CO2 Tank Monitor

Monitor CO2 cylinder pressure and predict depletion times to prevent gas outages in cell culture incubators, particularly during weekends when laboratories are unmanned. Provides automated alerts and consumption tracking for proactive cylinder management.

Key Capabilities:

• - Pressure-Based Depletion Prediction: Calculate remaining cylinder life based on current pressure and consumption rates
• Weekend Risk Detection: Identify if depletion will occur during weekends when no staff is available
• Multi-Cylinder Support: Handle different cylinder sizes (10L, 40L) and specifications
• Automated Alert System: Generate status reports with actionable recommendations
• Simulation Mode: Test monitoring scenarios with randomized data for training

---

When to Use

✅ Use this skill when:

• - Monitoring cell culture CO2 incubators that require continuous gas supply
• Setting up automated daily checks for cylinder status (e.g., via cron jobs)
• Planning cylinder replacement schedules to avoid service interruptions
• Training new lab members on gas monitoring procedures (use simulation mode)
• Managing multiple incubators with different consumption rates
• Pre-holiday assessments to ensure adequate gas supply during lab closures
• Creating standard operating procedures (SOPs) for gas monitoring

❌ Do NOT use when:

• - Monitoring liquid nitrogen tanks for cryogenic storage → Use freezer-sample-locator or specialized LN2 monitors
• Tracking compressed air or N2 cylinders for non-CO2 applications → Adapt parameters but verify compatibility
• Real-time IoT sensor integration is available → Use actual sensor API instead of simulation
• Cylinder uses weight-based measurement instead of pressure → Modify calculation logic accordingly
• Emergency gas leak detection is needed → Use gas detection safety systems with alarms
• Managing bulk tank installations with automatic switchover → These systems have built-in monitoring

Related Skills:

• - 上游 (Upstream): equipment-maintenance-log, INLINECODE2
• 下游 (Downstream): eln-template-creator, INLINECODE4

---

Integration with Other Skills

Upstream Skills:

• - equipment-maintenance-log: Track CO2 incubator maintenance history that affects consumption rates
• INLINECODE6: Monitor cylinder inventory and replacement schedules
• INLINECODE7: Record calibration dates for pressure gauges

Downstream Skills:

• - eln-template-creator: Generate experiment templates that include gas monitoring checklists
• INLINECODE9: Forecast CO2 gas costs based on consumption trends
• INLINECODE10: Coordinate empty cylinder disposal with replacement

Complete Workflow:

Daily Morning Check → co2-tank-monitor → equipment-maintenance-log → lab-inventory-tracker → Replacement Scheduling

---

Core Capabilities

1. Pressure-Based Depletion Prediction

Calculate remaining cylinder lifetime based on current pressure readings and historical consumption patterns.

CODEBLOCK1

Parameters:

Parameter	Type	Required	Description	Default
INLINECODE11	float	Yes	Current cylinder pressure in MPa	8.0
INLINECODE12

float | Yes | Average daily consumption rate (MPa/day) | 1.5 |
| capacity | int | No | Cylinder capacity in liters (10 or 40) | 40 |

Calculation Method:
CODEBLOCK2

Best Practices:

• - ✅ Calibrate consumption rate based on historical data (monitor for 1-2 weeks)
• ✅ Account for usage variations - higher on heavy culture days, lower on weekends
• ✅ Check pressure at consistent time (e.g., 9 AM daily) for accurate trends
• ✅ Update consumption rate seasonally - incubator usage may vary

Common Issues and Solutions:

Issue: Inaccurate consumption estimates

• - Symptom: Predicted depletion date consistently off by several days
• Solution: Monitor actual consumption over 2-4 weeks; adjust daily_consumption parameter

Issue: Pressure fluctuations causing false alerts

• - Symptom: Inconsistent pressure readings due to temperature changes
• Solution: Take readings at same time daily; allow cylinder to equilibrate after use

2. Weekend Depletion Risk Detection

Identify if cylinder depletion will occur during weekends when laboratory staff is unavailable.

