contact resistence vs use vs time 1

[Tmoose]

A fellow here has witnessed a drop in resistance of a crimp type joint after the first application of current, and figures it is a contact resistance effect fromn welding.

Is there a name for this phenomenon?

-----------
Most all sources publish a resistance increase for bolted or crimped connection joint over time, or with temperature or service cycles. This would suggest the electrical "sweetening" is at best a one time event.

 

[VEBill]

Thermal expansion?

 

[Tmoose]

"Thermal expansion?"

The eye witness describes the effect as nearly instantaneous and visible on an oscilloscope, and permanent, at least short term in the lab. He was pretty certain it was an improvement in contact resistance. I'll try to find out more.

 

[VEBill]

Since measuring resistance also involves applying current (normally a small current), perhaps the measurement (Vdrop vs I, starting from zero, to Imax, and back to zero again) could be combined with the stimulus and plotted.

 

[IRstuff]

Maybe, simply a burn-in.

A crimp is a mechanical compression of two metals together. Whatever was on the surfaces is now between two metals. Sufficient current could potential burn off volatiles and break down other contaminants and allow better metal to metal contact.

Addtionally, since most surfaces only contact at the tips of asperities, the current density at those points might be high enough to cause deformation and broadening of the contact points, thereby lowering the net resistance.

TTFN

FAQ731-376

 

[Comcokid]

What are the metals and platings involved on both sides of this "crimp type joint"? Are we talking stranded copper in a tin/brass crimp terminal, or something more non-typical like a nichrome wire crimped into a stainless eyelet? (Note-my examples here are just random ones, and not examples where I know of unusual effects.)

How much current are we talking about that is being used?

 

[bogeyman]

I do not doubt the observed changes, I have some reservations as to the standard to which the assembly was built.
Ordinarily the changes to one single connection as described would be so small that they would go quite unnoticed even using very sensitive measuring equipment.
Just how dirty does a crimp tag (and there are some very tarnished ones waiting to be used on many shelves)need to be to have such a noticable effect? I suppose the same could be true of some cables having grotty surfaces on their wire cores.
In a correctly crimped connection the interface of each face effectivly welds to the mating face giving a gas tight and therefore somewhat stable and corrosion resistant joint.
Dirty contacting faces will prevent the joint from being properly formed and should have no place in anything which we would want to be reliable.

 

[HCBFlash]

I'm with IRstuff, especially if the crimp terminal is plated, or dissimilar to the conductor it is applied to, but don't recall ever hearing a particular unique term used for this particular phenomenon.

 

[electricpete]

It may be possible due to thermal expansion or the other stuff mentioned above, but I am certainly skeptical of any sudden instantaneous change in resistance.

I am a little suspicious of the meausurement method and would like to know exactly what was measured (voltage, current, resistance), using what meter and what test setup.

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[electricpete]

Thinking again... sometimes circuit breaker contacts read lower with ductor reading (high current) than with multimeter (low current). It may have something to do with oxide layers as was mentioned...

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[electricpete]

A low current/low voltage dc measurement (multimeter)can also be susceptible to a variety of errors including voltages from disimilar metals, especially if temperature varies accross the circuit.

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[electricpete]

And they had better have been using a four lead measurement method.

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[flexoprinting]

Electricpete

Four leads?! Do you mean measure voltage, current and use
ohms law? Or is there a procedure I need to learn about?
Please explain.

Thanks

 

[VEBill]

4-wire measurement of R.

Basically the current is applied using one set of wires and the resultant voltage is measured with another set of wires. This makes for a much more accurate (and thus more sensitive as well) measurement of R.

The instrument drives an accurate current down the stimulus leads where voltage drops don't matter anyway. And there is virtually no current in the voltage leads (Hi-Z input) so there isn't any voltage drop to speak of.

In practice, instead of using probes, you use two clips. Each side of the clip is a separate connection to the instrument.

End result is a lovely stable R measurement with lots of digits to the right of the decimal point.

 

[flexoprinting]

VE1BLL

Thank you!

 

[IRstuff]

The 4-wire technique is sometimes known as Kelvin sensing:

http://en.wikipedia.org/wiki/Four-terminal_sensing.

The approach reduces the impact of the probe resistance from the measurement of the voltage drop.



TTFN

FAQ731-376

 

[MacGyverS2000]

Temp sensors (such as the venerable P100 platinum doodad) often use the 4-wire sensing method. Talk about accuracy... 0.01C accuracy!

Dan - Owner

http://www.Hi-TecDesigns.com