Shielded Cable Gland: How to Match EMC Glands to Cable Screens
Match an EMC cable gland to the cable screen: foil, braid, spiral and combination constructions, termination geometry, bonding rules and sizing on jacket OD.
قراءة المزيد

An electrical enclosure arrives with a protection rating that was established on a test piece. Every hole drilled into it after that is a modification, and the gland that fills the hole restores the rating only if the hole, the face and the fixing are all correct. This is the part of enclosure work that gets the least attention and causes the most failures, because a gland is bought as a cable accessory and installed as a structural interface.
Most leak and ingress complaints on electrical enclosures trace back to that interface rather than to the gland. The gland is usually the right part. It has been fitted into a wall that is too thin, a face that is not flat, a thread too short to reach the locknut, or a knockout that left a step under the washer.
The same gland fitted to four different walls gives four different results. What changes is not the seal at the cable — that part is identical — but the behaviour of the wall around the thread.
| Enclosure wall | Behaviour under the gland | Fixing approach |
|---|---|---|
| Sheet steel, thin | Deflects under torque; the face can bow away from the washer | Reinforce or use a locknut and a stiff washer; keep torque modest |
| Moulded polycarbonate or ABS | Wall pulls into a cone; cracks at the thread root if over-tightened | Wide washer face; support the inside; avoid nut-and-gland squeeze |
| GRP or SMC | Hard and brittle; machined face is flat but chips at the edge | Drill, do not punch; deburr; use a flat gasket face |
| Cast aluminium | Thick, rigid, machined face; takes high torque | Thread directly; washer optional if the face is machined |
| الفولاذ المقاوم للصدأ | Rigid, but galls when a stainless locknut is used dry | Lubricate or use a dissimilar nut; do not over-tighten to stop galling |
Stainless deserves a note. A stainless gland into a stainless wall with a stainless locknut is a galling risk, and the usual outcome is a joint tightened far past spec because it felt loose. It then either cracks the weld or work-hardens and cannot be dismantled. A nickel-plated brass locknut on a stainless thread is common practice for this reason.
Gland thread length has to cover the wall plus the washer plus the locknut, plus any gland plate. Two failures follow from ignoring it. If the thread is too short, the locknut cannot engage fully and the gland is held by a few turns — enough to feel tight and not enough to stay tight under vibration. If the thread is too long, the gland protrudes into the enclosure and fouls terminals, DIN rail or the door swing.
The locknut question is often answered by habit. Three conditions require one, and one condition removes it.
| Condition at the wall | صامولة القفل |
|---|---|
| Wall thin enough to distort under torque | Required — spreads the load and stiffens the joint |
| No access to the inside face | Required — this is the only way to clamp the gland |
| Conduit or heavy cable that could rotate the gland | Required — locks the body against turning |
| Thick cast face with a machined thread | Not required — the thread is the fixing |
What a locknut is never a substitute for is a flat sealing face: adding a nut does not flatten a bowed wall. Long-thread variants exist for thick walls or gland plates, and specifying one is cheaper than fitting a spacer.


An enclosure knockout is convenient and is not a hole. The material is removed in a step, leaving a rim that sits proud of the surrounding face, and a sealing washer placed over that rim contacts the high point and bridges the rest. The result is a joint that passes a visual inspection and fails a hose test.
Three remedies follow. Where a knockout is used, check the rim with a straight edge and file it flat if it stands proud. Where the hole is drilled, use a punch or a hole saw with a pilot and deburr both sides — the inside burr matters as much as the outside one, because the inside face is what the locknut bears on. Where the enclosure is brittle GRP or a moulding, drill rather than punch, since a punch can star the gelcoat.
Hole size is its own subject: an oversized hole cannot be sealed by any washer, and an undersized one cross-threads the body. The hole size chart gives the drilling dimension for each thread, including the tolerance that separates a thread that fits from one that leaks.
An IP rating on an enclosure applies to the enclosure as tested, with its door closed and its entries either fitted with the specified accessories or not present. Fitting a gland transfers a rating only in the sense that the gland has its own rating and the joint between them has to be sound. The two ratings do not multiply, and they do not compensate for each other.
This has two consequences. An IP68 gland in an IP54 enclosure produces an IP54 enclosure with a very good entry, and if the door gasket is the weak point the entry work has bought nothing. And the weakest entry sets the level: one open knockout defeats every correctly fitted gland on the box.
The door is where most enclosure ingress actually happens. A flattened gasket, a warped door, a missing breather filter or a hinge that no longer pulls the door square will all put water inside a box with perfect entries — and the gland gets blamed because it is the part someone can see water arriving at.
Where several cables enter close together, a gland plate is usually better than holes in the enclosure wall. The plate is a removable panel, typically thicker than the wall, that carries all the entries and bolts to a flange on the enclosure. It solves four problems at once: it provides a flat, rigid sealing face; it can be machined on a bench; it allows the whole cable group to be disconnected and re-entered; and it protects the enclosure’s own coating.
The cost is an extra gasket line between plate and enclosure, which is itself a leak path. A plate bolted to a painted flange, with no gasket and no bare metal contact, is both a leak path and a broken earth path — two faults in one joint.