CODEBLOCK3

Weekend Risk Scenarios:

Scenario	Risk Level	Action Required
Depletion on Saturday/Sunday	🔴 High	Immediate replacement or weekend duty
Depletion Monday morning

🟡 Medium | Replace Friday afternoon |
| Depletion Friday evening | 🟡 Medium | Monitor closely; replace if possible |
| Depletion mid-week | 🟢 Low | Schedule routine replacement |

Best Practices:

• - ✅ Set alertdays = 2 for standard work weeks (alert by Wednesday for Friday replacement)
• ✅ Plan Friday replacements for any cylinder with <4 days remaining
• ✅ Create weekend duty roster for critical experiments during high-risk periods
• ✅ Consider holiday schedules - extend alertdays before long weekends

Common Issues and Solutions:

Issue: False weekend alerts for Monday depletion

• - Symptom: Alerts triggered for Monday morning depletion when lab reopens
• Solution: Adjust logic or manually review; Monday AM depletion may be acceptable

Issue: Missing Friday evening depletion risk

• - Symptom: Cylinder runs out Friday night after staff leaves
• Solution: Check Friday 5 PM status specifically; consider Friday evening as weekend

3. Multi-Cylinder Size Support

Support different CO2 cylinder specifications commonly used in laboratory settings.

CODEBLOCK4

Cylinder Specifications:

Capacity	Typical Use	Full Pressure	Duration (@1.5 MPa/day)
10L	Small incubators, backup	~15 MPa	~10 days
40L

Main incubators, heavy use | ~15 MPa | ~40 days |

Best Practices:

• - ✅ Use 40L cylinders for main incubators to reduce replacement frequency
• ✅ Keep 10L cylinders as emergency backup for critical experiments
• ✅ Record cylinder capacity in monitoring logs for accurate predictions
• ✅ Standardize on one size per facility when possible for simplified inventory

Common Issues and Solutions:

Issue: Wrong capacity setting causing prediction errors

• - Symptom: Predictions consistently off by 4x (10L vs 40L confusion)
• Solution: Verify cylinder label; update capacity parameter in monitoring script

Issue: Mixing cylinder sizes in same calculation

• - Symptom: Different incubators on different cylinder sizes
• Solution: Run separate monitoring instances for each cylinder/incubator pair

4. Automated Alert System with Status Levels

Generate color-coded status reports with specific recommendations based on urgency.

CODEBLOCK5

Status Levels:

Code	Icon	Status	Condition	Action
0	🟢	Normal	Days > alertdays + 2	No action needed
1

🟡 | Caution | Days within alertdays + 2 | Monitor closely |
| 2 | 🔴 | Danger | Days ≤ alert_days or weekend depletion | Replace immediately |

Return Codes for Automation:

Exit Code	Meaning	Automation Action
0	Normal	Continue routine monitoring
1

Caution | Send email notification |
| 2 | Danger | Send urgent alert; page on-call staff |

Best Practices:

• - ✅ Integrate with alerting systems using return codes for automated notifications
• ✅ Customize recommendations for your lab's specific procedures
• ✅ Escalate based on status - email for caution, SMS/call for danger
• ✅ Log all status changes for trend analysis and audit trails

Common Issues and Solutions:

Issue: Alert fatigue from frequent caution alerts

• - Symptom: Too many caution-level notifications causing ignored alerts
• Solution: Adjust alert_days threshold; batch daily reports instead of immediate alerts

Issue: Missing critical weekend alerts

• - Symptom: Weekend depletions not detected when they span multiple days
• Solution: Ensure willdepleteon_weekend() logic covers multi-day weekend periods

5. Simulation Mode for Training

Generate randomized cylinder data for training staff without affecting real monitoring systems.

CODEBLOCK6

Training Applications:

Use Case	Simulation Parameters	Learning Objective
New staff training	Various random scenarios	Recognize different alert levels
Weekend risk awareness

Force weekend depletion scenarios | Understand critical timing |
| Consumption calculation | Different pressure/consumption combos | Practice manual calculations |
| Emergency response | Low pressure scenarios (<3 MPa) | Learn urgent procedures |

Best Practices:

• - ✅ Run simulations weekly to keep staff familiar with alert formats
• ✅ Create scenario library with specific training cases (weekend depletion, low pressure, etc.)
• ✅ Include simulation in onboarding for new lab members
• ✅ Test alert pathways using simulation to verify notification systems

Common Issues and Solutions:

Issue: Simulation data too random for consistent training

• - Symptom: Trainees see different scenarios each session
• Solution: Set random seed for reproducible scenarios; create predefined training datasets

Issue: Confusion between simulation and real data

• - Symptom: Staff mistakes simulation for actual cylinder status
• Solution: Clearly label all simulation outputs; use distinct formatting

6. Automated Scheduling Integration

Integrate with cron jobs or task schedulers for automated daily monitoring.

