Light bulbs

[FlyingDude]

Hi guys,

What is the PN of the light bulbs used in the torpedo-style panel lights in my 67 E model? The light assemblies are #36 on the below diagram with PN A1425-R.  I searched for them online and the picture I found matches what I have. The rheostat is shown as #115 and has PN 919006-1, which is an OHMITE H-8-F2-352A, and that checks too.  So, the light assy PN should match too, but I couldn't figure out what bulb goes in there.  The old inop bulb totally crumbled as I removed it.  This is the closest I got:

https://lasar.com/lights/ge-lamp-14v-1940

Next question will be, what's an LED equivalent?

Thanks...

 

[ShuRugal]

Those are the only bulbs on my plane I haven't swapped to LED, because i don't feel like ripping out the dimmer to replace with a PWM module for dimming an LED.

If you have a 14 volt system, those should be GE-89 bulbs.

[FlyingDude]

Guys, I found : https://www.aircraftspruce.com/catalog/pnpages/13-17741.php

anybody used them?

[Greg Ellis]

I have these in my airplane.  I am not sure if this guy is even still in business.  It was many years ago when I had these installed.

 

Shoulder Lights Ad.pdf

[FlyingDude]

2 hours ago, ShuRugal said:

because i don't feel like ripping out the dimmer to replace with a PWM module for dimming an LED

Why wouldn't you dim the LED with the theostat? I personally don't like PWM controllers.

[65MooneyPilot]

The lowest setting on the dimmer is still to much current to dim the led.

[EricJ]

50 minutes ago, FlyingDude said:

Why wouldn't you dim the LED with the theostat? I personally don't like PWM controllers.

Many LEDs don't "dim" like an incandescent will.   Some LEDs are specifically marked "dimmable", which means it has an internal PWM controller that does the dimming function based on supply voltage.   These are typically polarity insensitive as well, just like an incandescent, so that they are genuinely drop-in replacements for older bulbs.

But if you don't have actual dimmable LEDs, the usual dimming function doesn't work very well.

[PT20J]

Also, an incandescent bulb will get very dim but generally LEDs cannot get as dim without flicker, so they limit the dimming range. But LEDs dim without changing the color temperature. 

[Ragsf15e]

If you do go led, then you can’t use tge lights as a substitute for cabin heat like the old ones…

I switched to led, kept the same dimmer and it doesn’t dim well as eric and Skip mentioned, however, it does provide a nice red light for reading maps or checklists.  I turn it off when not in use.

[FlyingDude]

An LED is a diode and it has a forward drop voltage. You can regulate its intensity by regulating the current through it. The simplest way is to add a series resistor that would drop the voltage difference between the voltage source and the diodes forward voltage at the required current. R = (V - Vf) / I

@EricJ what you explained are the household dimmable LEDs. Household dimmers don't vary voltage. They chop the sinusoidal wave using triacs. So some sort of electronics is needed. 

The best LED dimmers are constant current sources. That corrects for variations in line voltage and diode drop voltage due to temperature, ageing, etc. Constant voltage sources with inline resistors can be used. They can also regulate LEDSs in parallel with incandescents. 

I thought I can give it a try and if I don't like it and still want the dimmability, I could replace the pot. Much higher resistance, much lower power dissipation...

 

[EricJ]

12 minutes ago, FlyingDude said:

An LED is a diode and it has a forward drop voltage. You can regulate its intensity by regulating the current through it. The simplest way is to add a series resistor that would drop the voltage difference between the voltage source and the diodes forward voltage at the required current. R = (V - Vf) / I

That's a really inefficient and ineffective way to dim an LED, though.  Current adjustment doesn't vary LED brightness very much once it is forward biased, and it has a very short response curve between forward biased and off.  The efficient and useful way to dim an LED is to give it full current on a variable duty cycle.

 

12 minutes ago, FlyingDude said:

@EricJ what you explained are the household dimmable LEDs. Household dimmers don't vary voltage. They chop the sinusoidal wave using triacs. So some sort of electronics is needed. 

Yes, that's the PWM (Pulse Width Modulation) controller that can replace the usual rheostatic or transistor controller even for DC systems.   Good dimmable LEDs that are true replacements for incandescent bulbs have built-in chopping power supplies and PWM controllers that respond to the input supply voltage.    

12 minutes ago, FlyingDude said:

The best LED dimmers are constant current sources. That corrects for variations in line voltage and diode drop voltage due to temperature, ageing, etc. Constant voltage sources with inline resistors can be used. They can also regulate LEDSs in parallel with incandescents. 

I thought I can give it a try and if I don't like it and still want the dimmability, I could replace the pot. Much higher resistance, much lower power dissipation...

Resistors generate heat so aren't the preferred method of dropping voltage for dimming the current through incandescent lamps.   Many Mooneys have a transistorized controller that provides a more efficient means to vary the input supply voltage to the lamps.   Ultimately the best way to dim LEDs is with PWM, so either replacing the dimming controller with a PWM controller to drive plain LEDs, or using dimmable LEDs with the PWM controller built-in and using the legacy controller are the typical ways to do it.   Either works, but if the old controller is used or the system mixes incandescent and LED lamps, using  dimmable LEDs will increase the chances that it will work as expected, as it should have the PWM controller built-in.

[rbp]

38 minutes ago, FlyingDude said:

An LED is a diode and it has a forward drop voltage. You can regulate its intensity by regulating the current through it. The simplest way is to add a series resistor that would drop the voltage difference between the voltage source and the diodes forward voltage at the required current. R = (V - Vf) / I

@EricJ what you explained are the household dimmable LEDs. Household dimmers don't vary voltage. They chop the sinusoidal wave using triacs. So some sort of electronics is needed. 

