Tuesday, October 13, 2009

Not So Bright, Avionics Guys

Over dinner the pilots brief me on the maintenance situation. The oil leak on the right engine turned out (after two o-ring changes) to be the governor leaking. Each time they changed the o-ring, the associated sealant held off the the leak for about ten hours, but now it's fixed, using the governor from the airplane that still hasn't gone to Kansas, because it's now waiting for an on-order governor. (Its papers turned up in someone's briefcase). The CHT is still deferred, as they have apparently determined that the replacement gauge was faulty.

The electronic tachs work well, except -- and my co-worker knew me well enough to preface this with "you're not going to like this" -- the numeric readout does not dim in response to the dimmer switch.

He's right. I don't like it. What is the matter with avionics manufacturers these days? I don't think I've had a newly manufactured avionics item installed in an aircraft that has hooked properly and completely into the dimmer rheostat. In this case the circumference LEDs dim, but the display on the face of the tach doesn't. Add that to the fuel transfer light and much of the specialized mission equipment, and the cockpit looks like a fricking pinball machine when I want it to look like London in the Blitz. And not the parts that were on fire. I need my night vision for finding things like conflicting traffic and airports, and I don't appreciate being half blinded because some equipment designer never considered that someone might fly an airplane in the dark and have better things to look at than their blinking lights.

There's a Canadian Tire in town. I'm going out tomorrow morning to get some of that static cling film you put on your car windshield to declare that you love Siamese cats or brake for shoe sales, and I'm going to cut bits the right size to go over the offending pixels at night.

14 comments:

dpierce said...

When you're done making your tach shades, take a picture and send it to someone high-up at the instrument manufacturer. And maybe someone in sales. Sales guys hate product deficiencies, and use customer complaints to berate engineers into making changes.

You never know what you might get out of it. (Of course, they might tell you they SELL a tach with a nice dimmer function, but your company was too cheap to order it ...)

Marc C. said...

I think I'd be calling the manufacturer of the new tachs myself and making sure they truly aren't dimmable and if not, giving them a piece of my mind. That's just dumb.

borealone said...

this is my pet peeve - it's not like we need the bright lights...i want my panel to be just barely illuminated, unless of course I decide it needs to be brighter. Are you listening, avionics guys?!?

nec Timide said...

It isn't just avionics. Everything in my 2003 Suzuki Grand Vitar (which I otherwise love) dims on the reostat, except the radio and the clock/OAT guage. But thanks for the great idea. I know its probably too late but this might look a bit more professional.

Aviatrix said...

Yeah, Nec Timide, that's the stuff. I've haven't done a lot of flying lately, and was wrong about the Canadian Tire, so I haven't got any yet, but that's exactly what I wanted.

jk said...

From your earlier description I know who the manufacturer is. They are, from what I've seen, a very customer service oriented company that stands behind their product. You might, or suggest that your PRM, might contact their support staff.

They helped me out on a piece of equiptment of theirs I had, was removing, and then sold on ebay, due to upgrading gear...

Wayne Farmer said...

The clock display on my 2001 Subaru Forester is above the windshield and is separate from the rheostat-dimmable console, so of course it has to have different rules: it dims when the headlights are on. Not what you want when driving with headlights on for visibility in the daytime.

brian said...

Having had to design a panel to dim, I can attest to the netoriously bad cockpit lighting buses. I was given design requirements to work on a 14VDC bus and 28VDC bus. I did so. The designed lighting curve resembled something like:
High brightness when bus was zero, a little bit above that it went low and gradiated up to high at a couple of volts below max.
The first flight test aircraft we put it in had a lighting bus that adjusted about 1-2 volts centered around 8 volts. The pilots complained that the unit did not adjust brightness. I was told that the new aircraft coming off the line have a different lighting bus (pulse-width modulated, no rheostat) and that our design wouldn't work for that. These days it's hard to come up with something that will work for everything. The engineers know your pain.

Kevin said...

I think I know what manufacturer and tachs you have.

http://www.buy-ei.com/Pages/R-1/R-1_Downloads.html

If you look carefully at the installation and operation manual, you will find that they have their own rheostat - its part number is CP-1.

I had the same problem with a slew of new gauges in my old Cherokee Six, and the CP-1 actually dims ALL the EI gauges. You need to ask your avionics tech to please install a CP-1 and route the two tachs' backlight wires through it. Then you can control the intensity of the display backlights with that control, and the tach LEDs will work on the cockpit instrumentation rheostat.