CODEBLOCK7

Integration Patterns:

Integration Type	Trigger	Action	Return Code Handling
Daily cron	9:00 AM daily	Check status	Log results; alert if ≠0
Pre-weekend

Friday 5 PM | Weekend risk check | Force alert if weekend risk |
| Real-time | Sensor threshold | Immediate check | Page on-call if danger |
| Manual | Lab staff | Ad-hoc check | Display full report |

Best Practices:

• - ✅ Use --quiet mode for automated runs to reduce log noise
• ✅ Capture return codes for alerting and logging
• ✅ Test cron jobs manually before deploying to production
• ✅ Monitor the monitor - ensure scheduled checks are running
• ✅ Implement heartbeat - alert if monitoring script fails to run

Common Issues and Solutions:

Issue: Cron job not executing

• - Symptom: No monitoring logs or alerts
• Causes: Path issues, permissions, environment variables
• Solution: Use full paths; test manually first; check cron logs

Issue: Sensor data unavailable

• - Symptom: Cannot read pressure from sensor log
• Solution: Implement fallback to manual input; alert on sensor failure

---

Complete Workflow Example

From daily monitoring to replacement scheduling:

CODEBLOCK8

Python API Usage:

CODEBLOCK9

Expected Output Files:

CODEBLOCK10

---

Common Patterns

Pattern 1: Daily Morning Monitoring Routine

Scenario: Lab technician checks all CO2 cylinders every morning at 9 AM.

CODEBLOCK11

Workflow:

1. 1. Read pressure gauges on all cylinders at 9 AM
2. Run monitoring script for each cylinder
3. Log status codes and any alerts
4. If any cylinder shows status ≥1, notify lab manager
5. Schedule replacements for caution-level cylinders
6. For danger-level, initiate immediate replacement procedure

Output Example:
CODEBLOCK12

Pattern 2: Pre-Holiday Weekend Assessment

Scenario: Before a 3-day weekend (e.g., Memorial Day), ensure adequate gas supply.

CODEBLOCK13

Workflow:

1. 1. On Friday before holiday, run assessment with extended alert_days (4)
2. Check all cylinders with weekend risk detection
3. For any cylinder with depletion during closure period:

- Prioritize immediate replacement
- Consider emergency weekend delivery
- Transfer critical cultures to backup incubator

1. 4. Document gas status in holiday handoff notes
2. Set up emergency contact for gas supplier

Output Example:
CODEBLOCK14

Pattern 3: New Lab Member Training

Scenario: Train new technician on CO2 monitoring using simulation mode.

CODEBLOCK15

Workflow:

1. 1. Run simulation mode to generate 10 random scenarios
2. For each scenario, trainee manually calculates remaining days
3. Compare trainee calculation with script output
4. Discuss appropriate actions for each status level
5. Practice emergency procedures with "danger" scenarios
6. Review actual monitoring logs from past month
7. Shadow experienced technician for first week

Output Example:
CODEBLOCK16

Pattern 4: Multi-Incubator Facility Management

Scenario: Large facility with 6 incubators on different CO2 cylinders.

CODEBLOCK17

Workflow:

1. 1. Create monitoring dashboard tracking all 6 cylinders
2. Run monitoring script for each cylinder daily
3. Implement staggered replacement schedule to avoid all cylinders needing replacement simultaneously
4. Track consumption trends to optimize delivery schedule
5. Maintain 1-2 backup cylinders in inventory
6. Negotiate bulk pricing with gas supplier based on usage patterns
7. Monthly review of consumption trends and cost optimization

Output Example:

Facility CO2 Dashboard - 2026-02-09
====================================

Main Incubators (40L):
  Inc-01: 🟢 18.2 days | Last replaced: 2026-01-22
  Inc-02: 🟡 4.5 days | REPLACE THIS WEEK
  Inc-03: 🟢 22.1 days | Last replaced: 2026-01-15
  Inc-04: 🟢 15.8 days | Last replaced: 2026-01-28

