Ask Dr. Bob

Contact Resistance: Key to Success
By Max Peel, Contech Reseach

The basic sole purpose of a connector is to allow a signal or current to flow from point A to point B. To achieve this function in a continual manner, the electrical resistance has to be as low as possible and remain so for the life of a system.

In measuring the electrical resistance of a separable mated contact, four elements are being measured;

1. Termination resistance, which is generally less than one million, whether it is crimp, solder, compliant, etc., unless the termination process is faulty.

2. Bulk resistance of the contact material. This will not change except due to normal temperature fluctuations, and should not be the result of any degradation process.

3. The resistance at the separable interface, which may change due to “something” happening at the interface, such as loss of normal force, corrosion issues, etc.

4. Film resistance, which normally is not a problem, since it is wiped or displaced during the mating operation.

Two additional elements will impact the resistance: normal force stability and wear. For resistance stability to exist, the normal force has to be stable. Unstable normal force will result in unstable resistance. The second additional element is wear. Wear is a complex factor and is contingent on and interacts with the magnitude of normal force and the contact geometry. If the plating system is disrupted, significant changes at the separable interface may occur. More on these issues will be forthcoming in future discussions.

As indicated in the previous article, there are two basic resistance evaluations performed—low level circuit resistance (LLCR) and voltage drop, or contact resistance at rated current. Both of these variables are performed with the same basic technique, but expressed differently.

For signal contacts, it is normally expressed in milliohms. For power contacts, it is normally expressed in millivolts, but could also be expressed in milliohms (contact resistance at rated current). Both should use a four wire system. Two probes are for current input and output. The other two probes are the voltage leads. Thus the current leads may be placed at any convenient location. The voltage probes should be placed at a specified distance from the terminations involved. This will minimize bulk resistance of conductors, traces, or other features which do not impact the interface. In all instances, the voltage probes must be forward of the current probes.

LLCR, which is used to measure signal contacts, is normally performed with a 100ma maximum test current and a 20 mV open circuit voltage. This combination will evaluate the system under conditions where the applied voltages and currents do not alter the physical contact interface. The key is the O.C. voltage. This voltage will not puncture or disrupt any damaging films or oxides that may be forming within the interface area. The current is not a major factor.

Some feel that higher currents will “burn” through thin films, but this is exactly what one does not want to happen. Burning through a film or oxide will or could generate heat, which may result in other problems. Current will not impact LLCR except for possible resolution issues.

Voltage drop or contact resistance at rated current used for power contacts is normally performed at a constant current generally at a rated current. The O.C. voltage is rarely specified and is allowed to fluctuate up to the power supply limitation. Contingent on the magnitude of the O.C. voltage, it can possibly electrically displace or rupture through thin films and oxides.

With increase usage of signal contacts being used for power purposes (via parallel circuit setups), both contact and LLCR is required in many specifications. This sets up some basic guidelines which should be followed to avoid masking a potential problem.

1. Whenever possible, LLCR and contact resistance should be performed on separate contacts, never on the same contacts.

2. If this is not possible, then contact resistance should be performed before LLCR at the initial measurement interval and after the final LLCR has been performed at the final measurement interval. Contact resistance should not be performed at any intermediate interval in order to prevent high voltages being applied to assure no disruption of any films or oxides which may have been initiated by the environments within a test sequence.

For power applications and to assure proper measurements are performed, it is also important that the conductor leads and test board traces be designed to handle the current levels being applied.

The above are the basics for the two key measurements. Another important criteria is the number of data points required. Most military, EIA, and IEC specifications only require between five and 10 measurements per connector, regardless of size. For today’s connectors, and to some degree the increase in the use of gold flash, this is gross undertesting. The classical argument is that the cost of testing is too high. With the availability of advanced equipment and computer technology, this argument is simply wrong. With proper test board layout and use of common buss concepts, up to 400 measurements per hour using manual techniques, and up to 3600 measurements per hour using automatic test concepts are possible. Many user proprietary and special interest groups are now beginning to significantly increase the number of data points required. This is being driven by new and varying applications and longevity issues.

A good general rule to use is to measure a minimum of 25 positions, or 25 percent of the contact population, whichever is more. In the event of connectors which are equal or less than 25 positions, then all positions should be measured. Measurements should be divided up between rows in a connector, including end and middle positions.

Many specifications in the past have specified actual resistance values that must be met. There will be an initial value and another value at subsequent intervals (such as 10 milliohms max. initially and up to 50 milliohms max. final, or somewhere in between, contingent on the qualifying agency).

This is gradually changing with many user specifications allowing initially a maximum value and specifying a maximum change. The basic question is how much change should be allowed. This point becomes a matter of experience and thought to adopt a logical value. The consensus allowed change is a resistance change less than 10 or less milliohms, with some specs also requiring a maximum average change. This is the trend for signal contacts.

Signal contacts are more forgiving then power contacts to resistance changes. For power contacts, resistance increases will cause increases in temperature rises, which could result in failure or thermal runaway. Thus only smaller magnitude changes should be considered for these types of contacts.

To interpret data relative to resistance changes, the following is offered:

a.       To evaluate contact resistance characteristics of the contact systems under conditions where applied voltages and currents do not alter the physical contact interface and will detect oxides and films which degrade electrical stability.

b.       These variables are monitored after each preconditioning and/or test exposure in order to determine said stability of the connector material systems, as they progress through the applicable test sequences.

c.        All contacts so monitored should be identified and excluded from any supplementary test that may impose a high voltage, thereby disturbing the conditions at the contact interface.

d.       The electrical stability of the system is determined by comparing the resistance value after a given test exposure to its initial value (prior to any exposure). The difference is the change in resistance occurring ­in which magnitude establishes the stability of the interface being evaluated.

e.       In order to categorize the change occurring in the resistance as the test sequences are performed, the following guidelines may be used:

·          <5.0 milliohm change                :         stable
·           5.1 to 10.0                                    :         stable with minor changes
·           10.1 to 15.0                                  :         stable with significant change
·          15.1 to 50.0                                   :         marginal
·           >50.1                                             :         unstable

f.     The above categories, in combination with data frequency, trends, and time of occurrence, allow a final determination to be established.

g.    A stable resistance measurement is defined as not fluctuating more than ± 1.0 mW. For item d) values, any drifting >2.0 mW should be noted.

From an experienced point of view, one should observe the measuring equipment if the changes start to exceed a +15.0 milliohm change. If significant “drifting” of the measurement occurs, then stability is in question. Changes for power contacts are generally 50 to 75 percent less than signal contacts relative to resistance requirements.

As this series of articles continues, additional comments will be forthcoming to address issues in resistance change impact and interpretation contingent on the various unique effect of different environmental stresses applied to contacts.

So in the meantime, relax, enjoy a good cigar and a glass of vintage brandy, and watch your well-tested connectors march to perfection.

Send your connector testing questions to AskDrBob@ConnectorSupplier.com.

Max Peel is a Senior Fellow at Contech Research, an independent test and research lab located in Attleboro, Massachusetts, U.S.A. For more information, visit www.contechresearch.com.