A manufacturer of medical devices in eastern Massachusetts brought me in, to identify causes of inconsistent and unstable calibration in a patented operating room device that was in pilot-production at the time.

    The calibration required knowing the resistance, at all parts of the operating cycle, of a very small ceramic resistance-heating element at the end of a long thin probe. The device was designed such that the only access for resistance measurement was at the connecting wires at the base of the probe, several inches away from the heater.
    After some testing and analysis I concluded that the variation in resistance as the heating element and the wires heated and cooled during operation of the device was not predictable enough to keep the element temperature stability in spec.
  The proper way to measure the resistance in this case would have been to add two wires, or a flex-circuit, inside the probe to achieve 4-wire (Kelvin) connection at the heating element, allowing the instantaneous resistance to be calculated. Together with measured current this would allow the micro-processor to calculate and adjust power at any time during the cycle.
    Unfortunately, expensive production mold tooling had already been delivered and paid for. The client eventually decided, for this and other reasons, to cancel the project.

    Not properly considering how to calibrate a critical part of the device contributed to this otherwise promising project being canceled.




Back to Troubleshooting page.