Home › Blog › Medical › Medical Electronics Sealing, Thermal & EMI Protection › Thermal pump-out and EMI seam failures
Pump-Out and Open Seams: Why a Medical-Device Thermal Pad Dried Out and an EMI Gasket Stopped Shielding
By the H-O Products applications engineering team · Updated June 2026 · ~10 min read
Two medical-device-electronics parts can be correctly chosen and still fail over time: a thermal interface that loses contact as it pumps out under thermal cycling, so the board runs hotter month over month, and an EMI gasket that stops shielding because its seam opened or it grounded against a coated surface. This article walks both failure modes and the design choices that hold contact over the life, with methods named by designation.
The thermal pad that pumped itself out of the gap
A thermal interface works only while it stays in continuous contact with both surfaces, and thermal cycling works against that. Each time the board heats and cools, the two surfaces expand and contract at different rates and the gap breathes slightly, and a softer or less stable interface material can be slowly worked out of the gap, a process called pump-out, until contact is lost at the edges and the thermal path degrades.
The board then runs hotter month over month for no obvious reason. The defense is an interface material and a mounting that hold contact over cycling: a boron-nitride silicone thermal gap pad chosen with enough conformability to stay seated, or a graphite thermal interface where a thin, stable, well-clamped path resists pump-out. The mounting pressure and the gap control matter as much as the material.
Why mounting pressure and gap control decide pad life
A gap pad has a working compression range, and life depends on landing inside it. Too little compression and the pad does not fully wet both surfaces, so contact is marginal from day one; too much and the pad is over-stressed and more prone to pump-out and set. The gap has to be controlled so the installed compression sits in the pad's recommended range, which means the stack tolerances and the clamp load are design inputs, not afterthoughts.
A pad sized to the nominal gap with no allowance for tolerance can end up barely touching at the loose end of the stack, which is contact loss before any cycling even begins.
The EMI gasket that stopped shielding
An EMI seam fails the same way the imaging page describes, and it is worth repeating because the medical-device enclosure has the same trap. A conductive EMI shielding elastomer only carries interference current to ground if it touches a conductive surface, so a gasket landing on a painted or anodized flange is installed but not grounded.
And the seam has to stay continuous and stay compressed: a lid that relaxes over time, a fastener that loosens, or a corner left open all break the conductive perimeter. A conductive foil tape can bridge a specific discontinuity, but the design intent is an unbroken, compressed, bare-metal-contacting seam. Shielding effectiveness is verified on the assembled enclosure by the device builder; the gasket only provides the path.
Contact over the life is the shared theme
Both failures are about maintaining contact over the service life, not about the first day. The thermal pad has to stay seated against pump-out and stay in its compression range across stack tolerance; the EMI gasket has to stay grounded against a conductive surface and stay compressed across relaxation and loosening. A correctly chosen material defeated by lost contact is the pattern, which is why the mounting pressure, the gap control, the contact-surface finish, and the long-term clamp retention are design inputs named before the part is cut.
The material does its job only while contact holds.
Thermal and EMI Failure Modes and the Design Defense for Each
| Symptom | Cause | Design defense |
|---|---|---|
| Board hotter month over month | Thermal pad pump-out | Conformable pad held in compression range |
| Marginal cooling from day one | Too little compression / loose gap | Gap control to the pad's range |
| Gasket installed, still emits | Grounded against paint / anodize | Bare-metal or conductive contact zone |
| Shielding faded over time | Seam relaxed / fastener loosened | Retained clamp, continuous compressed seam |
Illustrative and relative only; confirm each grade on its current TDS. Thermal behavior per the maker data; the grounding scheme, clamp retention, and thermal and emissions qualification belong to the device builder.
The pattern: the material was right, the contact was not held
Neither of these is a material defect. The thermal pad was the right interface but lost contact to pump-out or a loose gap; the EMI gasket was the right conductive grade but lost its path to a coated surface or a relaxed seam. The fix in both cases is to design for contact over the life: hold the pad in its compression range across stack tolerance and choose it to resist pump-out, and ground the gasket on bare metal with a retained clamp and a continuous seam.
Name those inputs and the chosen material keeps doing its job.
How H-O Products helps
H-O Products is a precision die-cut converter, not a raw-material producer and not the device maker. We die-cut, kiss-cut, laminate, and kit boron-nitride and graphite thermal interfaces and conductive EMI gaskets and foil to your drawing, with material traceability and lot-code TDS records under our ISO 9001:2015 certified quality management system in Winsted, Connecticut.
We help you reason about pump-out, compression range, gap control, grounding contact, and seam retention, and we cut the parts so the material holds contact over the life, with grade-level values confirmed against the maker's current data.
The mounting design, grounding scheme, and thermal and emissions qualification stay with the device builder.
Frequently asked questions
What is thermal pad pump-out and why does it make my board run hotter over time?
Pump-out is the slow loss of a thermal interface from the gap under thermal cycling. Each heat-and-cool cycle, the two surfaces expand and contract at different rates and the gap breathes, and a softer or less stable material can be worked out of the gap until contact is lost at the edges and the thermal path degrades, so the board runs hotter month over month. The defense is an interface chosen to stay seated and a mounting that holds it in its compression range across cycling.
How do mounting pressure and gap control affect thermal pad life?
