Menu Sign In Contact FAQ
Banner
Welcome to our forums

Random avionics internals

Peter wrote:

So, maybe an alternator fault detector in a 12V aircraft (whose bus voltage is 14V)?

That would be very well possible. At least Socata always had quite clever electrics, compared to Cessna or Piper for example. The avionics bus switching is a good example, using paint over well cleaned ground connections, are things that are non excistent Cessna and Piper of the same era.

JP-Avionics
EHMZ

NARCO AR-850 altitude encoder.

Component date codes suggest it was made 1987. The ICL7109 is a 12-bit A-D converter and they appear to be using a 2716 EPROM to convert to Gray code, with a lookup table.

I can’t see the pressure sensor type.

Oddly enough, while they chose the LT1008 op-amp very well (an amazing spec device in those days) including its temp range, the ICL7109CPL is not specced below 0C

Good old Narco and their TSO approvals






The pressure sensor appears to be heated by the 2N6307 power transistor, hence the big blue insulation block.

Total parts cost in 1987 would have been about $50 (100+ batch), plus whatever the pressure sensor cost.

Administrator
Shoreham EGKA, United Kingdom

PS Engineering PM-501 intercom.

From component date codes, made around 1996. Might be a much older product since it doesn’t use surface mount parts.

The two pots are interesting. I wonder who makes them?

A quite well made product.


Parts cost maybe $50 (100+ batches).

This one may be working so if anybody wants it, they can have it for the cost of the postage.

Administrator
Shoreham EGKA, United Kingdom

the ICL7109CPL is not specced below 0C

Ah, a 7109. I remember it well. I’m not sure the 2716 is rated below 0C either. Maybe the 2N6307 put out sufficient heat to keep it all warm on a chilly day!! It does raise a reliability question though, for a unit rated -20C to +55C!!

PJL
EGMD, EGKA

This shows that avionics approvals mean nothing if the vendor wants to cheat, ot just makes a mistake. But choosing components with the right temp range is the most basic thing… This designer didn’t have a clue.

I used the slightly similar 7107 back in 1978 – in the 501D. Same era as the above avionics really

I don’t think that heating arrangement would have warmed up the internals much. Anyway is it possible to meet the TSO environmental requirements with a heater? I somehow doubt it, because it would be a fantastically easy solution to stuff like some autopilot components (servos) which need to work (in some airframes) down to -55C and there you are facing a poor choice of components. It is quite tricky to design a brushless version of say a King servo wholly with such components. Even buying the motor with the manufacturer specifying that temp range can be tough, which is probably why the King servos use the crappy Globe motors.

I read somewhere that a 737 has a heated gearbox for the rudder but somebody there would have had a lot more $$$ to throw at it.

Administrator
Shoreham EGKA, United Kingdom

Peter wrote:

I read somewhere that a 737 has a heated gearbox for the rudder

Isn’t the 737 rudder actuated by a hydraulic PCU?!

No idea – maybe @A_and_C knows; he has an engineer license (as well as a type rating) for all this stuff. It could be some other type.

Here is another altitude encoder – an AMERI-KING AK-350.

It turns out to be a copy of the NARCO one above Same 7109 chip, same Gray code translation with an EPROM but this time it is an OTP (windowless) EPROM i.e. a “PROM”. Made around 1998 so this is ultra modern but still built conventionally. Probably the 7109 never came out in an SMT version and if you need to flow solder one big chip, plus the connector, then you may as well stick to old stuff. This design is really dirt cheap, and with a list price of about 20x the parts cost everybody in the pipeline is very happy installing these

Note that also like the NARCO one the TSO certification is fake, due to the temp range of the A-D converter. Another fake-TSO indicator is an ICL7660CPA which is 0-70C also

This one may even be working and I will keep it for experimentation in case I one day build something which needs a pressure altitude.



Obviously there is no factory calibration in these encoders, apart from the zero and span. Yet, lots of altimeters etc fail the FAA 2-yearly static check due to a nonlinearity which can’t be tweaked by tweaking the subscale position. If this design was implemented the obvious way i.e. with a microcontroller then you could factory-load a whole calibration curve which would give you a head start.

I wonder how many of these crappy encoders get chucked on the scrap heap because of this. But I suppose it is good business for everyone – so long as they all make the same design and nobody rocks the boat.

Administrator
Shoreham EGKA, United Kingdom

KI266 remote DME indicator – a real vintage piece from c. 1976

The Positronic connectors – originally developed in the Apollo era and AFAIK for that programme (or similar)

Some milspec (-55C to +125C) ceramic package logic chips

What looks like moulded carbon resistors… these were not accurate (maybe 20%) and not stable, so the designer had to be competent. I don’t know that that large round module is – probably a transformer or an inductor, with a shield around it

The usual fluorescent / luminescent displays, and a significant investment ($10k+) in the diecast front panel

This shows the level of diligence in those days: a layout diagram on the inside of the covers

The strange thing is that for all the robust design etc, and the very nontrivial price which this would have sold for (probably 20x the parts cost), no attempt has been made to protect from moisture. Not even a conformal coating on the PCB. EDIT: they have applied a conformal coating on the “analog” PCB but not on the logic PCB…

Administrator
Shoreham EGKA, United Kingdom

Finally there is a very interesting thread for me! I wonder why did it take six pages until the old trusty GNS-430 appears here :-). I hope to see more interesting pictures coming. I should have some SR22 Avidyne internals documented at home, will try to post here later. It is nothing more than industrial PC.

LKHK, Czech Republic

Here is the well known King KA-33 avionics fan, which is used all over GA to blow air into the back of the various boxes in the avionics stack

I got my hands on a scrap one. It was actually working but the fan unit was rubbing on something. Out of interest I contacted the maker of the fan unit itself

which is fairly obviously not an aviation product, but they are evidently one step ahead of the game, having been getting many enquiries from places like Angola, Jupiter, and maybe even Europe…

Of course it is delightful to be aware that the one they sell you for that outrageous price is ITAR compliant

Opening up the motor was fun. The case is made of soft steel; possibly mu-metal, and there was no evident way to get inside without damage:

The motor uses electronic commutation (no brushes; they use Hall effect sensors apparently) but the control “electronics” is the worst POS I have seen in GA:

In fact the whole construction is utter crap and I am astonished that (a) it ever got certified (well, I know the answer to that one) and (b) King/Honeywell put their name on it.

You can buy the complete unit on US Ebay for $150, in very good condition, but Euro-reg owners can’t take advantage of that (officially).

Administrator
Shoreham EGKA, United Kingdom
Sign in to add your message

Back to Top