COD Analyzer

What a COD Analyzer Is and What Is Being Decided

The measured quantity deserves one paragraph of honesty before any hardware. Chemical oxygen demand is an operational definition, not a substance: it is the oxygen equivalent of whatever a specific oxidant, under specific acid and heat conditions, manages to oxidize in two hours, and different oxidants oxidize different fractions of a real wastewater. Dichromate with a silver catalyst and mercury masking, the regulatory mainstream, recovers nearly all common organics and a good share of reduced inorganics; permanganate, the older and gentler reagent, oxidizes a smaller and different fraction, which is why permanganate-index values run lower than dichromate values on the same water and why the two conventions are never directly interchangeable. The laboratory reference methods, ISO 15705, ASTM D1252, Standard Methods 5220, and the national HJ series in East Asia, each fix reagent, digestion, and detection so that numbers travel between laboratories. An online COD analyzer is machinery that reproduces one of those defined chemistries, or correlates to them, at the fence line, unattended, and the purchase question is always which of those three claims, reproduce, correlate, or pretend, the instrument is actually making.

What is actually being decided in this purchase is four guarantees. First, method standing: if the permit or contract names a method, the analyzer must run it or hold a documented, defensible equivalence, and the UV correlation route is accepted only where the regulation allows continuous surrogate monitoring, which many do for process and many do not for compliance. Second, range and matrix fit: the instrument must measure the actual stream, municipal influent at 300 to 1,000 mg/L with solids and grit, industrial effluent with salinity or color interferences, or clean surface water at single digits, within its native or dilution range, and the datasheet's fine print on chloride tolerance decides more industrial applications than the headline range does. Third, operational reality: wet chemistry analyzers consume reagents and produce chromium or manganese waste with its own disposal obligations, and the cost per analysis and the waste drum arithmetic belong in the purchase decision, not in the first quarter's operating surprise. Fourth, data integrity: cycle time, calibration and validation provisions, and the outputs, 4 to 20 mA, relays, digital bus, decide whether the number arrives in time to matter and stands up when the regulator asks.

Against its neighbors the COD analyzer occupies deliberate ground. A TOC analyzer measures organic carbon after oxidation to carbon dioxide, a closely related pollution strength parameter with different chemistry, faster analysis on the combustion side, and its own standing in permits; the two numbers correlate loosely per site and are never a single instrument family. An online water analyzer is the umbrella term for parameter suites, COD among them with several others, and the umbrella page is where multi-parameter packages are compared. Specific neighbors cover specific parameters: an oil in water analyzer for hydrocarbons, a residual chlorine analyzer for disinfection, a hardness analyzer, silica analyzer, ammonia analyzer, phosphate analyzer, sodium analyzer, and hydrazine analyzer for the water chemistry panel, and the simple sensors, a conductivity meter, a pH meter, an ORP meter, a turbidity meter, and a dissolved oxygen meter for the field measurements that surround the oxygen demand story. BOD, the five day biological test, is the parameter COD most often substitutes for on speed, and the substitution logic belongs in the selection chapter. The COD analyzer takes the narrow territory of automated oxygen demand measurement, and every neighbor either measures something else or measures it slower.

The standards that organize the datasheet are the reference methods themselves and the monitoring regimes built on them. ISO 15705 and Standard Methods 5220 fix the dichromate basis; ASTM D1252 carries the closed reflux colorimetric variant; the HJ series, HJ 828 for dichromate and HJ 132 for the online fast digestion cousins, governs where Chinese discharge permits apply, and JIS K 0806 family addresses the permanganate and online instruments in the East Asian convention. Continuous monitoring acceptance, calibration frequency, and audit requirements come from the permit authority, USEPA program rules in the United States, the HJ 353, 354, 355 acceptance series in China, and MCERTS in the United Kingdom, and these regimes, not the chemistry, usually decide whether a UV correlation instrument is eligible for the duty. A datasheet that names no method basis is a price tag, not a specification, and a quotation claiming regulatory acceptance without naming the regime is asserting, not evidencing.

