Menu Sign In Contact FAQ
Banner
Welcome to our forums

Random avionics internals

Prob99 a marker beacon receiver, which is some tens of MHz.

Yes for sure whoever designed this just loved using a lot of components A better way would be a little CPU scanning the buttons and switching some muxes and op-amps.

Interesting they use relays and not analog switches. Relays are robust whereas muxes are easy to blow up with static, +24V, etc.

Administrator
Shoreham EGKA, United Kingdom

Yes, it is a marker beacon receiver, and the yellow wires on the right go to white/blue/amber lights.

The use of relays probably also makes it easier to implement the fail-safe feature (no power to audio panel => pilot is connected to COM1)

Last Edited by Ultranomad at 11 Jun 09:20
LKBU (near Prague), Czech Republic

Iranian drone air data computer:



Andreas IOM

Nice work

I am using a more powerful version of that CPU and neither is in any way export controlled (plus the chinese make copies) which shows that the export control business is dead in the water if you want to stop people making cruise missiles and such. All that’s left is hi res (say 1024×1024) thermal sensors with a high frame rate (say 50fps) and high grade inertial nav devices.

Administrator
Shoreham EGKA, United Kingdom

(I think this belongs in the “old avionics” thread but I can’t find it)

My friend has a Twin Commanche, whose Janitrol CO-generator has been replaced by an electric heater, It has its very own 120A 36V alternator. It doesn’t work, seems to have blown the 7A fuse in the power rail. I thought I’d attach the circuit diagram and photo here for some amusement. The circuit is a challenge to understand. The “aux” position was a “selling point” that you could also use it as a backup alternator. Assuming, of course, that you’re OK with connecting a device capable of producing 36V to your 12V avionics.

The guts of the circuit seems to be a discrete-component regulator that modulates the alternator field current to produce the requested voltage. The constructon quality is a bit dubious, especially the hefty diode soldered in place across a couple of heatsinks.


LFMD, France

Moved to that thread

The 3 phase alternator and rectifier can be clearly seen on the right.

The contactor is a SPCO switch which directs the output, via the 120A fuse, to either the heater or the AUX bus.

The alternator output also feeds the top of the field coil through a 7A fuse. The field current is controlled by the MOSFET Q3 whose gate is driven by op-amp U2.

U1 is unknown but R3/R5 deliver a feedback voltage to the op-amp. The setpoint is selected by S1B. R10 and R8 are adjusted to produce either the heater voltage or the bus voltage, respectively.

The construction is horrible and shows that you can get even total s**t certified if you know your way around the process – because certification is a box ticking exercise which has no relationship to build quality, reliability, etc For example, that heatsink is just a totally in-your-face bodge (especially as it is touching a capacitor) and the trimpots (one of which happens to set the bus voltage!) are of terrible quality. This junk reminds me of the utter crap which Sinclair Radionics used to churn out… didn’t stop Clive getting a knighthood though

If you want to get stuck into critique some more, look at the way the legs of that diode are bent

They were bent when the diode was held by its plastic package, so the diode junction is stressed. Together with the obvious heating, here is a great mechanism for a thermal cycling failure.

The date codes are 1988 or so but even then there was no reason to build it as badly.

One is not actually connecting 36V to the avionics bus. The circuit is supposed to regulate some lower voltage in that mode e.g. 12V or 24V. In any normal plane, the alternator is capable of producing way over 100V if you give it max field current (field winding connected directly across the alternator output) but present little load, and that failure mode could obviously blow up all your avionics, which is why a decent voltage regulator (see the Lamar ones in the Lamar thread) also contains an overvoltage trip. This circuit is a voltage regulator without an overvoltage trip.

If the 7A fuse is blowing then I’d say the field coil is short-circuit. Disconnect the field coil and see if it still blows.

Administrator
Shoreham EGKA, United Kingdom

Thanks @Peter! I think U1 and probably U3 are voltage regulators – there’s a 7808 visible in the photo. U1’s job seems simple, to supply a stable reference voltage for U2. But U3…? Apparently I1 is a “failure” indicator, so maybe K1 pulls in as long as there is sufficient voltage across the field coil?

VR1 is “interesting” too. I THINK VR means a Zener diode in this circuit. So it’s just connected straight across the main bus, with no current limiter, which means it could potentially be passing tens or hundreds of amps. No wonder it needs a giant heatsink.

the trimpots (one of which happens to set the bus voltage!) are of terrible quality.

Quite – you’re depending on one of these things to not fry your avionics. I take your point that any alternator can produce way too many volts, but hopefully the regulators are of higher quality than this (although who knows). And evidently there’s no separate over-voltage protection here.

The use of carbon composition resistors in 1988 is pretty amazing, too. And not exactly a sign of quality. Maybe they bought thousands of them when they were still in normal use, say 1960?

LFMD, France

Peter wrote:

This circuit is a voltage regulator without an overvoltage trip.

Isn’t VR1 plus the 5 A fuse the overvoltage trip? Any overvoltage will be shunted by VR1 and if it is more than transient, then the fuse will blow, deenergising K3 and disconnecting the alternator from the aircraft bus.

Not a great design, I agree…

ESKC (Uppsala/Sundbro), Sweden

Any overvoltage will be shunted by VR1

I guess, assuming VR1 doesn’t vaporize first. It’s not actually clear what the 5A thing is, most likely some kind of circuit breaker rather than an actual fuse since the symbol for the 7A fuse is different.

The 7A fuse is the yellow component just visible above the bigger heat sink. From inspection it seems to be some kind of semiconductor fast-blow thing. But that won’t be affected by the zener.

Looking closer, since the 7812 bottom left has a cap directly across its pins (and a nifty bit of soldering on the other pin), U3 is probably the 7808 in the centre.

The meter seems to come from the DC10 school of system design – it measures what the regulator is being asked to do, rather than what it is actually doing. (Like the baggage door closed detector on the DC10). So any component drift or failure in the regulator means it won’t be telling the truth. Not important in heater mode, you can feel what the heater is doing. But matters rather more for the alleged alternator backup mode.

Now my friend understands what this is all doing, and assuming he can persuade the thing to work again, his first step will be to remove the wire linking this Heath Robinson design’s output to the bus.

Last Edited by johnh at 27 Nov 11:20
LFMD, France

Isn’t VR1 plus the 5 A fuse the overvoltage trip?

Yes; well spotted! Crap way to do it though, because that zener will see a potentially massive power dissipation just below the CB-opening condition. Let’s say it is a 35V zener; at 5A that is 175W which will melt it or set it on fire within seconds – even if it has that “heatsink”. So that “circuit” will work only if the alternator goes properly crazy and does it fast.

U3 is probably the 7808 in the centre.

This is another thing. Look up the absolute max input of the 78xx regulators But so many people in avionics screw up on this. CAV (TKS control box) screw up too. These regulators are not suitable for a direct connection to a vehicle bus, 12V or 24V, where there is a starter motor.

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

Back to Top