Problem I encountered during calibration of specialized test equipment and Ques

[mellowtron]

So I work at a medical device company, we make a device that measures electrolytes in the blood. Blood drop goes in a test card, test card goes in the test card reader, test card reader has circuitry to read voltage and current from the blood reacting with reactants on the test card.

The potentiometric/amperometric circuitry of the test card reader must be calibrated.

A test card simulator exists for this. It is a printed circuit board connected to a specialized power supply. On the test card simulator, there are various precision resistors in series with individual reed relays. The specialized power supply can interface with a computer, and the computer can tell it to close internal relays. When an internal relay is closed, 5V goes through the individual reed relay coil and closes the relay. When this happens, the test card simulator will present the resistance to the test card reader for it to measure.

The test card simulator requires calibration, so there is a process to close the test card simulator power supply relays, and consequently a relay on the test card simulator, and then measure the total resistance. If it is out of specification (0.11%) then a repair must be made.

On this recent occasion in question, I measured the total resistance on the 200k resistor circuit to be about 200.2200 Ohms, above spec.

I measured the resistor in-circuit as in-spec (i don't remember but very close to 200k)

I measured the ON resistance of the relay contacts as less than one ohm. (0.4 or so)

I measured the resistance between the relay contacts, resistor, and contact pads as less than an ohm.

I cannot explain where the extra resistance comes from.

Then, as part of troubleshooting, I replaced the cable that connects the test card simulator power supply to the test card simulator (a specially-made 12-conductor cable assembly), and the resistance problem disappeared. But since the cable is not part of the circuit, this makes me confused. The cable was merely supplying 5v to the reed relay so that the test card reader could read its 200k resistor.

Anyone have any insight? Please note this is a DC circuit and I always wait for readings to settle before taking a measurement.

[ajb]

Have you inspected the power supply cable that was replaced?  Perhaps the cable was limiting power to the relay coil due to a bad connection, causing the relay to not pull in with as much force as normal.  Lower contact pressure would certainly cause increased contact resistance.  If you read the relay contact resistance under different conditions (for example, fewer total relays on?), that could explain why you didn't find an issue then.  Or it's possible that the fault is intermittent or temperature dependent or something. 

[moffy]

If there is any way a small voltage could bleed from the cable into the circuit it would look like a change in resistance.