> The part might look "wrong" without this information. One case was a 3.15A fuse in >series with a 27R resistor at the 28V supply input. The fuse can never blow (no the >aircraft didn't have 115V 400Hz supplies). > The reason was a pater exercise to obtain intrinsic safety approval without formal >testing. The rules said a specific type of fuse must be used at the input, the minimum >rating of that approved type of fuse was 3.15A! I've been down a similar road. We designed a low power radio remote control to run some Coal Mining equipment. The regulatory bureaucrats in the UK Coal Mining industry for intrinsic safety told us we *had* to use a particular power diode, don't recall the number any more, or they would not approve our product. That diode leaked more current that our product took while transmitting! If we wanted to use any modern diode, they would let us do that if we paid them nearly a hundred grand to "get it tested and approved", and listed by them as an approved component. So we used their diode and put up with the constant complaints about short battery life. :-( -- http://www.wearablesmartsensors.com/ http://www.softwaresafety.net/ http://www.designer-iii.com/ http://www.unusualresearch.com/

On 9/27/09 1:54 PM, "Magnus Danielson" <magnus@rubidium.dyndns.org> wrote: > Javier Herrero wrote: >> Yes, I'm also familiarized with space qualified EEE components, and >> lately I hear the word "heritage" more frequently that I would like ;) >> The only design I've made for aircraft electronics was for a commercial >> aircraft application, and the components selection was a lot less >> restringent than for space (I think also that the system for which I >> worked was also not classified as very critical), so perhaps the >> decision to use a 4046 at its limit was driven by that heritage reasons, >> and that really there was not too much better things around :) The >> venerable 4046 has been around for quite a long time... > > Surely would other alternatives to the 4046 be possible even within the > scope of heritage components? DBM and active mixers should be available > and useable. The CCO (it's actually current controlled with a V/I > conversion) could also be replaced. Logical gates can also be used to > replace the phase-detector. Not as handy, but the aspect of the 4046 > causing trouble could be circumvented if needed. Another aspect is to > improve its performance by design. > > But maybe I misinterprent the heritage aspect her in that it is designs > rather than components that is inherited? Design AND component... Easy to claim heritage if all you do is change a resistor or a capacitor. Tougher to claim if you change the active components, particularly if it's functionally different (e.g. Substituting a 2n3904 for a 2n2222 isn't a huge deal.. Changing a NPN to a PNP and reversing bias, or using FETs instead of bipolars.. That would be a big deal) Keeping those 30 year old databooks is important.. (as we tell the folks who say: "why do you need all that shelf space.. Surely those datasheets are available on line").. 1980s design, modified in 1990s, using parts first available in 1970s, launched in 2000s, gets to Saturn in 2010s, has inflight anomaly.. Oddly, the 4000 series CMOS are still heavily used. Back in the 70s and 80s, I hated those parts because they were ESD sensitive (compared to straight or LS ttl), but by today's standards, they're rock solid, have huge junctions, work at any supply voltage. Jim

Jim, Lux, Jim (337C) wrote: > Design AND component... > > Easy to claim heritage if all you do is change a resistor or a capacitor. > Tougher to claim if you change the active components, particularly if it's > functionally different (e.g. Substituting a 2n3904 for a 2n2222 isn't a huge > deal.. Changing a NPN to a PNP and reversing bias, or using FETs instead of > bipolars.. That would be a big deal) > > Keeping those 30 year old databooks is important.. (as we tell the folks who > say: "why do you need all that shelf space.. Surely those datasheets are > available on line").. 1980s design, modified in 1990s, using parts first > available in 1970s, launched in 2000s, gets to Saturn in 2010s, has inflight > anomaly.. > > Oddly, the 4000 series CMOS are still heavily used. Back in the 70s and > 80s, I hated those parts because they were ESD sensitive (compared to > straight or LS ttl), but by today's standards, they're rock solid, have huge > junctions, work at any supply voltage. I have been a fan of keeping old databooks. Keeps collecting them. For instance, in practice my most used databook is the RCA CMOS databook from 1982. Contains some very useful insights into the 4046 design which the online datasheet does not reveal. I don't have all the things I would wish I had, lacking alot of analog databooks. Cheers, Magnus

Gentlemen, in http://www.elektronik-kompendium.de/public/schaerer/pll4046.htm which unfortunately is available in German language only (the illustrations however speak for themselves!) the author makes an in depth analysis of the "self biasing amplifier" circuitry of the 4046's input pin 14 and explains the dos onts donts with this input. Apart from differences of the vco range due to different cmos technologies most of the chip to chip differences seem to have it's origin in this region. Best regards Ulrich Bangert P.S. > I don't have all the things I would wish I had, lacking alot > of analog databooks. Are you looking for something special? I have a number of old data books that I did not use for at least 15 years, actually they are much older. > -----Ursprungliche Nachricht----- > Von: time-nuts-bounces@febo.com > [mailto:time-nuts-bounces@febo.com] Im Auftrag von Magnus Danielson > Gesendet: Montag, 28. September 2009 00:38 > An: Discussion of precise time and frequency measurement > Betreff: Re: [time-nuts] 4046 variations (was EPE GPS....) > > > Jim, > > Lux, Jim (337C) wrote: > > Design AND component... > > > > Easy to claim heritage if all you do is change a resistor or a > > capacitor. Tougher to claim if you change the active components, > > particularly if it's functionally different (e.g. Substituting a > > 2n3904 for a 2n2222 isn't a huge deal.. Changing a NPN to a PNP and > > reversing bias, or using FETs instead of bipolars.. That would be a > > big deal) > > > > Keeping those 30 year old databooks is important.. (as we tell the > > folks who > > say: "why do you need all that shelf space.. Surely those > datasheets are > > available on line").. 1980s design, modified in 1990s, > using parts first > > available in 1970s, launched in 2000s, gets to Saturn in > 2010s, has inflight > > anomaly.. > > > > Oddly, the 4000 series CMOS are still heavily used. Back > in the 70s > > and 80s, I hated those parts because they were ESD > sensitive (compared > > to straight or LS ttl), but by today's standards, they're > rock solid, > > have huge junctions, work at any supply voltage. > > I have been a fan of keeping old databooks. Keeps collecting them. > > For instance, in practice my most used databook is the RCA > CMOS databook > from 1982. Contains some very useful insights into the 4046 > design which > the online datasheet does not reveal. > > I don't have all the things I would wish I had, lacking alot > of analog > databooks. > > Cheers, > Magnus > > _______________________________________________ > time-nuts mailing list -- time-nuts@febo.com > To unsubscribe, go to > https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts > and follow the instructions there.