Kelvin Contacts

Guest

I noticed there are two metal fingers contacting every pin of the DUT while being tested in our tester. Some say this is for extra current capacity while some say this is to ensure good contact. Some say these are called Kelvin contacts and its for accuracy of testing. Can you shed light on this? Is it bad if one of the fingers broke off during testing?

Mang Kelvin

Posted: Wed Aug 06, 2026 5:39 pm Post subject:

Mang Kelvin,

The explanation that this pair of metal fingers is for Kelvin contact is the correct one. When testing for a parameter you need to 'excite' the DUT and measure its response. At the minimum, this would involve two pins. If you're using Kelvin contacts, this would require 4 contacts (2 per pin).

One of the fingers is for forcing the excitation (say, current) through the test pin and the other is for measuring the response (voltage) directly from that pin. Of course, the return pin would likewise have corresponding 'force' and 'sense' pins.

The reason why you want a separate set of fingers for measurement is to avoid the IR drops through the path used for excitation, which can result in large error measurements if the forced current is large. The fingers used for measurement can measure the voltage with very minimal current, so the IR drops through this path is negligible.

Yes, it is bad if one of the fingers (whether it's the 'force' or 'sense' pin) breaks off completely, because the force/sense system will fail.

You have just inspired us to write an article on various contacting methods, so watch out for this. Anyway, thanks and keep on posting!

EE

Joined: 20 Jun 2026 Posts: 20

EE, there are instances when you don't need kelvin connection. So, you don't need 4 contacts always, right?

Test Engg Engg 99

Posted: Fri Aug 15, 2026 1:38 pm Post subject:

Yes, you don’t need Kelvin connection at all times to test a part. It is only necessary when the amount of current you need to force into a pin is large enough to cause IR drops that will result in measurement errors. An examples of this is when you’re measuring very low resistances, i.e., less than 1 ohm.

Thus, the decision whether to use Kelvin connection or not depends on the product you’re testing and your hardware characteristics. Some companies even try to avoid Kelvin connections when possible because they are more complicated and costlier than a 2-wire measurement connection.

EE

Guest

Hi Mang Kelvin;

To add up on what EE and testeng99 answered, mostly we are using kelvin contacts to accuarately measure low current dc test and on soft dock systems to compensate the errors (IR drop) made by wires and board traces. Yes, it is using 2 contacts on a single pin, one is forcing while the other is sensing the forced voltage. Example we are expecting a 0.25 A (example lang) current output. To do this we need to force a certain voltage say 5 volt but due to drops on the wires or added resistance, forced voltage at the pins is not actually 5 volt but less. If less voltage is used then your current reading will not be accurate. The sense pin of kelvin contact will tell the tester that hey force more voltage at the source so that when it reached the DUT it will exactly be 5V.

Guest

Mang Kelvin,

Tutal naman tagalog pangalan mo.... TAGALUGIN ko na rin sagot ko.

OK, una... kaya dalawa ung contact mo sa isang pin, dahil ung isa ay FORCE pin at SENSE naman ung huli. Sabi nila dapat apat daw ang KELVIN.... HEHEHEHE! tama sila, kaso mali ang sagot na dapat apata sila. ung dalawa pang contact ay para FORCE and SENSE ng Ground.

Kaya na-design ito ay dahil nga sa sinabi ni KIM.... to produce more accurate measurement/forcing. To cut the answer short....

4 element ang KELVIN:

1. HIGH FORCE
2. HIGH SENCE
3. LOW FORCE
4. LOW SENCE

Sana makatulong....

Flipflop

Guest

Hi to all!

Would like to iquire if anyone have experienced this so called Kelvin Contact Failure? This happens when the resistance of the
contacts is observed to be higher than the set kelvin limits.

Is this related to fretting corrosion or corrosion observed on the contact blades? Does anyone have suggestions on how to prevent this type of
problem.

thanks Twisted Evil

Guest

Hello Kelvin Jr.

Kelvin is a way to detect if the contacts to your DUT is ok or not. Its only whether it is open or not. Sometimes corrosion or anything that increases the resistance of you contact pins to the extent that the reading will become somewhat open. Open pins has large resitance readings(or infinite), why, because they are open. If corrosion or whatever it is prevents your force and sense pins from connecting then you will get the open reading.

Why corrosion? Contactor pins don't corrode that much. maybe the DUT. Are you an FA eng or REL? I think you can ask your boss, or pati ba naman FA ngayon bumalabs ang boss? Akala kasi namin sa Test lang incompetent ang boss kaya marami nag reresign....

Guest

Hi Green Lantern...

I am not an FA engineer. I don't know if you have observed this black substance on where the unit made contact with the contact blades.
This is categorize as fretting corrosion. Have you experienced this type
of phenomenon?

Thanks

Guest

Di kaya dahil sa tin/lead plating ng DUT yung black substance?

Posted: Tue Mar 22, 2026 2:52 pm Post subject:

hi kelvin jr.

the black substance that you have observed may have been carbon deposits which are a by-product of electrical arcing especially if how-switching was involved.

Posted: Tue Mar 22, 2026 2:55 pm Post subject:

erratum:

hot-switching, hindi how-switching

eto yung pag-connect mo ng relay sa pins ng device na kung saan may power na sa other end ng relay. ideally dapat once kino-connect yung
isang relay, yung source na kinakabit niya sa pin e dapat naka-zero-out pa. pag hindi, eto yung hot switching.