indicate that zeroing occurred when it should not have (there was overpressure), or perhaps a sensor fault.

Suggestion when warning about chamberpressurezero:

flowmeterzero

This value is the offset of the main flow meter. Expected value is between 0.5 and 1.4 V. This can be set in the flow zeroing calibration screen. A value out of that range would indicate that zeroing occurred when it should not have (there was flow), or a sensor fault.

Suggestion when warning about flowmeterzero:

flowazero, flowbzero

These are offset signals to zero the IRGA exhaust flow sensors (a = sample, b = reference). The expected range is between 0 and 0.8 V. These can be set in the Flow/Pressure calibration screen. A value out of that range would indicate that zeroing occurred when it should not have (there was flow), or a sensor fault.

Example: suggestion when warning about flowazero:

pca, pcb

These values are used to correct chamber overpressure for fan speed and flow rate in chambers that have apertures (and a modified internal flow path). The expected values are based on what chamber type the system thinks is attached. The values are available on the Chamber Setup screen.

Table 5‑1. Values for pressure correction coefficients for each chamber type.

Chamber	pca	pcb
6800-01 Fluorometer	0	0
6800-01A Fluorometer	4.3321E-10	1.0215E-5
6800-12 3×3 cm Chamber	0	0
6800-12A 3×3 cm Chamber	2.6520E-10	1.3064E-5
6800-13 6×6 cm Chamber	4.3998E-11	1.4792E-5
6800-17 Small Plant	0	0
6800-19 Custom	0	0

Suggestion when warning about pca or pcb:

Details for each chamber type are given in Chamber info.

Oxygen

This test checks the system constant for a reasonable value for oxygen (2 to 22%). The value is settable in system constants.

Suggestion when warning about Oxygen:

Sensor check

This test checks some sensors for proper zero, and some other sensors and settings for reasonable values.

With the pump off, mixing fan off, and temp control off, all flow (if there is any) is routed to the chamber, and after a 10 second wait, the reference cell IRGA flow is tested for zero.

Flow is then routed to the Reference IRGA, and after a 5 second wait, the following zeros are checked:

Falling outside of the stated tolerance is a warning. It becomes an error if the sensor is out by more than 4 times the tolerance. The remedy is always:

Tests 5 and 6 check that the IRGA's internal temperatures are being controlled to 30 °C. They should be correct by default, but they are changeable through alternative interfaces.

A failure on either of these IRGA control temperature tests is an error, with the suggestions:

A sanity check on temperature sensors follows. At least one leaf temperature sensor is expected (unless configured for the soil or aquatic chamber). Any other failure in this group would indicate a failed sensor, or that you are in conditions out of the operating range of the instrument.

A failure on Test 7 is a warning, with the suggestion:

A failure on Tests 8 through 12 is an error, with the suggestion:

A failure on Test 13 (Ta, Tb, and Tirga block matching) is a warning, with the suggestion:

Flow valve fct

The flow valve functionality test checks for basic functionality of the match/flow split control unit by doing a simple sequence and looking at the flow meters:

1. With all the air routed to the chamber and the match valve normal, measure Flow.
2. That value becomes the target for this test. Flow_s (if the chamber is closed and not leaking badly) should be within 30% of the target value, and Flow_r would be 0.

2. Now route all the flow to the reference cell.
3. Flow_r should read close to the target, and Flow_s should be 0.

3. With all the flow still going to the IRGAS, route it to the sample.
4. Flow_s should be close to the target, and Flow_r should be 0.

If a flow rate is off by more than 30% of expected, the following are suggested:

On failures of the no-flow sensor being off by more than 7 µmol s^-1, the following are suggested:

Pump speed

The pump speed checks that the 4 active pump settings (minimum through high) generate the flow expected, within 20%, adjusting for ambient pressure.

A failure is an error, with these suggestions:

Valve leak

The valve leak test does a systematic leak detection test on the flow split / match valve. The methodology is to put the valve into a particular state, then ramp the pump from low to high, and look for any trends up in the flow meters that should NOT be getting any flow. Specifically, on a plot of leaked flow plotted as function of main flow, we are looking for slopes of less than 0.2%. More than that might be a leak.

Step 1 has all the flow going to the chamber, with the match valve crossed to return air from the chamber is vented to the case. Flow_s should be getting nothing.

Step 2 routes all flow to IRGA and Flow_s. Flow_r would be getting nothing.

Step 3 routes all flow to IRGA and Flow_r. Flow_s should be getting nothing.

