Physical Address
304 North Cardinal St.
Dorchester Center, MA 02124
ASME CRTD-81-2006 — Consensus on Operating Practices for the Sampling and Monitoring of Feedwater and Boiler Water Chemistry in Modern Industrial Boilers
🔬 ASME CRTD-81-2006 is an ASME research report prepared by the Sampling and Monitoring Task Group and the Water Technology Subcommittee of the ASME Research and Technology Committee on Water and Steam in Thermal Systems. It provides consensus recommendations for the sampling and monitoring of feedwater and boiler water chemistry in industrial, utility, and heat-recovery steam generation facilities. The report stresses that all units and sub-systems in the power train must be adequately sampled and monitored: besides the obvious effects of failing to maintain design production of treated water and steam on plant production costs, inadequate monitoring can cause failure of major system components and unplanned outages. It is a companion to the ASME consensus documents on feedwater and boiler water chemistry control, the practical guide to avoiding steam purity problems, and the combined-cycle/cogeneration water and steam chemistry consensus.
💡 Sampling and monitoring are two distinct layers: monitoring covers parameter selection and continuous online measurement, while sampling answers “where, how, how much, and how often.” This report unifies both into one system.
1. The Importance of Effective Monitoring 📋
The report opens by stressing why effective monitoring matters: without it, chemistry deviations go unnoticed until scaling, corrosion, and carryover accumulate into major equipment failures. The objectives of monitoring are to detect abnormalities early, evaluate treatment effectiveness, and guide operational adjustments. The monitoring scheme should match the plant type (industrial, utility, or heat-recovery boiler), the makeup-water treatment (softened or high-purity), and historical water-quality records.
2. Parameters to Be Monitored 🧪
The report discusses a broad set of parameters: pH, conductivity, dissolved oxygen, hardness, phosphate, alkalinity, silica, iron, copper, suspended solids, dissolved solids, and sodium (for carryover and steam-purity assessment). Parameter selection depends on boiler type, operating pressure, and the treatment program; online instruments and laboratory analyses complement each other to form a complete monitoring system.
3. Sampling: Sample Point Locations and Nozzle Installation 📍
Sampling-system design is central to the report. Recommended sample point locations are presented for both softened-water and high-purity makeup systems, covering makeup water, deaerator outlet feedwater, boiler water, and saturated and superheated steam. The report also gives sampling nozzle installation requirements: nozzles should be installed where main-line velocity is high enough to ensure the sample velocity matches the mainstream, avoiding dead zones that would make the sample unrepresentative.
| Sample Location | Purpose | Installation Key Points |
|---|---|---|
| Makeup water (post-treatment) | Monitor incoming quality, evaluate pretreatment | Locate at pretreatment outlet, away from mixing dead zones |
| Deaerator outlet feedwater | Monitor dissolved oxygen, pH, hardness, etc. | Install nozzle perpendicular or 45° facing the flow |
| Boiler water | Assess internal chemistry regime and concentration cycles | Sample both steam space and water space of the drum for comparison |
| Saturated / superheated steam | Monitor steam purity and carryover | Extend nozzle into the mainstream; prevent wall water-film contamination |
| Condensate (as applicable) | Detect condenser leakage and contaminant ingress | Choose points per system configuration |
⚠️ Sample representativeness depends on sampling-system design. Nozzle size, sample-line length and velocity, and cooler capacity must be sized for the required flow; otherwise a sample can look perfectly clean while being thoroughly unrepresentative.
4. Purge Times and Representative Samples ⏱️
Table 3 of the report provides the minimum purge times required for representative sampling of water. Because water retained in the sample line may differ from the mainstream (especially after shutdown, flushing, or load changes), the line must be thoroughly purged before sampling to expel the stagnant water. The purge time depends on line length, internal diameter, flow rate, and sampling-system configuration.
🚨 When online readings and laboratory results disagree, check the sampling system first: insufficient purge, leaking coolers, and blocked sample lines are the three most common causes. Do not rush to blame the instrument or the analyst.
5. Sampling Frequency Guidelines and Test Methods 📊
Tables 1 and 2 give sampling frequency guidelines by location and test for sodium-zeolite (softened) makeup and high-purity makeup systems respectively; Table 4 summarizes the corresponding test methods. The general principle: critical parameters (feedwater dissolved oxygen, boiler-water pH) are monitored at high frequency, others are relaxed according to risk and stability, but startup, shutdown, and load disturbances call for intensified monitoring.
| Test Parameter | Softened Makeup | High-Purity Makeup | Typical Method |
|---|---|---|---|
| pH | Per shift / daily | Online continuous or daily | pH meter / colorimetric |
| Conductivity | Daily | Online continuous | Conductivity meter (cation-conductivity where applicable) |
| Dissolved oxygen | Per shift | Per shift / online | Indigo carmine or electrochemical probe |
| Hardness | Per shift | Daily | EDTA titration |
| Phosphate / alkalinity | Daily | Daily | Titration |
| Silica | Weekly | Weekly / daily | Colorimetric / spectrophotometric |
6. Engineering Practice Essentials 🛠️
- 📍 Location determines representativeness: cover the full chain — makeup → feedwater → boiler water → steam → condensate — instead of watching only boiler water.
- ⏱️ Institutionalize purging: purge per Table 3 before every grab sample and write the purge times into the sampling work instruction.
- 🔧 Maintain the sampling system: periodically inspect coolers, pressure reducers, and nozzles for leakage and blockage.
- 📈 Online plus offline: use online instruments for trend alarming and laboratory analysis for baseline verification; reconcile the two data sets regularly.
- 📋 Adjust frequency dynamically: adapt sampling frequency to water-quality history and risk, intensifying during special periods.
7. Frequently Asked Questions ❓
❓ Q: What happens if there are too few sample points?
A: If only boiler water is monitored while makeup and feedwater are ignored, source-side quality problems go undetected; by the time boiler-water limits alarm, the boiler may already have suffered deposition or corrosion damage.
A: If only boiler water is monitored while makeup and feedwater are ignored, source-side quality problems go undetected; by the time boiler-water limits alarm, the boiler may already have suffered deposition or corrosion damage.
❓ Q: Why purge the sample line before sampling?
A: Water retained in the sample line may differ from the mainstream, especially after shutdown or load changes. Purging until fresh water flows ensures the sample represents current mainstream quality; purge times follow Table 3 of the report.
A: Water retained in the sample line may differ from the mainstream, especially after shutdown or load changes. Purging until fresh water flows ensures the sample represents current mainstream quality; purge times follow Table 3 of the report.
❓ Q: Do softened- and high-purity-makeup systems have the same sampling frequency?
A: No. Softened makeup fluctuates more, so key tests such as pH and hardness run at higher frequency; high-purity makeup is more stable and leans on continuous online monitoring, allowing less frequent laboratory analysis.
A: No. Softened makeup fluctuates more, so key tests such as pH and hardness run at higher frequency; high-purity makeup is more stable and leans on continuous online monitoring, allowing less frequent laboratory analysis.
🔍 Q: Why must the sampling nozzle not be mounted flush against the pipe wall?
A: Near the wall there is a boundary layer and a stagnant water film that does not represent the mainstream. The nozzle should extend into the region of sufficiently high main-line velocity, with the sample velocity matched to the mainstream.
A: Near the wall there is a boundary layer and a stagnant water film that does not represent the mainstream. The nozzle should extend into the region of sufficiently high main-line velocity, with the sample velocity matched to the mainstream.