They decide whether the pad stays in its working compression range. Too little compression and the pad does not fully wet both surfaces, so contact is marginal from day one; too much and it is over-stressed and more prone to pump-out and set. The installed compression has to sit in the pad's recommended range across the stack tolerance, so the clamp load and the gap tolerance are design inputs. A pad sized to the nominal gap with no tolerance allowance can barely touch at the loose end of the stack.
Why is my EMI gasket installed but the enclosure still emits?
Because the gasket grounds against a painted or anodized surface, so it never reaches bare metal and the interference current has no path, even though it compresses. Anodize and paint are insulators, so the contact area has to be masked to bare metal or a conductive finish. The gasket also has to compress enough for continuous contact and the seam has to stay continuous. A bench coupon on clean metal grounds fine; the assembled enclosure with a coated flange does not.
Why did my shielding fade over time even though it worked at first?
Usually the seam relaxed or a fastener loosened, so the gasket lost the compression that maintained continuous contact, or corrosion or contamination built up in the contact zone. An EMI gasket has to stay compressed and stay in conductive contact for the life, so a lid that relaxes, a clamp that backs off, or a contact surface that degrades all fade the shield. A retained clamp and a continuous, compressed, bare-metal contact are the defenses.
Can a more conformable thermal pad resist pump-out better?
Conformability helps the pad stay seated as the gap breathes, but it is a balance: a pad must be soft enough to fill the gap and stay in contact, yet stable enough not to be worked out under cycling. A boron-nitride silicone gap pad chosen at the right softness for the gap, held in its compression range, resists pump-out; a graphite interface resists it differently, by being thin, stable, and well clamped. The right answer depends on the gap, the clamp, and the cycling, which we size together.
Does sterilization affect the thermal pad or EMI gasket?
It depends where they sit. An internal thermal pad or EMI gasket inside a sealed enclosure may see only the internal environment, while an exposed part sees the sterilization route directly, so the duty differs by location. Tell us which parts are exposed and which are internal and we grade each to its actual environment. The sterilization validation across the assembly stays with the device builder.
Is shielding effectiveness something H-O can guarantee?
No. Shielding effectiveness is a system property of the assembled, grounded enclosure measured by the device builder in an emissions scan, not a property of the gasket alone. H-O converts the conductive gasket and foil to your drawing and flags the grounding-contact and seam-continuity features the material needs to work, but the shielding effectiveness and emissions qualification are verified on the finished product by the device builder and its test houses.
How do I keep the thermal path working for the life of the device?
Design for contact over the life: choose an interface that resists pump-out, control the gap so the installed compression stays in the pad's range across stack tolerance, and retain the clamp load. A thermal path is only as good as its continuous contact with both surfaces, so the mounting that holds that contact is as important as the pad. Tell us the gap, the tolerance, the clamp, and the thermal cycling and we size the pad and flag the gap control needed.
Can H-O kit the thermal pad and the EMI gasket together for the line?
Yes. We die-cut each part to its own drawing, kiss-cut on a liner for placement, and kit them under one part number so the line receives the thermal interface and the EMI gasket as a reconciled set rather than loose pieces, with material traceability and lot-code TDS records. The conductive and insulating parts are kept clearly separated so they are not interchanged. The mounting and grounding design stay with the device builder.
What do I send H-O to design a thermal interface or an EMI seam that lasts?
For a thermal interface, send the gap, the stack tolerance, the clamp or mounting pressure, the heat load, and the thermal cycling, and we size a pad that holds contact in its compression range. For an EMI seam, send the seam geometry, the mating-surface finish, the closure force, and how the clamp is retained, and we cut the conductive part and flag the bare-metal contact and continuity needed. Add quantities, and we quote a die-cut part. The companion how-to-spec article on this estate is a full checklist.
Have a thermal interface or EMI seam reviewed
Send the gap, stack tolerance, and clamp for a thermal interface, or the seam geometry and contact-surface finish for an EMI seam, and we will size the part and flag the gap control or grounding contact it needs, quoted to your drawing.
Samples typically ship in 3 to 5 business days for common die-cut configurations on materials we keep on hand; standard production runs about two weeks after drawing approval; expedited service is available. MOQ varies by material and part.
Family-owned in Winsted, Connecticut since 1971 · ISO 9001:2015 certified · OEMs worldwide send us a drawing — and get back their part.
Related reading
Request a quote for your medical-electronics part
Send H-O the gap or seam geometry and the duty—engineering reviews every RFQ and converts samples on your geometry.
Last updated · Reviewed by H-O Products engineering (ISO 9001:2015 certified converter, Winsted, CT).
References
- ASTM D149 (dielectric breakdown voltage and dielectric strength of solid insulating materials), cited by designation where a dielectric is involved.
- ISO 10993 series (biological evaluation endpoints), cited by designation where a part contacts a patient-facing surface.
- Manufacturer thermal data and technical data sheets for the boron-nitride and graphite thermal interfaces and the conductive EMI elastomer and foil grades named on the parent application page.
This article summarizes material-selection and design principles for medical-electronics thermal management and EMI protection using the families and published test methods cited on the parent application page. Comparisons are qualitative; per-grade values are maker technical-data-sheet figures and must be confirmed on the manufacturer's current TDS at your installed condition. Standards (ISO 10993, USP Class VI, ISO 11607, IEC, ASTM, UL) are cited by designation for context, not as a claim of qualification or registration.
H-O Products is a precision die-cut converter, not a raw-material producer and not the device maker; biocompatibility testing, sterilization validation, device clearance, and final-assembly certification belong to the device owner and its test houses.