Failures of the first two tests are warnings, since they don't really compromise measurement integrity. The suggestions are:

A failure of the last test is serious, and an error. The suggestions are:

A leak in the flow split (step 1) wouldn't matter for normal operation, and a leak while matching (step 2) wouldn't matter, since the valve is not fully closed, but directing air to both cells. However, the critical one is the third test, with the match valve in normal operating position. A leak here represents potential cross contamination between cells, which will corrupt your measurement. A leak in step 3 needs to be dealt with.

Chamber pressure

The chamber pressure test attempts to hold the chamber overpressure at 0.1 kPa with the pump cycling through three flow rates. If there is a chamber leak, the lower the pump speed, the harder it is to hold that high an overpressure. If the issue is not a chamber leak, but a failed pressure control valve, then the test would fail at all pump speeds.

A normal chamber:

A leaky chamber:

Failed valve (or open chamber):

Sample cell cal

The LI-6800 does real time chamber leak detection by comparing the main flow meter to the sample cell flow meter. Since the latter is of much lower performance, it usually needs to be calibrated to the main flow meter. (This calibration only comes into play when computing leak corrections - it does NOT change the displayed / logged values of Flow_s.)

If a preliminary leak test indicates a leak of greater than 2%, an error is triggered, with the suggestion

H₂O limits & match

This test checks the control range for water, and thus checks the state of the desiccant and humidifier tubes. It also checks how well the IRGAs are matched for water over that range at three points: driest, ambient, wettest.

The final graph is a view of the H₂O operating range, and what the match correction looks like over that range. If ambient relative humidity is very low at the beginning of the test, the change may be too small and LI-6800 may report a false failure. You can ignore this, typically.

With working chemicals:

With bad desiccant:

With a dry humidifier tube:

CO₂ limits & match

This test checks the control range for CO₂, and thus checks the state of the soda lime and CO₂ cartridge. It also checks how well the IRGAs are matched for CO₂ over that range at three points: scrubbed, ambient, full CO₂.

The final graph is a view of the CO₂ operating range, and what the match correction looks like over that range.

An open or extremely leaky chamber will produce an error with these suggestions:

Test results when everything works as expected:

Test 1 catches zeroing issues...

... and soda lime issues:

Test 2 catches empty CO₂ cylinders:

If Test 3 fails, and the offset looks like a zeroing issue:

If Test 3 fails, and the offset looks like a spanning issue:

If Test 3 passes, there will be a Test 4 to catch an obvious match need:

Chamber leak

The chamber leak test is a two-part leak test, comparing the Flow_s readings with all air going though the chamber, with all air going directly to the sample cell bypassing the chamber. This is potentially more accurate than the live leak indicator, since the former uses the same flow meter, while the latter uses separate flow meters and depends on their inter calibration.

Leaks less than 1% are pass the test; between 1 and 2% will generate a warning, and above 2% is an error,

Leaks below 50% merit these suggestions:

Above 50% produces these suggestions:

Chamber fan

The chamber fan test sets the fan to a series of 10 setpoints, each randomly chosen in a fixed interval. For each setpoint, there is a typical expected range of resulting fan speeds. If your results have a point or two just outside the expected range, you can safely ignore it. If all or most points are well out (such as all at 0), then there is clearly a problem. If there seems to be a problem with the fan not coming on at lower power settings (or at any setting), then power the head off, remove the chamber, and look for debris obstructing the fan.

If the fan does not appear to be running at any power above 30%, an error with this suggestion is produced:

Any other violation is a warning with this suggestion:

Heat exchanger

The heat exchanger test does a three setpoint test using the heat exchanger, to make sure it is working. If there are failures, contact LI-COR.

Press valve test

This tests attempts to detect harmonic vibrations set up in the pressure control valve that might influence the Flow_s flow sensor. The pressure control is ramped up to 0.2 kPa and back, while the Flow_s sensor is monitored. Normally there is little or no effect on Flow_s, but a vibrating valve can cause a sudden apparent drop in the Flow_s reading.

A normal valve looks like this:

A vibrating valve looks like this:

An open or extremely leaky chamber will produce an error with these suggestions:

Soil chamber test

The soil chamber test is only available when the soil chamber is attached. The test checks for chamber opening and closing, and circulation between the chamber and head.

These actions can also be tested manually from the Environment > Flow page.

Problems here are usually due to improper plumbing, kinked tubing, etc., and produce these suggestions: