Solenoid valve leakage: why it is rarely the seal
First separate internal leakage from external leakage — they have completely different causes. If it is internal, check seat concentricity and spring eccentricity before you touch the seal. Seal material gets replaced constantly because it is visible, cheap and available, which is exactly why it is the most common wrong answer.
Why this happens
Seals get replaced because they are the easiest thing on the valve to replace. They are visible, cheap, in stock, and replacing one feels like progress. The result is that seal material is the most common wrong answer in valve leakage troubleshooting, and the real cause survives.
Two things hide behind the word “leakage”, and they need to be separated before anything else:
- Internal leakage — media passes the closed seat from inlet to outlet. The valve fails to shut off.
- External leakage — media appears at a body joint, a stem, or a lead exit. The valve fails to contain.
Confusing these guarantees a wasted first attempt. Internal leakage is about seat geometry and force. External leakage is about compression, surface finish and thermal cycling.
Check these in order
1. Separate internal from external. Blank the outlet, pressurise the inlet, leave the coil de-energised, and watch for pressure downstream and for traces at the body. Do this before you disassemble anything — the symptom you observe cold will guide the whole investigation.
2. Check seat concentricity. A seat that is not perpendicular to the plunger axis makes contact on one side first. The seal cannot compensate for a geometric error indefinitely; it deforms locally, takes a set, and leaks. This is the first thing I measure, and it is the most frequent real cause of “leaky seal”.
3. Check spring eccentricity. A return spring that is not square, or a spring pocket that is off-centre, applies a side load. The plunger tilts, the seat contact becomes asymmetric, and leakage appears at one point on the circumference. Note where on the circumference the witness mark sits — a single-sided mark is a strong hint towards eccentricity rather than material.
4. Check compression. Under-compression leaks. Over-compression destroys the elastomer — it takes a permanent set, then leaks. Both leave a deformation mark, which is why “the seal looks deformed” is weak evidence. Compare actual compression against the material’s recommended range.
5. Check seat surface finish and contamination. A particle embedded in the elastomer holds the seat open. A seat that is too rough wears the seal; one that is too smooth can prevent sealing with some elastomers because there is nothing for the material to flow into.
6. Check differential pressure and its direction. Most valves seal in one direction. Reversed pressure unseats the seal, and this is common in field installations where nobody checked the intended flow direction against the actual plumbing.
7. Check temperature and compression set. This is where material genuinely becomes the answer. High temperature plus continuous compression produces permanent set, and the seal stops recovering. But note the order: this is the seventh thing to check, not the first.
8. Only now: material compatibility. Chemical compatibility tables solve the chemistry. They do not solve temperature, compression set, or permeation. A material can be fully compatible with the media and still be the wrong choice.
What actually to change
| Finding | What to change | Why not the seal |
|---|---|---|
| Witness mark on one side only | Fix seat perpendicularity or spring squareness | A new seal takes the same set within weeks |
| Over-compression | Correct the pocket depth | Material change does not change geometry |
| Particle embedded in the seat | Filtration upstream, seat flush | Replacing the seal restores function briefly |
| Reversed pressure | Correct the installation or specify a bidirectional design | No material seals against reversed pressure |
| Permanent set at high temperature | Now the material is the answer: change grade or reduce temperature | This is the one case where the seal is the fault |
| External leak at the lead exit | Potting or a vented seal design | This is a containment problem, not a seat problem |
When it IS the harder problem
The leak follows the body, not the seal. Swap seals between a leaking and a non-leaking unit. If the leak follows the body, the seal was never the problem, and you have saved yourself a material qualification. This is a ten-minute test that settles an argument that otherwise runs for weeks.
Intermittent and temperature-linked. If it seals cold and leaks hot, you are seeing differential expansion between the body and the seat, or compression set accumulating in real time. Measure the seat geometry hot, not cold.
Only after a certain number of cycles. If it leaks from new but not after a hundred cycles, the assembly process is bedding the seat in. If it seals when new and leaks after a hundred thousand cycles, you have wear or fatigue, and the answer is surface finish or stroke, not material grade.
Nobody has confirmed the pressure direction. I have lost more time to this than to any material question. Confirm the intended flow direction against the actual installation before investigating anything else.
A note on what this page is
This is a personal notebook, not a product page. I write down the checks that actually decide the outcome and the order to run them in, including the ones I got wrong first.
solenoid valve leakingsolenoid valve internal leakagesolenoid valve seal materialsolenoid valve not sealingvalve seat concentricity
Frequently asked
- How do I tell internal leakage from external leakage?
- Pressurise the valve with the coil de-energised and blank off the outlet. Any rise in pressure downstream of a closed valve is internal leakage past the seat. Any trace at the body joint, stem or lead exit is external. They have different causes and different fixes, and lumping them together is the most common reason the first repair attempt fails.
- The seal is deformed. Is that not proof the material is wrong?
- Not on its own. A seal can deform because it was over-compressed, because the seat it sits on is not concentric, or because the temperature exceeds the elastomer's range — all of which leave the same witness mark. Check the seat and the compression first; if those are correct and the deformation persists, then question the material.
- Why does a new valve leak and an identical old one not?
- Usually a seat or guide tolerance at the edge of the band, not a seal batch difference. If you must triage quickly, swap the seals between a leaking and a non-leaking unit. If the leak follows the body rather than the seal, the seal was never the problem.
- Can differential pressure direction cause leakage?
- Yes, and it is regularly overlooked. Most valves are designed to seal in one direction of pressure. Reversed pressure unseats the seal and no material change will fix it. Confirm the intended flow direction against how the valve is actually installed.