Below are the questions asked during the live event, along with their respective answers.
Q: How do you handle requirements that are not applicable in your planning?
A: The DIDs normally want each requirement addressed. If a particular requirement is not applicable the procedure should identify the requirement, state not applicable, and provide the rationale for classifying it as not applicable. Failing to include this could be considered as an omission in addressing the requirements.
Q: How do you get information about design changes during production to update the control plan?
A: This process can be very difficult to obtain information unless compliance / regulatory role is part of engineering change orders. The compliance authority should review the change request and upon approval note the change in the control procedure (maybe just a reference to the change order and the supporting information on why the change was approved.
Q: Which tests are most important for A/C Radars? Which 461G tests?
A: Defining which tests are most important requires an analysis of the threats and vulnerabilities. Often RE testing is one of the first to get examined because control measures to reduce emissions may contribute to hardening potential susceptibility issues and it is normally not destructive. I consider CE testing as an early test too for the same reasons and it could identify that the RE is cable related. So the importance is not easily defined but the sequence can be used to identify more issues or get the highest threat addressed early – most thin the more benign tests first progressing to the higher probability of damage. Of course, that approach may lead to more retest actions as problems are identified.
Q: Could you give a highlight of something that is left out of test plans that would benefit successful testing?
A: I often see reports that omit measurement of the power current – the report DID specifies inclusion of this information. The test plan should include a step to measure this current. This simple item provides key information about the operation of the item during test. If, for example, a device current at 60 Hz was 3-amps during qualification testing and later a sample to investigate a change was demanding 5.8-amps we would need to understand why the difference.
Q: The EMI control plan DID provides a good outline of suggested topics but are there public domain examples of real EMI control plans available? If so, please suggest a good example.
A: Since most DIDs support contractor format as acceptable, a “good” template would be different depending on reader perspective. I have templates that I have developed but the detail is unique to the product. I normally look at the DID outline and expand the discussion on each element to provide the design guidance. Preparing a good control procedure is a highly technical process – but the DID outline provides a good approach. I also layout the procedure in the order of the DID – this helps the reviewer locate specific information because it follows the DID topic sequence.
Q: Should you have a waiver strategy during the planning process in-mind in case there are non-compliances during testing?
A: A waiver strategy should be included in test procedures to Identify how waiver requests are to be processed (approval authority, review team, etc.); Procedures should also include what I call “escape clauses” to identify what to do if the test article dies from a non-test related failure, a test related failure, non-repeatable anomaly, etc. This should be part of the acceptance criteria section.
Q: Is there a MIL-STD 461 EMC test that you see is more problematic in the qualification process than others? What advice can you give to design for this test to ensure fewer trips for testing?
A: Any of the tests can be problematic if the design failed to consider the compliance requirements. I often encounter issues from failing to consider parasitic effects. For example, a CS114 issue could really be a radiated susceptibility problem – CS114 induced current could create a high-level field radiated from the cable in close proximity to the device. So what appears to be an interface issue could be an aperture issue. In-depth training courses on MIL-STD-461 testing explore these potential issues.
Q: In your experience, how long does the implementation of this EMICP life cycle takes, and how costly it could be?
A: This depends on the complexity of the product and the availability of a conceptual design. Control procedure initial preparation time ranges in the 24 to 160 work hours region with updates needing a few hours. As I mentioned, the initial release is needed 60-90 days prior to a preliminary design review (PDR). This document is part of the PDR and sets the stage for contractor-customer discussions on high-risk elements and may feed into contract modifications. With many years of working on these issues, I find that the value is worth time spent when products meet the compliance requirements first time in the lab.