The optical scheme of the pyrometer includes two optical channels:
measurement channel;
sighting channel.

Copper Freezing Point Calibration Results

Although formally TSP pyrometer is being delivered uncalibrated, it was found convenient to use the copper point BB for the pyrometer characterization, specifically for assessment of its reproducibility, short term stability and resolution. Calibration of TSP pyrometer has given the following results:
(a) 650 nm head: signal 0.14821 V, 8th range of amplifier,
(b) 900 nm head: signal 5.201V, 8th range of amplifier.
Those data were used later on for comparison with the transfer standard lamp. A sample graph of the fixed point calibration is shown in figure below.

To confirm that the contract requirements to pyrometer stability and resolution are met, we have performed an experimental study of the matter, which has shown that all specs are outperformed.
The most reliable and impressive data have been received during TSR calibration by fixed point BB, which is already mentioned in the Section 2.1 above.
During repeated measurements of the fixed point temperature the day-to-day reproducibility was found to be better than 20 mK with 650 nm detector head and 30 mK with 900 nm detector head.

In our case a good quality iris diaphragm was positioned immediately after BB opening. Results, obtained by both independent methods, were found to be consistent with maximum SSE correction reaching 0.27% for 15 mm diameter circular source. Certainly, high accuracy measurements, for example, of lamps, would require additional measurement with rectangular mask, but the main purpose of our investigation was to check the general quality of the optics.
The obtained data is presented in the figure below.

As for resolution, it could be easily illustrated by the following two graphs, where signal drift for 15 minutes stayed within 5 mK corridor (with 650 nm head - 10 minutes).

Size of Source Effect Correction

Size of Source Effect (SSE) could have significant adverse influence on the accuracy of measurements involving comparison of sources with considerably different geometry. That is why at both design and characterization stages it received a special attention.
Technical steps to minimize the SSE are beyond the framework of present document, so only SSE measurement methods and results are be covered.
SSE was measured using 2 complementary methods - the first giving integral value with a good accuracy, and the second providing a detailed data with lower resolution.
The fist method employs a small mask (we used mask with diameter of 1.1 mm), obscuring the central part of the source, letting go all other radiation, which ideally should make a zero contribution to the signal. This method gave us a 0.27 % SSE while viewing a blackbody (BB) with exit aperture exceeding 16 mm.
Another method is based on the direct measurements of the signal dependence on the size of source.