The RFQ should also name the sampling and data station so the package checks out as a system. The analyzer sees only what the sample system delivers: a sample conditioning system with intake, filtration or settling appropriate to the method, and a flow-controlled stream at temperature and pressure the instrument tolerates, and on hard slurries a fast loop driven by a centrifugal pump with the analyzer tapping the loop. Multi-stream installations add a stream selection manifold with its cross-contamination story. Data lands through 4 to 20 mA into the SCADA or recorder, through Modbus or the local equivalent where the plant is digital, and to the discharge telemetry where the regime demands it, with alarm relays for range and fault. Stating the whole station in the RFQ lets one supplier check matrix, filtration, and data compatibility in a single pass.

How Digestion Chemistry and UV Correlation Produce the Number

The dichromate instrument reproduces the reference chemistry in miniature and automation. A metered sample, or a diluted aliquot on high strength streams, mixes with the reagent package: dichromate in concentrated sulfuric acid, silver sulfate catalyst, and mercury sulfate to mask chloride interference. The mixture digests at about 148 to 175 deg C for a fixed period, in classic instruments two hours mirroring the laboratory reflux and in fast versions 15 to 30 minutes with the acceptance basis carrying the deviation; the orange dichromate is consumed in proportion to the oxidizable matter, and the photometer reads the remaining oxidant, at about 445 nm directly or at 600 nm class through the chromium three-plus green, converting absorbance to milligrams per liter through the stored calibration. Spent reagent, now containing chromium six plus reduced to chromium three plus and whatever mercury the package carried, drains to a waste container whose handling is part of the plant's hazardous waste program, at a few milliliters per analysis accumulating to real liters per month.

The permanganate instrument follows the same architecture with the gentler oxidant: potassium permanganate in acidic, sometimes alkaline, medium, digested at 100 deg C class for 30 minutes to an hour by convention, read photometrically at 525 nm family through the permanganate color loss. The chemistry suits low strength waters, river and lake monitoring and drinking plant intake, where dichromate's aggression is unnecessary and the permanganate index is the conventional parameter, and it avoids silver and mercury while producing manganese waste, a lighter disposal story than chromium. The critical honesty is the same as in the laboratory: permanganate index and dichromate COD are different measurements, correlating per site through paired data, and an analyzer specification must name which convention the number serves because the permit or contract names one.

The UV correlation instrument removes chemistry from the measurement entirely. A filtered sample flows through a cuvette; the instrument reads absorbance at 254 nm, the aromatic and unsaturated organic band, optionally with turbidity compensation at a second wavelength, and converts absorbance to COD through a calibration the site builds from paired laboratory COD values across the load range. The physics is honest, the speed is real, seconds of response and cycles per minute where the wet chemistries deliver one number per 30 minutes to two hours, and the reagent and waste lines disappear; the weakness is equally structural, because UV254 sees only the UV-active fraction of the organics, and a stream whose UV-inactive load varies, sugars, some acids, changed industrial inputs, drifts the correlation silently. The instrument earns its place on process control and on streams with stable character, and on compliance duty only where the regime explicitly accepts continuous surrogate monitoring with periodic laboratory confirmation, which the RFQ must verify in writing rather than assume.

Cycle time and delay set the control value of the number. A dichromate analyzer reporting every 30 to 120 minutes describes the load that arrived in the last hour plus; aeration control on that number runs an hour behind the diurnal curve, which is why many plants run the fast digestion variants or pair a UV sensor for control with the wet chemistry for the record. Detection limit and precision follow the chemistry too: online dichromate instruments quote limits around 3 to 10 mg/L class and repeatability of a few percent of reading above the low range, permanganate similar at its ranges, and UV correlations limited by their calibration quality rather than optics. Where the duty is load billing at an interface, the precision statement and the validation regime, standard checks at intervals, audit trails, decide disputes before they start, and the purchase should demand the precision basis: percent of reading at what concentration, on what standard.

