Contact Us
  • Home
  • BLOG
  • Why Do Press-Fit Pins Fail at the Hole, Not the Pin?

Why Do Press-Fit Pins Fail at the Hole, Not the Pin?

A press-fit joint fails at the hole, not at the pin. The pin is a compliant spring and it does its job; what decides the connection is the barrel it was pressed into, because IEC 60352-5 defines the joint by an interference band that the hole has to hit. That band is narrow: a 1.15 mm drill, a finished hole between 1.02 and 1.10 mm, hole copper of 25 um minimum, and plating that does not crack when the pin goes in. Miss it and the failure is silent, because a joint that is not gas tight still passes a continuity check at goods in and drifts a year later.

Two force figures bracket the whole problem. Below roughly 40 N of press-in force per pin the contact is not gas tight, and above roughly 115 N the pin and the hole plating are damaged on the way in. Both come from the same hole, and the same connector pressed into two different boards will sit at opposite ends of that window.

What a cross section actually measures

IEC 60352-5 does not accept the joint on insertion force alone. The check is destructive: the hole is sectioned and the deformation of the drilled hole contour has to stay under 70 um, the remaining plating at the thinnest point has to stay above 8 um, and there must be no crack anywhere in the barrel or in the conductor it connects. That is the real acceptance criterion, and it is measured on a first article rather than on a sample picked off the press bench. A board can pass pull force and still fail this section, because pull force says nothing about a crack that has not opened yet.

Why the band is so narrow

The eye of the needle is a beam. It holds the joint through the normal force it puts on the hole wall, and that force comes from how much the beam is compressed, which is the interference between pin and h 25 um is compliant and gives the beam room to deform the barrel; copper at 50 um resists, so the same hole diameter produces a different force and a different deformation. This is why the hole cannot be specified by finished diameter alone. A shop that drills nominal and plates thick, and a shop that drills small and plates thin, can both report the same finished hole and deliver two different joints. The drawing has to carry the drill, the hole copper and the finished diameter, with the drilling position and the panel position of the connector called out.

The position matters as much as the size. Hole to hole pattern tolerance in the connector field is usually held around plus or minus 0.1 mm, and the pins are rigid in that plane. A field that is off by 0.1 mm loads one row of pins against the other, so half the connector is under pressed and half is over pressed, and the cross section at one end looks nothing like the cross section at the other end.

Why the finish decides the force

Immersion tin inside the hole, usually 0.8 to 2 um, is what most press-fit specifications ask for. It is soft, it shears under the pin, and it gives a force that repeats pin to pin. ENIG is a hard surface and pushes the press force up for the same hole. Lead-free HASL puts a tin layer that varies from about 1 to 40 um across the same panel, and that variation shows up directly as force variation across the connector. On the pin side, tin over a nickel underlayer is the combination that keeps whiskers from growing out of the coating and landing on a neighbouring contact.

Reflow before pressing is not the problem. Connector makers test modules pressed into boards that have already been through reflow and find retention unchanged. The second press is the problem, and so is a board that has been pressed once, pulled apart and used again in service, because the barrel has already been deformed and the second insertion starts from a different interference.

Where builds go wrong

Specifying the finished hole only. The drawing says 1.05 mm and nothing else. Two suppliers then deliver two joints, one at 30 N and one at 120 N, and both claim conformance.

Qualifying on a coupon instead of the panel. A separate test coupon presses beautifully. The connector sits in the middle of a 400 by 500 mm panel, where drill registration and plating thickness are at their worst, and that is the location that decides the batch.

Re pressing a serviced board. Rework studies on high speed press-fit connectors show three insertions survive when the correct tooling and process are used, and also show that hole plating and pin geometry, not the number of insertions, decide the loss of strength. A fourth press into a hole measured in minimum condition is where the plating gives up.

How we hold the hole

SUPERB PCB holds press-fit hole tolerance at plus or minus 0.05 mm, tighter than the standard plated through hole tolerance on the same build. Holes for press-fit fields are drilled, not milled, immersion tin is specified inside the hole, and the connector field position is checked against the panel datum rather than against the drill origin. First article includes a cross section of the pressed field with the contour deformation and the remaining plating measured and recorded, and press force is logged per pin on a force displacement press so the window is a number on a chart rather than an operator's feel.

The same line runs backplanes up to 68 layers with back drilling and controlled impedance. Press-fit assembly on high speed switch backplanes follows the plated hole work described in plated hole copper and the limits in rework and repair limits.

What to send with the question

The page of the connector data sheet that gives the recommended hole, the stack-up with the copper weights, the finish you want inside the hole, the position of the connector on the panel, the press force window the connector supplier states, and the quantity per build. Six items, and the answer is a hole specification that both the board shop and the press bench can hold, not a trial batch that tells you in three weeks which way it drifted.

Need a quote or a review of your build against these limits? pcba@superb-tech.com · WhatsApp +86 133 9608 1443