HPLC
HPLC Pressure Problems: High, Low and Fluctuating Pressure Explained — Analytical Instrument Guide
What normal HPLC pressure looks like, and how to find the cause of high, low or fluctuating pressure on Shimadzu, Waters and Agilent systems, one connection at a time.
System pressure is the first number an HPLC user looks at, and for good reason. It tells you, at a glance, whether the flow path is open, whether the pump is delivering steadily and whether something has changed since yesterday. When the pressure is too high, too low or will not stay still, the cause is almost always somewhere you can reach with a wrench and a little patience.
This guide explains what normal pressure looks like, then walks through the three pressure faults we are called for most often: high pressure, low pressure and fluctuating pressure. It applies to Shimadzu, Waters and Agilent systems alike, because the physics of the flow path is the same on all of them.
Know your normal pressure first
There is no single "correct" pressure for an HPLC. The pressure your system shows depends on:
- Flow rate. Double the flow and the column pressure roughly doubles.
- Column length and particle size. A longer column, or smaller particles, give a higher pressure at the same flow.
- Mobile phase viscosity. Water and methanol mixtures are more viscous than either solvent alone, so the pressure rises in the middle of a methanol and water gradient and falls again at the ends. Acetonitrile mixtures are less viscous and run at a lower pressure.
- Temperature. A warmer column oven lowers the viscosity and the pressure. A lab that cools down overnight shows a slightly higher pressure in the morning.
The most useful habit is simple: write down the pressure for each method when the column is new, at the method flow and temperature, and keep it with the method. That number, not a figure from a brochure, is the reference you compare against when something seems wrong. A rise of a few percent over weeks is normal column ageing; a sudden jump is a fault.
Quick reference
| What you see | Most likely cause | Check first |
|---|---|---|
| Pressure suddenly much higher than your reference | A blockage: inline filter, guard column, column inlet frit, a capillary, the injector | Remove the column and compare |
| Pressure rising slowly over days | Particles or sample matrix building up at the column inlet | Guard column, inline filter, sample filtration |
| Pressure lower than normal, steady | A leak, the purge valve open, wrong flow or wrong mobile phase | Look for wet fittings, close the purge valve |
| Pressure zero or close to zero | Purge valve open, pump not primed, a large leak, empty bottle | Bottle level, purge valve, prime the pump |
| Pressure going up and down regularly | Air in a pump head, a sticking check valve, a worn seal | Purge each channel, degasser |
| Pressure jumping irregularly | Air reaching the pump, a failing degasser, poor solvent mixing | Degasser, solvent preparation |
| Pressure limit error and the pump stops | A blockage, or a pressure limit set too low for the method | Compare with your reference, check the limit setting |
High pressure: find the blockage by elimination
High pressure means something in the flow path is partly blocked. The fastest way to find it is by elimination, one connection at a time, starting from the column.
- Stop the flow and let the pressure fall to zero before you open any fitting.
- Disconnect the column inlet and route the line to a waste beaker.
- Restart the pump at a low flow and read the pressure.
- If the pressure is now low, the restriction is in the column, the guard column or an inline filter. Replace the guard cartridge or the inline filter first; they are there to block before the column does.
- If the pressure is still high without the column, move one connection closer to the pump: the column oven tubing, the injector or autosampler, then the pump outlet. When the pressure drops, the part you just took out of the path is the blocked one.
Common culprits, in the order we usually find them:
- The guard column or inline filter. Cheap, quick to change and the most frequent cause.
- The column inlet frit. Unfiltered samples, precipitated buffer and particles from wearing seals collect here. Reversing a column to flush it is allowed only for some columns; check the column maker's instructions before you try it, and never reverse a column into the detector.
- The autosampler needle, needle seat or rotor seal. A worn rotor seal sheds particles downstream, and a needle seat can block with sample matrix.
- A capillary. Narrow tubing blocks easily, especially after a buffer has been left to dry in it.
- Buffer precipitation. Switching straight from a phosphate buffer to a high percentage of organic solvent can make the salt come out of solution inside the system. Flush the buffer out with water before you go to organic.
