Menu Sign In Contact FAQ
Banner
Welcome to our forums

PWM for dashboard/instrument lights

Traditionally, the little light bulbs in instruments are dimmed with a rather solid rheostat. Nice analog simplicity, solid, dependable. Today it seems only natural to use a simple PWM-dimmer instead (bar certification issues) but would that not create too much interference into the various HF systems?

Apart from certification and HF interference, any other thoughts?
Are any G/A planes already so equipped?

Background: yes, I feel tempted to create and install such a contraption, even if I could never legally fly when instrument lights are needed or required. Just “because I can” !

EBZH Kiewit, Belgium

I have not seen switch mode dimmers but I have seen dimmers where the heat was dissipated in a remotely mounted power transistor instead of being dissipated in the dimmer pot (a rheostat) which is an unwieldy component with a fairly short life.

This is the Socata GT version

It is a bad design done without much understanding of electronics because the output voltage depends on the load and on the current gain (hfe) of the transistor. I don’t know why they didn’t do an emitter follower.

Socatas do get problems with interference from little red fluorescent lamps mounted on the underside of the top of the instrument panel. These are a bizzare solution; some would say a typically complicated solution to a simple problem. And expensive to fix if they go wrong. Obviously today these should be replaced with LEDs but some people are afraid that this is illegal.

Administrator
Shoreham EGKA, United Kingdom

Interference problems are likely to come from: either interfering with audio circuits (think of the GSM bip-b-b-bip-b-b-bip sound) or harmonics from the PWM circuit (a square wave will have harmonics going up far above the fundamental frequency, and the wires to the lamps may act as antennas). I’d have to imagine at typical PWM frequencies, any harmonics that start getting up into radio frequency are going to be pretty weak, though.

Andreas IOM

Socatas do get problems with interference from little red fluorescent lamps mounted on the underside of the top of the instrument panel.

These things can be found in a lot of other aeroplanes, but mostly white, not red. Cessna calls them “moonlight” in the Citations. But they are harmless compared to the high-voltage electroluminescent panels for switches and fuses. One can easily tell that “night mode” was switched on without looking at the switch: The constant humming sound in the intercom is indication enough. Therefore, I think that PWM will be a similar nuisance, unless every single cable is shielded full length.

Last Edited by what_next at 16 Feb 11:53
EDDS - Stuttgart

Thanks @Alioth, my own thoughts were rather similar.
@Peter, it seems that the more budget some engineers are given, the more nonsense they turn up. I doubt such a design would have been produced either in the USA or in Russia, both having some no-nonsense tradition in engineering.

EBZH Kiewit, Belgium

One can design a SW power supply which doesn’t interfere, where efficiency is not much of an issue. You just shape the switching waveform so it has slow edges. Normally that would be commercial suicide (in say a DC-DC converter) because the efficiency might be only 70% but in this app it is fine.

This is how I would do a lamp dimmer where say 10W can be safely dumped into some bit of the airframe

The lower resistor is chosen to make the lowest output voltage about 10V (with 24V bulbs) at which they will be really dim, so there is no point in going lower.

One should insert a resistor where the arrow points to make sure that – with “24V” bulbs – the output voltage is about 20V, for long bulb life. I don’t think there are actually “28V” bulbs.

The same thing will work in a 12V system.

The TIP121 is just one of many options and if I was doing this for a plane I would try to get hold of one of the old TO3 cased devices, for reliability, mounting it on a solid aluminium oxide washer.

There is a significant safety issue with instrument illumination because it is a single point of failure for flying the plane. At night, if it fails, you might be screwed. My LH EHSI dims off the instrument lighting bus but the RH EHSI (which has a switchable AI feature) is dimmed manually, with a pot to the RH of it – panel

Administrator
Shoreham EGKA, United Kingdom

Peter wrote:

You just shape the switching waveform so it has slow edges.

Yes, I thought along that line too. Something like an R/C/R T-filter between the square wave source (555?) and the switching FET. Efficiency is not totally irrelevant, given the limited amount of power available from a 912. But, again, this is really a gimmick only, for me.

Apologies to those unacquainted with electronics hardware.

EBZH Kiewit, Belgium

I’d actually use an Atmel ATtiny 13 rather than a 555 – they are about the same price as a 555, and you can control the PWM in software and use pushbuttons instead of a potentiometer (pushbuttons generally having more options to be made more reliable).

I’ve made switch mode power supplies using 555 timers though, proper PWM ones by having the feedback going to the control pin for running 170 volt nixie tube displays :-) Very versatile chip.

Andreas IOM

With push buttons, you need to store the last setting in an EEPROM. Easy enough but it’s amazing how many people don’t do that – e.g. the Davtron instruments which always lose their last selection.

One could argue that after a power cycle you want max brightness, but maybe not?

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

Back to Top