Hitting the target quantity precisely in a chemical transfer comes down to two things the batch controller does: it steps the flow rate down as the batch nears its setpoint, and it anticipates the extra product that will pass while the valve is closing. By learning the overrun measured in previous batches, the controller issues the close command slightly ahead of the target, while the meter is sized to stay within its measuring range during the slow final stage. When the transfer ends, the recipe ID, target and delivered quantity, temperature and density, start and end times and any alarms are gathered into a single batch record.
Where the batch controller sits in the overall loading chain is explained on the truck loading systems page. This page looks at recipe-driven chemical transfers and the factors that decide cutoff accuracy and record quality.
Where batch metering is used with chemicals
· Filling tankers or containers: Delivering exactly the ordered quantity into a tanker or transport container, no more and no less.
· Recipe feeds to a process: Charging a reactor or blend tank with the raw material quantity defined by the recipe; an error here spoils the product itself.
· Tank-to-tank transfers: Moving a set quantity from storage to a day tank or preparation tank and reconciling it against inventory.
Setting the target: mass or volume?
Chemical recipes are most often defined by mass. A Coriolis mass flow meter measures mass flow rate, density and temperature at the same time and derives volume from mass and density, so a mass-based batch can use a mass target directly without any conversion. With a volumetric meter, the measured volume has to be converted to mass or to volume at reference conditions using density and temperature, and the inputs to that conversion also affect batch accuracy. How the invoiced quantity is defined is a separate question, covered on the chemical custody transfer metering systems page.
What throws off the cutoff
Flow does not stop the instant the close command is issued. If the difference varies from batch to batch, the target will be missed. The main contributors and how to reduce them:
Factor | Effect on cutoff | How to reduce it |
Valve closing time | Product passing while the valve closes is added on top of the target | Drop to a low rate for the final stage; tune valve speed and closing characteristic to the product |
Flow rate at closure | The higher the rate, the larger the overrun during closing | Two or more stages of flow reduction |
Meter and controller latency | Pulse resolution and signal update interval make the controller see the target late | Adequate pulse resolution and a fast-responding signal path |
Pump stop sequence | If the pump stops after the valve, line pressure rises; if it stops first, flow becomes erratic | Define the pump and valve sequence in the control logic |
Viscosity and temperature changes | The same valve movement passes a different quantity, so the learned compensation drifts | Keep separate compensation values by product and temperature range |
Uncertain line contents | An empty or partly full line at the start creates a gap between what the meter sees and what is delivered | Define the line condition at batch start and end |
Staged closing and overrun compensation
In two-stage batch control, the transfer runs at high rate, the rate is cut once a set quantity remains, and the last portion is completed at low rate. Many batch controllers average the overrun measured over recent batches and recalculate the valve closing point automatically. That compensation works well under steady conditions; after a product change or a marked temperature shift, the learned value should be reset or kept separately for each product. And since closing a valve too abruptly can cause a pressure surge in the line, closing speed is set with line integrity in mind, not cutoff accuracy alone.
Meter accuracy in the low-flow stage
The low rate used in the final stage has to stay within the range where the meter measures accurately. At low flow, Coriolis meter accuracy is governed by zero stability: as flow drops, zero stability becomes a larger share of the flow and the percentage error grows. Pushing the rate too low to sharpen the cutoff can therefore cost measurement accuracy. The slow-stage rate should be chosen together with the low-flow performance stated on the meter's datasheet.
Line contents, product changeovers and line clearing
In chemical service, the line is often emptied between batches or flushed for a different product. For metering, this reduces to one question: what volume sits between the quantity the meter registers and the quantity that reaches the destination? Whether the line is full at batch start, and whether the section between meter and destination is cleared with nitrogen or another method at the end, should be spelled out in both the recipe and the control logic. Any product or solvent used for flushing should be recorded on its own and kept out of the batch quantity.
Batch records and traceability
In recipe-based chemical transfers, the record matters as much as the measurement. A typical batch record contains:
· Recipe or order ID, product and destination
· Target quantity, delivered quantity and the difference
· Average temperature and density, and the conversion method if one is used
· Start and end times, interruptions and alarms
· Meter factor applied and a log of parameter changes
· Operator sign-off and sample details where applicable
Batch controllers can store transactions with a date and time stamp and pass the data on to DCS, SCADA or enterprise systems. Where records feed invoicing or quality documents, an audit trail showing who changed which parameter and when is usually required as well.
What TLY Enerji contributes
For chemical batch transfers, TLY Enerji can provide engineering support in selecting the meter and control valve, integrating the batch controller into the plant's PLC or DCS architecture, setting up the recording and reporting flow, and in site installation, testing and commissioning. Responsibility boundaries within the control logic are agreed at project start based on the existing automation infrastructure.
What we need to size a batch metering system
· Range of batch sizes and the acceptable cutoff deviation
· Whether the target is mass or volume and how the recipe is defined
· Product list with viscosity and temperature ranges
· Transfer rate and maximum allowable transfer time
· Whether the line stays full or is emptied between batches, and the flushing and clearing method
· The system that will receive the records and the required report content
Related pages
· Chemical metering skids: Technology and material selection for the meter used in batching.
· Solvent batch transfer systems: Keeping solvent purity intact along with the quantity.
· Batch metering systems for pyrolysis oil: Traceability for a product whose quality varies between batches.
· Chemical custody transfer metering systems: Whether the invoiced quantity is based on mass or volume.
· Additive injection systems for chemical terminals: Adding the additive quantity to the batch record.
Batch metering FAQ
Why does the deviation from target change from batch to batch?
After the close command, the valve still takes time to shut, and the quantity passing in that interval depends on the flow rate at closure, product viscosity, temperature and the pump and valve sequence. As those conditions change, so does the overrun. Staged closing, product-specific compensation values and keeping the line condition the same for every batch all reduce that variability.
Can one overrun compensation value serve every product?
Generally not. The compensation is based on the average overrun measured in recent batches, and that value loses its meaning when you switch to a product with different viscosity or density. Keeping a separate value per product, or restarting the learning at each changeover, limits the deviation in the first batches after a switch.
How far can the final-stage flow rate be reduced to improve batch accuracy?
Down to the lowest rate at which the meter still measures accurately. With Coriolis meters in particular, accuracy at low flow is governed by zero stability, and the percentage error grows as flow falls. Going below that limit may give a sharper cutoff, but it makes the measured quantity less reliable. Set the slow-stage rate together with the meter datasheet and the transfer time target.
Is nitrogen or flushing product used to clear the line included in the measurement?
That decision has to be made up front and written into the recipe and control logic. Whether the section between meter and destination is cleared directly affects the delivered quantity. Product or solvent used for flushing should be recorded separately from the batch quantity; otherwise the recipe quantity and inventory reconciliation will not match.