Hope this helps!

Anonymous said...

All this talk of Rheostats, reminds me of rheostat in the C172 I flew. The panel would got out when the throttle was moved full forward. (common knowledge to others, but not to me.)

So here I am as a student learning about black hole landings without ground based vertical guidance. The approach predictably isn't going well, I declare going around to my CFI, simultaneously pitch to level off, and push the throttle forward, the horizon dips below the cowling, and the entire panel goes dark...

Denial, anger, acceptance.

Didn't move a muscle for a few seconds, figuring we are level pitch, full power, we are climbing, and peripheral vision says that that part of the horizon is staying put.

Eyes adjust to cabin lighting hitting instrument panel, and attitude indicator is happy.

CFI didn't say much the whole way home.

It was fixed before I came in the next day...

Verification word is barke.
Used in a sentence
"That's a nice bight you have on your barke"

Aviatrix said...

Anonymous your C172 makes no sense! (I believe you; I fly airplanes; I know they make no sense) but gah! I'm trying to think of any possible association that could cause that to happen. The throttle slidey thing (I can't quite call it a lever) in a C172 opens the throttle butterfly valve and triggers an accelerator pump -- mechainically --. The same thing happens while the master is turned off. It's not electric. My first guess is that that the throttle action was mechanically interfering with a connection for the panel lighting. Did you find out what it was, or only that it was fixed?

Anonymous said...

The fright school's C172 definitely made no sense that day.

They didn't say much about it, as I still had that look on my face which said, "anything you anything say other than 'it won't happen again' will result in injury to you".

As near as I can tell, there was a loose connection which the throttle linkage would sometimes act on. I am picturing a wire with a ring lug on the end with the nut not tightened down. It might have been related to airplane attitude ie nose down on approach made it easier to snag the wire. As I recall, we did regain panel lighting later on in the flight. It might have resolved itself when I retarded the throttle at cruise.

I think my mind took roughly the same thought process as you outlined, only compressed in time as it happened. The black hole was in front, and there was plenty of lit up civilization to the sides. I had peripheral vision that confirmed that we were climbing and the horizon was staying put.

To make maters worse, I could see/feel the CFI spring to life, and start fiddling with the Rheostat. My Right hand on throttle, his left arm crossed over mine, to reach the rheostat, left of the throttle. He had prior knowledg that there was a problem. This resulted in me wishfully thinking that the lights would magically come back on 'any time now' as the seconds ticked away. Denial

Then Anger, they aren't comming back on, and he's distracting me with his fumbling.

Acceptance, I better keep flying the plane, cause he's not gonna say "my controls" anytime soon, he's fixated on the rheostat.

As we were stable, the lights out was mostly a realistic distraction. Eventually I said, "quit playing with it and put a flashlight on the panel for me". By then we were had climbed a 1000' or so.

Since that day I wear a headlamp after dark. It puts out red or white light, and I can turn it on one handed.

The verification word is "lisortr" pronounced 'lie sorter'

You need a lisortr when you deal with some fright schools.

nec Timide said...

Yeah, Nec Timide, that's the stuff.

It may be the stuff, but my CT doesn't have any :\

zb said...

There's a circuit that works with both a variable voltage from a rheostat and a pulse-width-modulated (PWM) input, and I am sure you can build it with parts that cost not much more than a buck:

You take a sawtooth or triangle oscillator, compare the rheostat or PWM voltage with the oscillator signal and use the output of the comparator for the light source that needs dimming.

When the available voltage is varied by a rheostat, the comparator will turn on the light whenever the adjustable voltage is greater than the oscillator signal and produce a PWM output of the same frequency as the oscillator's.

When the available voltage already is a PWM signal, it will be bigger than the oscillator signal no matter where you happen to be on the sawtooth or triangle wave and the comparator's output will be on whenever the PWM is on. When the PWM is zero, it will always be smaller than any part of the oscillator's wave so the comparator's output will also be low: The output will always do what the available PWM input already does.

Some tweaking of offset voltages and amplitudes might be necessary, but I think the approach should work. Should I file a design idea at EDN? Maybe: Cheap Quad OpAmp Makes any Instrument Panel Display Dimmable.

Disclaimer: Hacking aviation-grade equipment might make it more useful but will void the aviation rating.