Cracked resistor?

I assume you've got your test prods directly on the pads so you know it's not just a dry joint?

X-posted to s.e.r (more appropriate group)

I had a Fluke ammeter once that had an open resistor due to the endcap de-bonding from the resistor material. It was invisible and only detected by poking the cap.

Bob

Don't know about SMT resistors but I've seen leaded resistors blown with almost no evidence. The cause was a *massive* overload (lightning damage in an agricultural application). And the resistors were metal film so they didn't arc.

--sp

Yes. But making that mistake isn't beyond me, by any means. It's a very well-assembled board (I didn't do it), but well designed, too (I did it

-- and I'm modest, just ask me).

I suppose it could be something like that. This resistor is pretty well isolated from the big ugly world -- the customer was doing prototyping, so it's more likely they set the board down on top of a metal tool or something.

I've always assumed that the black stuff you see on the top of a SMT resistor is the resistive material, and that it'd be smoked away if it were burnt. But, maybe not.

The board is well within normal operating parameters with the resistor replaced.

In the interests of Science I sacrified a 15.0 ohm 1% 0805 resistor to the cause (25V/1A applied).

The only visible sign of damage is a small hole near the center of the part.

Huh. Just, huh.

Thanks.

Perhaps a factual survey as to how and why low value carbon composite resistors became to be cracked may help. The extreme many times was not only cracked but also burnt. But there were times when there was no visual indication of over-heating. My wild guess is stress from over-current rating pulses; sort of like electro-migration.

Plenty of water and the whiskey bottle has been properly drained.

I think that there is a protective coating of some sort. May even be clear.

I've only one documented example:

YMMV.

The low value in your case does perhaps hint at, shall we say, dynamic failure. But who knows. Got an electron microscope? ;-D

Tim

-- Seven Transistor Labs, LLC Electrical Engineering Consultation and Contract Design Website:

Hand soldered SMT resistors can sometimes develop invisible open circuits. When this happened to me I discovered that the person soldering them would reflow one end and then push the other end down with the soldering iron tip as it was also reflowed. The second end was always slightly above the pcb surface because the unmelted solder balls acted as a spacer. The crack in the metallisation was usually invisible even with a very good quality binocular microscope. John

This is a professionally assembled board. I'm not ruling out assembly error but if there is one it's more subtle than that.

(For the past year or two I've been reflowing all my prototype boards, either component-by-component with a hot-air reflow gun, or all at once with a skillet. My defect rate has gone down, mostly due to those evil packages that put pads on the bottom only.)

The dissipation power is 200mW max.

So I lower than 0.3A.

If not, it burns !

No, most of them have a layer of shiny black (or other color) paint over the actual element, then the part value is screen printed on.

Jon

Hello, and the only similar problem I've encountered is with those sandy-colored dropping resistors (rectangular parallelepiped shaped) that have been used in radios and TVs. They can swell and/or crack over a period of time. Repeated heating and cooling no doubt stresses the resistor material. Sincerely,

Some of those are 'Fuse Resistors', and are designed to fail open if the equipment starts drawing too much current, or if it is running too hot. I replaced hundreds, if not thousands of those in consumer electronics. Some OEMs switched to the Belfuse 'Chemical Fuses', which had a chemically coated element that would flash and vaporize the element at a set temperature.

They were quite common, in spite of Phil Allison's claim to the contrary.

Not much safety margin for an 0603.