Need to correct my example: 10 devices fail in the first year. 10% per year. 50% would be at 5 years. So the MTBF is 5 years. Adam's explanation is likely also accurate.
On Sat, Aug 26, 2017 at 12:01 PM, Forrest Christian (List Account) < [email protected]> wrote: > My understanding is that MTBF is usually calculated by taking a certain > sample size, then seeing how many fail over a certain amount of time. > From this number you can then use statistics to determine how many years > the rest will last. > > A simplified example would be if you take 100 devices, and 10 fail in the > first year, then you assume that 5 would fail per year, and the mtbf would > be 5 years. > > Remember 'M' is mean. Or average. Which roughly means that only half of > the units will still be working in that amount of time. It doesn't mean > your particular radio will last that long, just that half of the radios > will last that long. Yours might fail in 10 days or a year or never.... > > Personally, I believe that this method is rather dubious since some > electronic parts exhibit wear-out. Electrolytic capacitors as an example. > Even if very few devices fail at 5 years, there is a good chance that > most will fail at 20 years after the electrolytic caps have dried out. > > On Sat, Aug 26, 2017 at 9:29 AM, Kurt Fankhauser <[email protected] > > wrote: > >> Where do these MTBF ratings come from by radio manufacturers? Are they >> just made up numbers the manufacturer "hopes" that the product can achieve >> or is actual testing done to get to these numbers? I thought i seen a radio >> once with a 90 year MTBF rating. How they hell can they determine that? The >> components in the radio didn't even exist 90 years ago. >> >> If a radio manufacture states in the spec sheets that the radio has a 40 >> year MTBF rating but then also admits that after 4 years expect to have >> problems due to a design flaw, what does that mean? Is the expected MTBF >> rating only good in a "lab environment" under "ideal conditions"? >> >> Seems to me the MTBF is just marketing fluff and actually doesn't mean >> crap.... >> > > > > -- > *Forrest Christian* *CEO**, PacketFlux Technologies, Inc.* > Tel: 406-449-3345 | Address: 3577 Countryside Road, Helena, MT 59602 > [email protected] | http://www.packetflux.com > <http://www.linkedin.com/in/fwchristian> <http://facebook.com/packetflux> > <http://twitter.com/@packetflux> > > -- *Forrest Christian* *CEO**, PacketFlux Technologies, Inc.* Tel: 406-449-3345 | Address: 3577 Countryside Road, Helena, MT 59602 [email protected] | http://www.packetflux.com <http://www.linkedin.com/in/fwchristian> <http://facebook.com/packetflux> <http://twitter.com/@packetflux>