A worked example fixes the arithmetic. A food plant effluent at 800 to 2,500 mg/L COD with 1,500 mg/L chloride wants a discharge record and a load signal for its equalization control. The chloride figure matters first: plain dichromate chemistry tolerates roughly 1,000 mg/L chloride even with mercury masking before positive bias creeps in, so the specification demands chloride handling, dilution as standard with the range to carry it, and the supplier confirms at the stated matrix. The duty splits: a fast dichromate analyzer on the discharge with monthly laboratory confirmation satisfies the permit where the authority accepts online equivalents, while a UV254 sensor on the equalization inlet drives the control an hour earlier than the wet chemistry could. Reagent consumption at one analysis per hour runs the operating cost line, roughly 0.7 to 2 liters of reagent per week by package, and the chromium waste lands in the hazardous drum at the corresponding rate. Every number in the example is family arithmetic for scale; the method, the chloride line, and the split duty are the deliverables.

Types, Construction, and Duty Boundaries

Three instrument families cover the catalogs. The online dichromate analyzer is the compliance workhorse, a cabinet instrument, Hach CODmax III class among others, with reagent reservoirs, a heated digestion reactor, metering pumps and valves, a photometer, and a controller running the sequence and the data; ranges commonly 0 to 10,000 mg/L with automatic dilution, cycle times 20 to 120 minutes by program. The online permanganate analyzer, Endress+Hauser CA80COD class among others, serves the surface water and East Asian permit convention with its gentler chemistry and lighter waste story. The UV254 correlation sensor is a flow cell and photometer, sometimes with multi-spectral optics for compensation, sold as a probe or compact panel unit with the calibration built site side, and it is the instrument of the control room rather than the permit file wherever the regime restricts surrogates.

Within the families, construction choices carry the warranty. Wet chemistry cabinets divide on reagent logistics: bottles changed by hand on a schedule, cartridges exchanged, or bulk reservoirs with level alarms, and the choice prices the operating labor and the outage risk of an empty reagent. Digestion reactors hold acid-wetted heat, so wetted materials, valve reliability at the reactor, and the service interval of the metering path, pumps and valves rebuilt on a schedule measured in months, are the datasheet rows that predict year three. Photometer stability and the automatic zero and standard checks decide data credibility between services. Enclosures run IP54 to IP65 panel or wall mount by family, with ambient limits that matter in unheated analyzer sheds, and hazardous area versions exist but are the exception. UV sensors divide on compensation sophistication, single wavelength plus turbidity correction versus multi-spectral with algorithms, and on cuvette cleaning, mechanical wipers or chemical cleans against biofilm, the maintenance item that dominates their cost of ownership.

Boundaries are worth stating bluntly. A COD analyzer does not measure BOD, the five day biological oxygen demand, and where a permit names BOD the plant runs correlations from COD with the laboratory confirming, or runs respirometers, but the substitution is a permit conversation, not a hardware feature. It does not measure TOC, whose instruments oxidize to carbon dioxide and count carbon; a TOC analyzer is the neighbor for permits and processes that frame pollution as carbon. The simple sensors answer different questions entirely: turbidity is not strength, conductivity is salinity not load, and dissolved oxygen is the receiving water or the aeration basin state, not the waste strength. And the online instrument never out-stands its laboratory basis: every wet chemistry analyzer is certified against the reference method, and every UV correlation is only as good as the paired data behind it, so the purchase includes the validation program, standards, schedule, records, or the number is decoration. The duty map below pairs the common stations with the family that survives them.

Key Specification Parameters

The table below decodes the datasheet lines that decide whether a quoted instrument matches the duty. Each row is a purchase defining parameter; the paragraphs after the table explain what to demand and what a weak answer looks like.

The method row deserves a paragraph of its own because it is the row that decides audits. Name the basis in the RFQ and require the quotation to name it back: dichromate under ISO 15705, ASTM D1252, SM 5220, or the HJ 828 family; permanganate index under the applicable national convention; UV correlation with the acceptance clause of the governing regime quoted. Where the regime demands equivalence documentation, the acceptance test protocol, paired laboratory comparison data, and the commissioning validation belong to the purchase scope. The same naming discipline applies to the precision row: percent of reading means something only with the concentration and the standard it was measured against, and billing stations should demand the basis in writing.