Prevention is cheaper than repair. Filter samples and buffers, use a guard column for dirty matrices, and flush buffers out with water at the end of every day.
Low pressure: look for the leak
If the pressure is lower than your reference and steady, solvent is going somewhere other than through the column, or the pump is not delivering what it should.
- The purge valve or drain valve is open. It sounds too simple, but it is the most common cause after a priming step. Close it fully.
- A leaking fitting. Run a dry tissue along every connection from the pump to the detector. A fitting that is wet or shows white salt deposits is leaking. Tighten gently with the pressure at zero, and replace a fitting with a crushed ferrule rather than forcing it.
- Worn pump seals. Solvent or salt crystals under the pump head point to worn piston seals. A seal leak also shows up as slowly shifting retention times.
- Wrong method settings. Check the flow rate, the mobile phase composition and the column in the method. A method copied from another system sometimes carries a different flow.
- An empty or nearly empty bottle, or an inlet line that has lifted out of the solvent.
Fluctuating pressure: usually air or a valve
A pressure trace that goes up and down is the pump telling you that it is not delivering a steady flow. Plot the pressure in your data system for a few minutes and look at the pattern.
- Regular ripple that repeats with each pump stroke points to a single pump head or channel: air in that head, a sticking or dirty check valve, or a worn seal. Run each solvent channel on its own to find which one. Purge that channel, then replace a blocked solvent inlet filter or clean or replace the check valve.
- Irregular jumps usually mean air is reaching the pump: poorly degassed solvent, a failing degasser channel, a loose inlet line or an inlet filter that is clogged and causing cavitation.
- Ripple that started after running buffers is often salt on a check valve or a piston. A long water flush sometimes frees it. A seal wash kit prevents most of this when you run buffers every day.
Fluctuating pressure also shows up in your chromatograms as a noisy baseline and shifting retention times. Our baseline noise and drift guide explains how to tell a flow problem from a detector problem with a simple stop-flow test.
Brand-specific notes
The approach above works on every HPLC, but each family has its own common weak points. For model-by-model details see:
- Agilent HPLC pump troubleshooting for the 1100, 1200 and 1260 Infinity pumps, including the purge valve frit.
- Shimadzu LC-2010C HT pump problems for high pressure, fluctuation and the "no pump home" error.
- The Waters Alliance e2695 troubleshooting page for the Alliance separations module.
Frequently asked questions
What is a normal pressure for an HPLC system?
There is no single normal value. It depends on the flow rate, the column length and particle size, the mobile phase and the column temperature. The right reference is the pressure your own method showed when the column was new, recorded at the method flow and temperature. Compare against that number, not against another lab's system.
Why does my HPLC pressure rise during a gradient and then fall?
Mixtures of water with methanol are more viscous than either pure solvent, so the pressure rises in the middle of the gradient and falls again as the composition moves towards either end. This is normal and repeats from run to run. A pressure curve that changes shape from one run to the next is worth investigating.
Can I keep running if the pressure is a little high?
If the pressure has crept up slowly and is still well inside the limits of the column and the system, you can usually finish the work and plan maintenance, such as replacing the guard column or inline filter. A sudden jump is different: stop, find the blockage, and do not raise the pressure limit to get past it, because that moves the problem into the column or the injector.
How often should pump seals and check valves be replaced?
It depends on how many hours the pump runs and what it pumps. Buffers and high flow rates wear seals faster. Rather than waiting for a leak, replace seals and check valves at planned preventive maintenance, based on the instrument's usage. An AMC puts this on a schedule.
When should I call an engineer?
Call when the pressure fault stays after you have purged, flushed and checked the column and fittings; when a valve, seal or sensor needs replacing and you do not have the parts or training; or when the pump shows motor, drive or electronics errors.
When to call us
Our engineers service HPLC systems for laboratories in Thane, Mumbai, Navi Mumbai, Pune and Ambernath. When you contact us, send the instrument model, the method flow and mobile phase, your reference pressure and the pressure you see now, and what you have already tried. A screenshot of the pressure trace saves time.
Brand and model names are used only to identify instruments. Vision Analytical is an independent service company and is not affiliated with any instrument maker.