Backup Incubators (10L):
  Back-01: 🟢 6.2 days
  Back-02: 🟢 8.4 days

This Week Actions:
  - Replace Inc-02 cylinder by Wednesday
  - Order replacement for delivery next week
  
Monthly Consumption: 42 MPa (28% under budget)

---

Quality Checklist

Pre-Monitoring Setup:

• - [ ] CRITICAL: Verify pressure gauge is calibrated and functional
• [ ] Confirm cylinder size (10L or 40L) matches monitoring parameters
• [ ] Establish baseline daily consumption rate through 1-2 weeks observation
• [ ] Set appropriate alert_days threshold (typically 2 for standard weeks)
• [ ] Configure automated scheduling (cron) for daily checks
• [ ] Set up alerting pathway (email, SMS) for danger-level status
• [ ] Create emergency contact list for gas supplier
• [ ] Ensure backup cylinder availability for critical incubators

During Daily Monitoring:

• - [ ] Take pressure readings at consistent time (e.g., 9:00 AM)
• [ ] CRITICAL: Record actual pressure value, not estimates
• [ ] Note any unusual consumption patterns (door openings, new cultures)
• [ ] Verify status code and understand implications
• [ ] Check for weekend depletion risk (especially on Thursdays/Fridays)
• [ ] Log all readings for trend analysis
• [ ] Update consumption rate if usage patterns change significantly
• [ ] Inspect cylinder and regulator for leaks or damage

Alert Response:

• - [ ] CRITICAL: Acknowledge danger-level alerts immediately
• [ ] For weekend depletion risk, arrange replacement before Friday 5 PM
• [ ] Notify lab manager of caution-level cylinders needing replacement
• [ ] Document all actions taken in response to alerts
• [ ] Verify replacement cylinder is available before removing empty one
• [ ] After replacement, verify incubator maintains proper CO2 concentration
• [ ] Update monitoring logs with replacement date and new cylinder info
• [ ] Return empty cylinder to storage area for pickup

Post-Replacement Verification:

• - [ ] CRITICAL: Verify new cylinder valve is fully open
• [ ] Check pressure gauge reading on new cylinder (should be ~15 MPa)
• [ ] Monitor incubator CO2 concentration for 30 minutes to verify stability
• [ ] Verify no gas leaks at regulator connections (soap bubble test)
• [ ] Update monitoring parameters with new cylinder start pressure
• [ ] Log replacement in inventory management system
• [ ] Notify relevant lab members of cylinder change
• [ ] Schedule next replacement based on predicted depletion date

---

Common Pitfalls

Monitoring Setup Issues:

• - ❌ Inconsistent reading times → Daily variations affect trend accuracy

- ✅ Take readings at same time each day (e.g., 9:00 AM ± 30 min)

• - ❌ Wrong consumption estimates → Inaccurate depletion predictions

- ✅ Calculate from actual usage over 2+ weeks; update seasonally

• - ❌ Ignoring temperature effects → Pressure varies with ambient temperature

- ✅ Allow cylinder to equilibrate to room temperature before reading

• - ❌ Not accounting for usage variations → Weekend vs weekday consumption differs

- ✅ Monitor separately for different usage patterns; use average

Alert and Response Issues:

• - ❌ Alert fatigue → Too many notifications cause important alerts to be missed

- ✅ Batch daily reports; only escalate urgent alerts immediately

• - ❌ Missing weekend alerts → Friday check doesn't catch Saturday depletion

- ✅ Always check weekend risk explicitly; use Friday 5 PM check

• - ❌ Delayed replacement → Waiting too long leads to actual depletion

- ✅ Replace when caution status triggered; don't wait for danger

• - ❌ No backup plan → Cylinder fails with no replacement available

- ✅ Maintain 1-2 backup cylinders; know emergency supplier contacts

Calculation and Data Issues:

• - ❌ Wrong cylinder capacity → 10L vs 40L confusion causes 4x error

- ✅ Always verify cylinder label; double-check capacity parameter

• - ❌ Pressure unit confusion → PSI vs MPa or Bar mixing

- ✅ Standardize on MPa; convert if gauge shows different units

• - ❌ Not tracking multiple cylinders → Mixing readings from different tanks

- ✅ Label cylinders clearly; use location-specific monitoring

• - ❌ Ignoring cumulative error → Small daily errors compound over weeks

- ✅ Verify predictions against actual depletion; calibrate regularly

Operational Issues:

