Downloaded from https://www.everyspec.com MIL-HDBK-816 NOTICE 1 TABLE VIII. Irradiation conditions. a. Name of the radiation test facility. b. Name and organization of radiation test engineer. c. List of dosimeter types and readout equipment. d. Date of irradiations. e. Irradiation run number. f. Number of samples in each irradiation. g. Placement of samples with respect to the radiation source. h. Radiation facility operational mode. i. For each irradiation, flux or dose rate, fluence or ionizing radiation dose, start and stop times, and ambient or device case temperature. j. Device bias conditions and whether in-flux measurements will be made. k.
Start and stop times for post irradiation measurements, including measurements made during a series of radiation exposures (particularly important for ionizing radiation dose because of the time dependent effects which can occur). 5.1.8.5 Estimates of device failure levels. To reduce costs and to optimize the usefulness of the data obtained, it is important to estimate the expected failure levels for the selected device type and radiation environment before full scale characterization measurements are begun. Device failure will usually be defined as parametric failure but may sometimes be defined as functional failure.
In the case of parametric failure, it is well known that different failure limits can be defined for different device applications. For the present purpose, engineering judgment should be used to define a failure limit for each parameter being measured. The important point is that the lowest stress level which renders the device unacceptable should be considered as the failure level. Estimates of failure levels will have to be based either on previous experience and data or on measurements on a few parts from the available samples. The steps by which the failure points are estimated should be documented in the characterization report.
Because the definition of parametric failure depends on the intended device application, the estimated failure points will only be approximately correct. They will nevertheless be useful for adjusting the irradiation increments and the total fluences to match the expected device performance. The irradiation increments and the total fluences selected on the basis of the estimated failure points should be listed in the test plan. After the characterization measurements are completed, device failure points can be more accurately defined. 5.1.8.6 Gradual degradation.
For devices which degrade gradually as they approach failure, it will generally be useful to measure post-irradiation performance characteristics as a function of accumulated radiation fluence or ionizing radiation dose because such results can be used to interpolate or, in some cases, extrapolate the part’s performance to radiation levels other than those used in the tests. For such devices, the number of irradiations and the radiation increments should be made commensurate with the expected failure levels. It is worth noting that a gradual approach to failure may not always be monotonic; sometimes a parameter may first increase with radiation fluence and then subsequently decrease.
This behavior means that, in the military standard system and for the radiation levels shown in table I (see 4.2), a part type that has been qualified at a higher level may not meet the same RHEPL at a lower level. Qualification at a higher level should therefore not mean, automatically, that the part is also qualified at a lower level unless care is taken to select the RHEPL so that it does not show less degradation at the higher level. 5.1.8.7 Abrupt failure. Some devices give little or no indication that a radiation failure point is being approached, until they fail abruptly; usually abrupt failures occur as loss of functionality.
The claim has sometimes been made that there is usually some gradually deteriorating parameter which can be used as a predictor of abrupt failure (e.g., a flip-flop circuit failing functionally when the fanout capability of the circuit, as indicated by the sink current, degrades past a certain point). As a practical matter, however, a device must be considered to fail abruptly when there is no feasible way to pin-point its failure level by means of interpolation. An estimated failure level is also important for such devices and should be used for selecting the number and increment value for the irradiations such that the abrupt failure point will be narrowly bracketed.
Whenever possible, the most desirable measurement is to monitor the device while it is being irradiated to determine the exact stress level where failure occurred. Again, the estimated failure point will have to be based on previous data or on measurements on a limited sample size before full scale characterization measurements are made. SUPERSEDES PAGE 21 OF MIL-HDBK-816 21