The chloride and sample rows decide industrial applications more than any headline figure. Demand the tolerance at your stated matrix, with the masking chemistry named where mercury-free packages claim it, and let the dilution scheme carry the peaks; then align the sample conditioning to the method, filtration or settling appropriate to the digestion, flow and temperature at the instrument inlet, and a fast loop where solids load the line. The comparison table below compresses the evidence checks into the form a specification meeting can argue from.

The outputs and calibration rows close the specification. The instrument earns its price only when its number moves: 4 to 20 mA to the controller, relays for range and fault, a digital bus where the plant data path expects one, and telemetry where the regime polls the station; and the data stands only when its provenance moves with it, calibration records, standard checks, and audit trails exported on demand. The quotation that quotes the analyzer and leaves sample conditioning, data integration, and the validation program as site exercises is not a complete quotation.

Selection Decision Factors and RFQ Language

The decision sequence is short enough to run as a checklist in a specification meeting, and each step eliminates a class of quotations before price comparisons begin.

1. Read the permit or contract first: which parameter it names, COD, BOD, or both, which method or equivalence it accepts, and which regime, EPA program, HJ series, MCERTS, audits the station, because that reading eliminates families before any datasheet opens.
2. Characterize the matrix honestly: strength range at peaks, chloride, solids, color, salinity, and temperature, and let the tolerance rows, not the range row, do the eliminating.
3. Split the duties where the plant needs both: a wet chemistry instrument for the record and a UV sensor for control, which together usually cost less than forcing one instrument to do the other's job badly.
4. Set the cycle time from the control loop or the reporting interval, and accept the latency the chemistry imposes on record duty.
5. Price the operating reality: reagent consumption at the analysis rate, waste volume and its disposal class, and consumable service intervals, as annual lines next to the capital price.
6. Scope the sampling: intake point, fast loop with a centrifugal pump on slurries, filtration appropriate to the method, and the sample conditioning system delivered as part of the package.
7. Fix the data path: outputs, bus, telemetry, alarm relays, and the audit trail the regime will ask to see.
8. Fix the validation program in the purchase: standards, calibration and check intervals, paired laboratory comparisons where equivalence is claimed, and the records format.

Against the neighboring parameters, the comparison a specifier actually needs runs as follows, and it belongs here in the selection chapter rather than in the page title. BOD is the biological five day test the COD number usually substitutes for on speed; where the permit names BOD, the online COD becomes a correlation instrument with periodic laboratory BOD confirmation, and that program, not the analyzer, is the compliance object. TOC measures carbon and correlates to COD per site, faster on combustion instruments and cleaner on reagents, and a TOC analyzer is the standing alternative where the permit or the process frames the load as carbon. The simple sensors, conductivity, pH, turbidity, answer adjacent questions at a fraction of the cost and never replace the strength measurement. The COD analyzer wins when oxygen demand itself is the named or controlled quantity and the matrix tolerates one of its chemistries or holds a stable UV character; every alternative either measures a different quantity or needs this one's validation program to stand in for it. Choose from the permit paragraph, not from the parameter a supplier leads with.

The application map below turns that comparison into a table the project file can carry.

The RFQ block should be written once and reused. A complete block reads: matrix description with strength range at normal and peak, chloride, solids, color, and temperature; duty, compliance with the named permit and regime, process control with the loop latency, or billing with the precision basis; method and basis named, dichromate, permanganate index, or UV correlation, with the regime acceptance clause cited; range with dilution scheme and detection limit with its basis; cycle time and first-result latency; chloride tolerance and masking chemistry; reagent package type and consumption at the stated analysis rate, with waste volume, composition, and containment; sample requirements, temperature, pressure, flow, filtration, with the sample conditioning system in package scope and fast loop on slurries; outputs, 4 to 20 mA, relays, bus, telemetry, and alarm definitions; enclosure and ambient limits with heating or cooling where the shed demands; calibration and validation program, standards, intervals, records, and paired laboratory comparisons where equivalence is claimed; and service model, consumables schedule, and local support. A quotation returned without the method basis, the matrix confirmation, or the waste arithmetic is incomplete at any price.

