Quick answer
When a bag-filter compressed-air header loses pressure during pulse cleaning, determine whether the air source cannot supply the instantaneous demand or whether the cleaning system is consuming more air than intended. Diagnose compressor/header pressure, receiver volume, regulator and isolation valves, pulse-valve timing, diaphragm leakage, nozzle/orifice condition and simultaneous air consumers. Increasing compressor pressure is not the first step if valves are leaking or too many pulses fire together.
Problem / job
Pulse-cleaning pressure falls sharply, later rows clean weakly, filter DP rises or cleaning becomes inconsistent during a pulse sequence.
Evidence
Capture upstream plant-air pressure, local receiver/header pressure, pulse timing, number of valves firing, regulator position, compressor load and filter DP.
Decision
Separate insufficient air supply/receiver capacity from excessive pulse demand, leaking diaphragms, restrictions and poor sequencing.
As-found evidence capture
Trend the compressed-air pressure upstream of the local regulator and in the filter header/receiver during a complete cleaning sequence. Mark each pulse event. If upstream plant pressure stays stable while local pressure collapses, the restriction or demand problem is local. If both collapse, investigate the shared air system and concurrent users.
Listen for continuous leakage between pulses and inspect whether any pulse valve remains partly open. Record recent changes to pulse duration, interval or number of valves fired together.
Seven-step diagnostic workflow
1. Compare upstream and local pressure
Measure before and after the regulator/isolating components. A large dynamic pressure drop can reveal an undersized/restricted regulator, partly closed valve, clogged line or inadequate supply pipe.
2. Check receiver/header recovery
Observe how quickly local pressure recovers between pulses. Slow recovery suggests restricted supply or inadequate stored-air capacity for the programmed pulse interval.
3. Verify pulse timing
Confirm pulse duration, interval and whether multiple valves fire simultaneously. A PLC or timer change can create excessive instantaneous demand without any hardware failure.
4. Inspect pulse valves and diaphragms
A torn diaphragm, contaminated pilot or sticking valve can leak continuously or stay open too long. Compare audible/pressure response between rows and service faulty valves according to the manufacturer procedure.
5. Check regulators, filters and isolation valves
Verify valves are fully open as intended and regulators/air-preparation components are not restricting flow. Static pressure can look normal even when dynamic flow capacity is inadequate.
6. Check competing compressed-air demand
Determine whether pneumatic conveying, aeration or other large consumers operate simultaneously. A shared-header event may require sequencing or supply-system correction rather than modification of the filter alone.
7. Prove pressure recovery during a full cleaning cycle
After correction, run the normal cleaning sequence and confirm header pressure remains inside the filter/OEM operating range, each pulse is consistent and filter DP responds normally.
Evidence-to-decision table
| Evidence | Likely direction | Next check |
|---|---|---|
| Plant header stable, local header collapses | Local restriction/capacity/demand | Regulator, line, receiver, pulse sequence |
| Plant and local pressure collapse together | Shared air-supply issue | Compressor and concurrent users |
| Pressure falls even between pulses | Continuous leak/stuck valve | Pulse valves and diaphragms |
| First rows strong, later rows weak | Poor recovery/receiver capacity | Pulse interval and supply flow |
| Collapse began after PLC/timer change | Excess pulse demand | Duration, interval, simultaneous pulses |
Field measurements that make the diagnosis conclusive
Use a fast pressure trend rather than a single gauge reading. Record plant-air pressure upstream of the filter branch, pressure immediately before the local regulator, and receiver/header pressure at the filter. Align those traces with the pulse command sequence. A healthy local system should show a repeatable pressure dip and recovery pattern. If the upstream trace remains stable while the local header repeatedly collapses, concentrate on the branch restriction, regulator, receiver, pulse sequencing and valve leakage. If both upstream and local pressure fall together, the filter is probably exposing a shared compressed-air capacity problem rather than creating it by itself.
Compare the first, middle and last cleaning rows. Progressive weakening across the sequence points toward inadequate recovery between pulses. One row causing an unusually deep pressure drop suggests a leaking or slow-closing diaphragm valve, abnormal orifice/nozzle condition, or a control signal that holds that valve open longer than the others. Record the exact row and pulse timing before dismantling anything so the same condition can be reproduced after repair.
Minimum evidence sheet
- Plant-air pressure before the filter branch.
- Local receiver/header pressure before, during and after each pulse.
- Pulse duration and interval from the controller.
- Which valves fire together, if any.
- Compressor loading/unloading state during the test.
- Filter differential pressure before and after a full cleaning cycle.
- Any continuous audible leakage between pulses.
How to separate insufficient supply from excessive demand
Supply limitation and excessive demand can produce the same symptom, so use cause-and-effect timing. If header pressure is already low before cleaning starts, investigate the shared air system, branch isolation valves, regulator capacity and competing users. If pressure is normal before the first pulse but collapses only when the cleaning sequence begins, quantify the cleaning demand itself. Review whether pulse duration was increased, intervals were shortened, more valves were configured to fire simultaneously, or a valve is leaking continuously.
Do not compensate for a leaking or poorly sequenced cleaning system by raising the plant-air setpoint. That can increase energy use and stress components without correcting the cause. The correct endpoint is stable cleaning performance at the pressure range specified by the bag-filter and pulse-valve manufacturer, with predictable receiver recovery and no continuous leakage.
Return-to-service acceptance criteria
Accept the cleaning-air system when header pressure stays stable through a full pulse sequence, pressure recovers predictably between pulses, no valve leaks continuously, filter DP remains controlled and the system operates at the filter manufacturer’s required pulse pressure without unnecessary compressor demand.
Common mistakes
Do not raise plant-air pressure before checking leaking valves and timing. Do not diagnose from static pressure alone. Do not shorten pulse intervals without confirming receiver recovery. Do not ignore other large compressed-air consumers. Finally, use the filter/pulse-valve supplier’s pressure limits rather than generic settings.
Build a stronger cement maintenance library
The Infinity for Cement Organization technical package contains 900+ technical files for cement professionals. It is a one-time $249.99 purchase, with instant download immediately after payment and lifetime updates forever at no additional cost.