• - ❌ Gauges not calibrated → Inaccurate readings lead to wrong predictions

- ✅ Calibrate pressure gauges annually or per manufacturer recommendation

• - ❌ Leaks undetected → Higher consumption than predicted

- ✅ Regular leak checks; sudden consumption increases indicate leaks

• - ❌ Regulator problems → Pressure drops even with adequate gas

- ✅ Replace regulators per schedule; check for ice buildup in flow

• - ❌ No documentation → Cannot track trends or troubleshoot issues

- ✅ Maintain detailed logs; review monthly for optimization

---

Troubleshooting

Problem: Predicted depletion date consistently wrong

• - Symptoms: Actual depletion 3-5 days earlier or later than predicted
• Causes:

- Incorrect consumption rate estimate
- Changing usage patterns (more/fewer cultures)
- Temperature variations affecting pressure readings
- Leaks in system

• - Solutions:

- Recalculate consumption rate from recent actual usage
- Monitor for 2+ weeks to establish accurate baseline
- Take readings at consistent temperature conditions
- Check for leaks using soapy water at connections

Problem: Frequent false alerts (alert fatigue)

• - Symptoms: Receiving caution/danger alerts that don't require action
• Causes:

- Alert threshold too sensitive
- Normal fluctuations triggering alerts
- Multiple notifications for same cylinder

• - Solutions:

- Adjust alert_days parameter (increase from 2 to 3)
- Implement alert aggregation (daily summary vs immediate)
- Use quiet mode for automated checks; manual for full reports

Problem: Weekend depletion not detected

• - Symptoms: Cylinder runs out Saturday/Sunday despite monitoring
• Causes:

- Friday check missed weekend risk
- Depletion spans multiple days
- Holiday weekends not considered

• - Solutions:

- Add explicit Friday 5 PM check for weekend risk
- Extend alert_days before long weekends
- Use willdepleteon_weekend() logic that checks multi-day periods

Problem: Cannot read pressure gauge accurately

• - Symptoms: Inconsistent or unclear pressure readings
• Causes:

- Gauge needle stuck or damaged
- Condensation or dirt on gauge face
- Parallax error reading needle position

• - Solutions:

- Replace faulty gauges immediately
- Clean gauge face regularly
- View needle straight-on to avoid parallax
- Consider digital pressure sensors for better accuracy

Problem: Cylinder depletes faster than expected

• - Symptoms: Predicted 10 days, actually depleted in 6 days
• Causes:

- Increased incubator usage (more cultures, frequent door openings)
- Gas leak in system
- Regulator malfunction
- Temperature increase (higher consumption)

• - Solutions:

- Check for leaks at all connections
- Verify incubator door seals
- Inspect regulator for proper function
- Update consumption rate to reflect actual usage

Problem: Monitoring script not running automatically

• - Symptoms: No daily logs or alerts despite cron setup
• Causes:

- Cron job misconfigured
- Path issues in script
- Permissions problems
- Script errors not logged

• - Solutions:

- Test cron job manually first
- Use full absolute paths in cron and script
- Check cron logs: grep CRON /var/log/syslog
- Redirect stderr to log file for debugging

---

References

Available in references/ directory:

• - (No reference files currently available for this skill)

External Resources:

• - Cell Culture CO2 Guidelines: https://www.thermofisher.com/cellculture
• Gas Cylinder Safety: https://www.osha.gov/gascylinders
• CO2 Incubator Best Practices: https://www.sigmaaldrich.com/incubators

---

Scripts

Located in scripts/ directory:

• - main.py - CO2 tank monitoring engine with prediction and alerting logic

---

Pressure Conversion Reference

Unit	MPa	PSI	Bar
MPa	1.0	145.0	10.0
PSI

0.0069 | 1.0 | 0.069 | | Bar | 0.1 | 14.5 | 1.0 |

Typical Cylinder Pressures:

• - Full cylinder: ~15 MPa (~2200 PSI)
• Working pressure: 8-10 MPa
• Replace threshold: ~3-5 MPa
• Empty: <1 MPa

---

Last Updated: 2026-02-09
Skill ID: 182
Version: 2.0 (K-Dense Standard)