Installation, Standards, Failure Modes, and Maintenance

Installation quality decides whether the instrument ever sees a representative sample. Put the intake where the stream is mixed and characteristic, not in a quiescent corner or a grease layer; run a fast loop on solids bearing streams, a small centrifugal pump moving the sample quickly from intake to analyzer shed so solids never settle in the line, with the analyzer tapping the loop through its own filtration; and deliver the sample at the temperature, pressure, and flow the datasheet states. Mount the cabinet level with service access to reagents and the reactor, contain the waste line into the hazardous waste drum without gravity surprises, and keep the enclosure within its ambient limits, heated where winter threatens reagents and cuvettes. Wire the outputs to the points the data path expects, and prove the alarm chain, range and fault, end to end before handover.

Commissioning practice proves the number, not just the plumbing. Calibrate against certified standards across the working range and record the curve; then run paired samples against the laboratory reference method across at least a normal load week, and file the comparison as the equivalence evidence where the regime asks for it. Verify cycle time from sample intake to report at the duty program, check the automatic standard function against its tolerance, and walk the data path from analyzer to record to telemetry confirming timestamps and scaling at every hop. On UV correlations, build the calibration from paired data spanning the load range, document it, and schedule its refresh, because the correlation is the instrument. The commissioning file, curves, pairs, timestamps, is what an inspector or a counterparty is shown, and what the annual validation is compared against.

The standards context is the reference method plus the monitoring regime. ISO 15705, ASTM D1252, and Standard Methods 5220 fix the dichromate basis; the HJ series fixes the Chinese dichromate and online fast digestion methods and the HJ 353 to 355 acceptance regime for continuous monitoring stations; JIS K 0806 family carries the Japanese online instruments; MCERTS governs where UK certification applies; and USEPA program rules, 40 CFR Part 136 context, govern the American method approval and monitoring practice. The analyzer is certified against one of these, the station is accepted under the regime's protocol, and the validation records keep the standing alive. A specification that names the method and the regime transfers the burden of proof to the supplier, which is where it belongs.

Failure modes in service follow a short and predictable list. Sample system faults first: a silted intake, a starved fast loop, a clogged filter, and the analyzer faithfully reports the chemistry of a stagnant line while the plant reads a confident wrong number; flow and elapsed-time alarms are the countermeasure, and the daily check walks them. Reagent exhaustion and expiry: an empty or aged reagent shifts the curve silently until the automatic standard check catches it, which is why the check interval is set tighter than the reagent calendar; level alarms and dated bottles are the operational discipline. Photometer and cuvette fouling: wetted optics film over, UV cuvettes grow biofilm, and cleaning, mechanical wipe or chemical, runs on schedule or the drift reads as process. Valve and pump wear in the metering path: the wet chemistry's moving parts, rebuilt on the consumable schedule, fail as short samples or carryover, seen as scatter between duplicates. Waste line faults: a blocked drain backs waste into the cabinet at the worst hour, and containment plus a level switch in the drum is cheaper than the cleanup. Drift of the UV correlation: character change in the stream, a new industrial input, seasonal shift, walks the calibration off, and the paired laboratory check at its scheduled interval is the only witness; treat every correlation instrument as provisional between confirmations. And freeze damage in unheated sheds: reagents and cuvettes crack, and the heated enclosure was the line item someone cut.

Maintenance intervals scale with analysis rate and matrix more than calendar. Daily, the walk checks flow alarms, reagent levels, waste level, and the last standard check result; weekly, the standards verify and the intake and filter condition are read; monthly, the laboratory pair refreshes the equivalence file and the UV correlation, and the cuvette or optics cleaning lands where scheduled; quarterly to semiannually, the metering path kit, valves and pumps, is rebuilt on the manufacturer's interval, and the reactor seals inspected. Every service records standards used, results found, and parts replaced, and those records are the audit trail and the reliability data in one file. None of these checks costs a tenth of one disputed discharge quarter, and every one produces a number the next RFQ can use, which is how a purchasing file gets smarter instead of just older.