A metal gland in a metal enclosure looks like an earth path and often is not. Paint, powder coat, anodising, a gasket and thread sealant all sit between the gland body and the enclosure metal, and any one of them can raise the resistance of the joint above what a protective conductor needs.
Three cases need explicit attention: an enclosure with a coated flange, an enclosure where the gland is fitted through a plastic gland plate, and an installation where the gland will be dismantled for maintenance. In each case the earth path must be independent of the thread — an earth tag under the gland, a bonding conductor to the enclosure stud, or both. The earthing and bonding guide covers the measurement side of this in detail, and it is worth reading before an enclosure is signed off rather than after an inspection finding.
Entry orientation decides which failures are possible. A bottom entry sheds water away from the seal and is the default outdoors. A top entry delivers water directly to the dome nut, and if the cable arrives from above with no drip loop, the seal is stopping a stream rather than a splash.
The better answer is often a different route rather than a different gland: bring the cable up past the enclosure and enter from below, or add a drip loop outside the box. Where a top entry is unavoidable, a shroud over the gland extends life considerably, and a longer body keeps the seal further from standing water.
Condensation is the other half of this subject and is often misdiagnosed as a leak. A sealed enclosure that cycles in temperature draws moist air in and condenses it on cold metal, and the water appears at the bottom of the box near the lowest gland. A pressure-compensating breather or a drain addresses the cause; a new gland does not. The diagnostic sequence is in why cable glands leak, which starts from the direction the water arrived.
Most entries are added after the enclosure is built, often after it is energised. Four rules keep a retrofit from becoming a fault.
| Rule | السبب |
|---|---|
| Isolate before drilling, and mask the interior | Swarf is conductive, gets into terminals and is very hard to remove |
| Drill from outside with a pilot, then deburr both faces | Burrs on the inside face stop the locknut seating flat |
| Choose a flat area, clear of ribs, corners and the door swing | A washer cannot seal on a curved or interrupted face |
| Fit a gland sized to the cable, not to the hole you have | Adapting a gland to an existing hole is how leaks start |
Where the existing hole is already the wrong size, a reducing or enlarging washer is legitimate and a sealant-filled joint is not. If neither is possible, the correct fix is a new hole in a sound position and a blind plug in the old one — cheaper than the rework that follows a sealant repair.
Only if the thread makes bare metal contact and stays that way. Coating, gaskets and sealant break the path, and a dismantled joint breaks it again. Use a tag or a conductor where the path has to be guaranteed.
No. It is needed on thin walls, where the gland must be locked against rotation, or where access from inside is impossible. On a thick machined face it adds nothing.
You can, and it will not raise the enclosure’s rating. The enclosure rating is set by its weakest element, which is usually the door seal or an unfitted knockout.
Usually condensation rather than ingress. Check the distribution of the water and whether the cable sheath is wet internally before replacing any gland.
Whichever value the wall can take without distorting, which is usually below the gland’s own maximum. Tighten until the washer is in full contact; more torque narrows the contact ring instead of widening it.
Send the enclosure drawing with wall material and thickness plus the cable list, and we will mark which entries need a locknut, which need a longer thread, and where a gland plate beats nine separate holes. The nickel-plated brass range covers the general industrial case, and the M20 standard series is where most enclosure schedules start.