The best LED dimmers are constant current sources. That corrects for variations in line voltage and diode drop voltage due to temperature, ageing, etc. Constant voltage sources with inline resistors can be used. They can also regulate LEDSs in parallel with incandescents. 

I thought I can give it a try and if I don't like it and still want the dimmability, I could replace the pot. Much higher resistance, much lower power dissipation...

 

the accepted method in microelectronics to get linear dimming on LEDs is to use PWM. only hobbyists use analog outputs to vary the voltage.

[FlyingDude]

Nope... Anyway, no need to fight and I won't break NDAs to prove you're wrong and I'm right

[FlyingDude]

1 hour ago, EricJ said:

Current adjustment doesn't vary LED brightness very much once it is forward biased, and it has a very short response curve between forward biased and off. 

Please look at this diode V-I graph. The diode can pass any current (from very little to the max that would break it) at the Vf. Hence the external resistor to adjust current.

 

[rbp]

surely you can sketch out a simple circuit that demonstrates your proposal without breaking any NDAs

[FlyingDude]

2 hours ago, EricJ said:

Many Mooneys have a transistorized controller that provides a more efficient means to vary the input supply voltage to the lamps. 

Yes, that's when you use the transistor like a variable resistor. 

Look at your LED strips. There's a R for each batch of 4-5 LEDs. Its purpose is to set the current. 

As for efficiency... Well after using pots to vary incandescents,  driving an LED this way is a joke.

I'm not saying PWM doesn't work. I put it in my products. But if you don't want the flicker, then you need constant current outputs. 

[FlyingDude]

1 minute ago, rbp said:

surely you can sketch out a simple circuit that demonstrates your proposal without breaking any NDAs

I tried to type it above.

[EricJ]

13 minutes ago, FlyingDude said:

Please look at this diode V-I graph. The diode can pass any current (from very little to the max that would break it) at the Vf. Hence the external resistor to adjust current.

 

The curve you really want is brightness vs current.    Sure it'll conduct more current once it's forward biased, but the brightness vs current curve with this method is not nearly as good, and especially not nearly as efficient, as using PWM.    This is why the world uses PWM to dim LEDs.

[rbp]

2 minutes ago, EricJ said:

brightness vs current curve with this method is not nearly as good, and especially not nearly as efficient, as using PWM.    This is why the world uses PWM to dim LEDs

and this is why I wrote:" to get linear dimming on LEDs is to use PWM"

the bigger question is how you propose to vary the current to control the brightness of LED

[FlyingDude]

6 minutes ago, EricJ said:

The curve you really want is brightness vs current.    Sure it'll conduct more current once it's forward biased, but the brightness vs current curve with this method is not nearly as good, and especially not nearly as efficient, as using PWM.    This is why the world uses PWM to dim LEDs.

Please search the sentence "As the current through the LED increases, the brightness also increases." https://www.sciencebuddies.org/science-fair-projects/project-ideas/Elec_p037/electricity-electronics/how-does-led-brightness-vary-with-current

Photons are emitted in an LED as holes and electrons recombine. That's exactly the very mechanism with which forward conduction occurs in a diode. 

(Yes I shared the first page that I found, but it's sponsored by GM so that gives credibility).

 

[EricJ]

20 minutes ago, FlyingDude said:

Yes, that's when you use the transistor like a variable resistor. 

Look at your LED strips. There's a R for each batch of 4-5 LEDs. Its purpose is to set the current. 

Yes, because you want to drive them at just the right minimum current for full brightness with every PWM cycle.

20 minutes ago, FlyingDude said:

As for efficiency... Well after using pots to vary incandescents,  driving an LED this way is a joke.

I'm not saying PWM doesn't work. I put it in my products. But if you don't want the flicker, then you need constant current outputs. 

The only thing you need to avoid flicker is to have the PWM period higher than human flicker sensitivity, so better than 30 Hz or so, which is very easy to do.  Varying the duty cycle from 0% to 100% above that frequency gives you full dimming range with no flicker. 

[EricJ]

Just now, FlyingDude said:

Please search the sentence "As the current through the LED increases, the brightness also increases." https://www.sciencebuddies.org/science-fair-projects/project-ideas/Elec_p037/electricity-electronics/how-does-led-brightness-vary-with-current

Photons are emitted in an LED as holes and electrons recombine. That's exactly the very mechanism with which forward conduction occurs in a diode. 

 

Yes, but the curve is not very linear, and not very efficient.  As mentioned, PWM is used as a simple solution to these shortcomings.

I've been dimming LEDs by PWM duty cycle for nearly forty years.   Mostly as status lamps on circuit boards.   

 

[Kelpro999]

That’s it…..  I’m using lamp shades provided I can find a top projecting LED that’ll fit in the lamp housing. 

[FlyingDude]

12 minutes ago, EricJ said:

I've been dimming LEDs by PWM duty cycle for nearly forty years.   Mostly as status lamps on circuit boards.   

I've worked on autonomous driving, incabin and driver monitoring and machine vision with cameras trying to read dimmed LED road signs. It's a mess because cameras see the flicker and in the end don't capture the whole frame. So I know what I'm talking about 

[FlyingDude]

15 minutes ago, EricJ said:

but the curve is not very linear,

Which curve? Current vs brightness? Please share the curve.