Application-Specific Notes¶
Qorvo devices¶
Qorvo device measurements require additional physical setup before you begin the calibration procedure above:
- Apply thermal paste and stack the device: a. Evenly spread thermal paste on the TC-100 thermal chuck and place the Qorvo mounting plate on top. b. Apply thermal paste to the thermocouple probe and place it into the hole on the Qorvo mounting plate. c. Evenly spread thermal paste on the Qorvo mounting plate and place the Qorvo device on top.
- Secure the device: a. Screw the Qorvo device and mounting plate securely onto the TC-100 thermal chuck. b. If there are no screw holes, mount the plate and device down with clamps instead.
- Connect power: Connect the Qorvo device to its power source.
[diagram/image in source — not reproduced here: photos of the thermal paste, mounting plate, and clamp setup]
For Qorvo devices, the documented example settings for the general procedure above are:
- Hardware calibration: High Temp 75 °C, Base Temp 120 °C, thermocouple type K
- Timing/Illumination: Stage Heating Time 40, Image Averages 50
- Movie Mode: chuck Set Point 80 °C, device drive voltage 10 V, DUT pulse 15 s, Movie Mode Length "Very Long (30 sec)"
Treat these as a documented starting point rather than fixed requirements — adjust pulse voltage, pulse width, and chuck temperatures for your specific device's power handling and expected self-heating.
TEA TTV samples¶
The TEA TTV guide documents several reference-sensor/coating options for TCAL calibration, with different accuracy and setup-time trade-offs. In general:
| Method | Absolute accuracy | Delta accuracy | Sample prep required | Typical time |
|---|---|---|---|---|
| Stage Cal (stage sensor) | — | — | No | ~5 min |
| Stage Cal (thermocouple sensor) | — | — | No | ~5 min |
| Stage Cal (external customer sensor) | — | Yes | Requires a customer sensor on or near the device | ~5 min |
| TCAL (using R1 emissivity from Stage Cal) | — | — | No | ~6 min |
| TCAL (R1 with black paint coating/dot) | — | Yes | Enter emissivity, or use Stage Cal for black-paint emissivity | 5–10 min |
| TCAL (R1 with black IRISS sticker) | — | Yes | Enter emissivity, or use Stage Cal for sticker emissivity | 5–10 min |
| TCAL (R1 with White Out) | — | Yes | 2-step, using Stage Cal for reference emissivity | 5–10 min |
| TCAL (R1 with Sharpie dot) | — | Yes | 2-step, using Stage Cal for reference emissivity | 5–10 min |
Notes on this table: - Stage-sensor and thermocouple-sensor stage calibrations can be affected by thermal expansion, sample movement/alignment, and difficulty placing the thermocouple precisely. - An external customer sensor gives absolute-accuracy results but requires the customer's device to already have a sensor on or near it. - The TCAL rows using a coating or sticker give delta-accuracy results and either take a directly entered emissivity value or a two-step process that derives emissivity from a Stage Cal measurement.
Specific guidance for reference regions on TEA TTV samples:
- Black paint or IRISS sticker: spray-paint a small window (cut into the PCB) black, or cut a piece of an IRISS emissivity sticker (approximately 0.95 emissivity) and place it onto the device with tweezers as your known-emissivity reference region.
- External thermocouple probe: apply a small amount of thermal paste to the probe tip, then place the probe on the corner of the TEA TTV sample.
- To obtain a calibration coefficient from SanjANALYZER+ for use as a "Single Emissivity"/"Calibrated from File" input: draw an ROI over a known, accurate high-emissivity reference region on the calibration image, read the coefficient from the "Mean" value, and enter that value as your calibration coefficient.
Additional resource referenced in the TEA TTV guide: FLIR, "Use low-cost materials to increase target emissivity" — https://www.flir.com/discover/rd-science/use-low-cost-materials-to-increase-target-emissivity/
Electroplated gold surfaces¶
Electroplated gold is called out separately because its surface finish makes calibration noticeably more sensitive to defocusing and misalignment than a smooth reference surface. A misaligned calibration on electroplated gold produces visibly poor results compared to a properly aligned one.
[diagram/image in source — not reproduced here: side-by-side CCD images comparing a misaligned vs. an aligned calibration on electroplated gold at 50x]
Guidance:
- Pay extra attention to focus and 3D alignment quality (Focus ROI and Capture Reference steps) when calibrating on electroplated gold, since small alignment errors have an outsized effect on this surface.
- Current alignment algorithms are designed to reduce misalignment-related edge effects and improve pixel-by-pixel accuracy on rough surfaces like electroplated gold; make sure your software is up to date before calibrating challenging surfaces of this kind.