In theory but who checks it, who tests it? You can still write crap software, so long as you “demonstrate” compliance 
Peter wrote:
In theory but who checks it, who tests it? You can still write crap software, so long as you “demonstrate” compliance
https://en.wikipedia.org/wiki/DO-178B
“The number of objectives to be satisfied (eventually with independence) is determined by the software level A-E. The phrase “with independence” refers to a separation of responsibilities where the objectivity of the verification and validation processes is ensured by virtue of their “independence” from the software development team. For objectives that must be satisfied with independence, the person verifying the item (such as a requirement or source code) may not be the person who authored the item and this separation must be clearly documented."
OK here is another one: a KN63 DME. This one is from year 2000.
Stripped down only partially because this is a working spare unit I have.
Under one of the boards appears to be a 40-pin microprocessor and an EPROM.
I don’t believe this product has changed since the early 1980s.
The RF section is interesting and not dissimilar to the Trig transponder in the video above.
It turns out that the KDI572 display just plugs in! I bet a lot of people in the business always knew this, but said nothing because there is no legal way to change them in Europe… it’s obvious that the simplest way to fix the four-figure priced KDI572 is to pick one up somewhere (they can be found for $100) and change either the display or the rest of it
Only these four little screws need removing. It is very delicate though
Same chip as posted above, this time from a KDI572 dated 1994, a duff unit out of my plane
This 1994 date proves, as I always knew from various indications, that a lot of my 2002 TB20 panel was built from old gear. Given that the avionics warranty bill (2 years’ warranty) came to best part of 100k (!!) I reckon Socata used up a pile of avionics which customers returned with intermittent failures. Even if they had just bought used avionics from various sources and generated JAR-1 forms for them, they would not have got so many failures.
There is no way you could built a plane in 2002, with new avionics, and ended up with stuff from 1994, given that these products were really mainstream back then, and sold in volume. Enough volume to make it worthwhile for BK to get the above custom chip done.
BTW on that DME I posted pics of above you can get a reduced exchange repair deal provided the RF transistors are not blown up!
One of them can be seen as MSC1536 and indeed it is practically nonexistent. It is so weird it doesn’t even google to a data sheet; only a description.
Here’s another little one, for those who get bored of the techy ones
And, yes, more of the junk we all fly with… the ubiquitous Whelen wingtip light assembly.
No real attempt has been made to make it weatherproof.
Well, the strobe portion has a bottom cast in silicone. That is a lot of fun; I used to work in such a shop in the 1970s and when the boss wasn’t looking we used to pour the excess into animal moulds, and occassionally into johnnies (as they were called then)
There are rubber seals under the two end lights but nothing really holds the glass against these, and the RH end (the tail light) is totally open.
The power supply is a bit better made:
Conformally coated PCB.
Construction is OK
Interesting use of custom potted modules
but those three power diodes, sticking up, is very poor practice because they fail in vibration
The above stuff is 20 years old. All going on Ebay shortly.
The LED version – sorry forgot to get photos – is not actually better in terms of waterproofing, despite having a beautifully CNC machined base
Well, IP68 is hard and needs more bulk to contain the seals, as well as stiffer construction to make sure the seals are compressed along their whole length. They acknowledge water will get in so they have a drain hole at the bottom. I fear the LED ones will last as long as other LED lamps last outdoors: until the PCB corrodes. OTOH they do get pretty warm – +65C or so on the bench – so moisture should get expelled.
Tangentally related to avionics, but there’s some commonality here:
Oh yes – similar era and similar ceramic chips to the KDI572 further above.
This is from the back of a Socata (pre-GT I believe) instrument cluster. I think these adjust the zero and scale of the various moving coil meters
Those black trimpots are horrible. They are almost the cheapest nastiest trimpots money can buy. They are certainly the cheapest nastiest enclosed trimpots money can buy. Not stable in the short term and definitely not stable over years.
The problem is that unless you can find the calibration procedure and the test set-up documented (it will all be Socata-internal) you cannot just change them. What should be possible, with considerable care, and assuming a fully working and accurate system, is to extract each one, measure its setting, and replace it with an identically adjusted high quality trimpot. Or better still with 2 resistors.
