HI All,
I'm no expert on making MTBF (Mean Time Between Failure) calculations. But I have participated in these studies as a Subject Matter Expert representing the electronics design and components used in the design (I was the designer). There are many elements involved in MTBF studies. The most important and significant issue is the raw parts count. IOW, the more parts, the more likely a failure. Additionally, every part that is involved in the design have their own MTBF anylysis. So the final calculation takes into consideration the number of parts, the MTBF's of the individual parts, and finally the assembled system MTBF which considers the method of assembly (shock and vibration resistance, temperature resistance, circuit board techniques, soldering and attachment methods, etc.). Then, a suite of tests are performed to verify the MTBF studies. Usually the MTBF studies are pretty accurate. The tests use accelerated stress test methods so the time frame of the test processes is not beyond reason. The accelerated test methods are well characterized and widely used in space and military applications. Once the basic reliability of the item is established, the design proceeds into production and the test suits are simplified since the design has been qualified during the design phase. Consumer electronics are hardly ever tested in this way. I would say never but I'm not sufficiently familiar with all of the companies and their engineering goals to make such a statement.
A given electronics system can be designed with high reliability in mind or no reliability considerations at all. Both methods show up on the market. However, the systems that are designed with high reliability as a design goal are always more expensive because they are more expensive to design, build and more expensive to test. This is the reason that MIL spec products are considered the most desirable but tend to be expensive.
Ham products have no formal quality control requirements other than the ones imposed by the manufacturers themselves. Just what these are is unknown to me. Generally, the only way to have a clue is the cost of the product. This actually makes sense in a profit driven market. There is no advantage to building a more expensive product unless the customer gets something for the added expense. Reliability is one of those items that some people are willing to pay for. This answers the old saying that states that you get what you pay for.
PLB's have formalized reliability requirements the manufacturers must answer to. These have been defined by the federal government. MTBF analysis and testing is driven by rules, results and documentation. Ham radio, have none that I know of. This difference tips the reliability balance towards the PLB's. Remember, reliability is a statistical function. It reports the Probability of a failure not the actual failure itself. It's one of the reasons that PLB's have always been more expensive than competing, non-ham and ham, systems which have no formal reliability requirements.
So, PLB hardware should be theoretically (and practically too) more robust than others. For me, this is reason enough to go the PLB route. However, this is not enough if the human systems are deficient. But, the PLB system includes methods to ensure the human systems work and work every time. The main reason the PLB scheme is so reliable is that it is basically simple. KISS applies here as long as it is sufficient to get the job done. This can only be determined over time. Time and many successful results have shown emphatically that the PLB human systems work, every time. Competing methods have not shown this level of reliability. Until they do, PLB's remain the best method for emergency response.
Sparky
I'm no expert on making MTBF (Mean Time Between Failure) calculations. But I have participated in these studies as a Subject Matter Expert representing the electronics design and components used in the design (I was the designer). There are many elements involved in MTBF studies. The most important and significant issue is the raw parts count. IOW, the more parts, the more likely a failure. Additionally, every part that is involved in the design have their own MTBF anylysis. So the final calculation takes into consideration the number of parts, the MTBF's of the individual parts, and finally the assembled system MTBF which considers the method of assembly (shock and vibration resistance, temperature resistance, circuit board techniques, soldering and attachment methods, etc.). Then, a suite of tests are performed to verify the MTBF studies. Usually the MTBF studies are pretty accurate. The tests use accelerated stress test methods so the time frame of the test processes is not beyond reason. The accelerated test methods are well characterized and widely used in space and military applications. Once the basic reliability of the item is established, the design proceeds into production and the test suits are simplified since the design has been qualified during the design phase. Consumer electronics are hardly ever tested in this way. I would say never but I'm not sufficiently familiar with all of the companies and their engineering goals to make such a statement.
A given electronics system can be designed with high reliability in mind or no reliability considerations at all. Both methods show up on the market. However, the systems that are designed with high reliability as a design goal are always more expensive because they are more expensive to design, build and more expensive to test. This is the reason that MIL spec products are considered the most desirable but tend to be expensive.
Ham products have no formal quality control requirements other than the ones imposed by the manufacturers themselves. Just what these are is unknown to me. Generally, the only way to have a clue is the cost of the product. This actually makes sense in a profit driven market. There is no advantage to building a more expensive product unless the customer gets something for the added expense. Reliability is one of those items that some people are willing to pay for. This answers the old saying that states that you get what you pay for.
PLB's have formalized reliability requirements the manufacturers must answer to. These have been defined by the federal government. MTBF analysis and testing is driven by rules, results and documentation. Ham radio, have none that I know of. This difference tips the reliability balance towards the PLB's. Remember, reliability is a statistical function. It reports the Probability of a failure not the actual failure itself. It's one of the reasons that PLB's have always been more expensive than competing, non-ham and ham, systems which have no formal reliability requirements.
So, PLB hardware should be theoretically (and practically too) more robust than others. For me, this is reason enough to go the PLB route. However, this is not enough if the human systems are deficient. But, the PLB system includes methods to ensure the human systems work and work every time. The main reason the PLB scheme is so reliable is that it is basically simple. KISS applies here as long as it is sufficient to get the job done. This can only be determined over time. Time and many successful results have shown emphatically that the PLB human systems work, every time. Competing methods have not shown this level of reliability. Until they do, PLB's remain the best method for emergency response.
Sparky
