What's new
Van's Air Force

Don't miss anything! Register now for full access to the definitive RV support community.

G3X discrete input with inline LED?

storkeye

Member
Friend
I'm just in the final stages of planning my panel. I'm going to install a few 12v 20ma LEDs on the panel to alert for things like Canopy unlatched and Low Oil Pressure.

It would be nice to have this show on the G3X as well as the LED light up.

Does anyone know if it's possible to wire these LEDs inline (in series) with the 'Discrete Input' so that the LED lights up as the signal is passed to the G3X?

I believe the discrete inputs are Active Low which I assume would mean the positive leg of the LED goes to the discrete input and the negative to ground. But will this configuration activate the LED?
 
Nope but

Hi,

The inputs on the G3X won't flow enough current. They are designed to sense rather than flow current so they have a lot of resistance.

If you see a diode designed to work with 12ish volts it has an internal resistor built in with it to slow the current (or it burn out really really fast). Normally when building a circuit you set a resistor to get a specific current, and then put the diode in between. (very googleable)

A diode without an internal resistor will only drop 0.3 volts or so which the G3X will still see as high, so you could use a diode and a separate resistor and tap off between the diode and the resistor.

........................ 12v.........11.7v............ 0v
+12 ---- Sensor ----- Diode ----- Resistor ----- GND
..........................................|
......................................... |
......................(Active High G3X in )

(Sorry about all the ..... - I needed them to make it line up.)

If you have an active low circuit the T comes off between the resistor and ground and goes to a G3X active low the sensor then goes between the T and ground.

Derek
 
Last edited:
Thank you

Thank you Derek, that's really useful.

The LED's I'm planning on using are 12v and have a built in resistor.

So if I understand it correctly I might be able to use the ground pin on the LED as an active low. I'll test it with a meter of course but that should be after the in-built resistor.

So the way I see it, I would supply 12v to the panel LED (with it's in-built resistor), T off the negative connector of that to the G3X and the micro switch on canopy.

In theory, ground will only be active IF the switch is activated which the G3X will see as an input.
 
I'm just in the final stages of planning my panel. I'm going to install a few 12v 20ma LEDs on the panel to alert for things like Canopy unlatched and Low Oil Pressure.

It would be nice to have this show on the G3X as well as the LED light up.

Does anyone know if it's possible to wire these LEDs inline (in series) with the 'Discrete Input' so that the LED lights up as the signal is passed to the G3X?

I believe the discrete inputs are Active Low which I assume would mean the positive leg of the LED goes to the discrete input and the negative to ground. But will this configuration activate the LED?

I did this for my door open light. The door sensor switch closes to ground. That ground went to both a relay coil (12V fed to the other side) to illuminate the LED lamp and the discrete input for the G3X. You could probably skip the relay. The discrete is always sending a positive voltage. Once that circuit completes a path to ground, the G3X will sense the voltage flowing out of the discrete and activate appropriate software function. I wouldn't wire this through an LED, as the input may not be designed to pass that much current and could cause damage. Also possible that the G3X sends 5 volts and not 12 on the discrete.

Larry
 
Last edited:
Abe,

Derek is sort of right. The voltage of a forward biased LED is more like 2V - 3V giving a wide berth for various types but the concept is essentially correct.

Lets talk about the G3x input detection and lighting the LED separately. Active low means the input needs to below the low voltage threshold to be considered active. So if your external switch is connected on one side to ground then if the other side is connected to the G3x input it will sense an active low when the switch is closed. This assumes that the input is in the in-active or high state. I'm not looking at the G3x install manual at the moment so I'll keep my comments somewhat generic. Ok, so for the input to not sense the active state all the time the input needs to be high when the switch is not closed. This is normally achieved by having a pullup resistor to the positive voltage. So normally the input is pulled high and when the switch is closed the input is pulled low. Sometimes internal pullups are available which can be configured. I'm not sure if the G3x has this feature or not since I'm not looking in the manual as I said.

Now lets move onto lighting the LED with that switch. You connect your LED anode end which on a 12V LED indicator may be denoted at +12V to +12V. Then as Derek said there is a series current limiting resistor internal to the indicator. Then you connect the cathode or - end of the indicator to the switch non-ground end. Then when the switch is turned on the - end of the LED will be connected to ground and the positive end will be connected to +12V and the LED will turn on.

Now just combine the two functions. I believe you will need a separate resistor to act as a pullup since the LED in series with the resistor will not have sufficient leakage current to bias the input high. I could help you figure out the required pull-up resistor value but I need more info from the install manual to select it properly.
 
Garmin says

Thanks Ray, I’d really appreciate any help with this.

The indicator I’ve chosen is this one: https://docs.rs-online.com/ce08/0900766b801bc97f.pdf

The Garmin manual says...

When configured as Active Low, the inputs conform to the following specification:

Active: Vin < 3.5 VDC, or Rin < 375 Ω
Inactive: Vin > 8 VDC, or Rin > 100 kΩ

When configured as Active High the inputs conform to the following specification:

Active: Vin > 8 VDC, or Rin > 100 kΩ
Inactive: Vin < 3.5 VDC, or Rin < 375 Ω

Is that what you needed from the manual?
 
Abe,

Refer to Figure 30-2.2 in the install manual. It shows a 5K resistor as the pullup for active low signals. 5K isn't a standard value. You can use a 4.7K, 5%, 1/4 Watt resistor or a 4.99K, 1%, 1/4W resistor. There is nothing critical about this value. If you go bigger (higher resistance) you might not meet the G3x voltage thresholds, so don't do that. If you go smaller though you will just be burning slightly more current when the input is active and the resistor has the full voltage across it. A 1/4W resistor has plenty of margin for its power rating.
 
G3X +GEA24

The GEA24 has 4 discrete inputs.
There are seven (GP1-7) that can be used as discrete inputs as well.
They need a pull up resistor. Check attachment.
Engine instruments like oil pressure transducer has a setting for low oil pressure. You don´t need a separate low pressure oil switch.
You can label all the inputs with whatever you like.
An alarm can be programmed as Master Caution or Master Warning.
If you get an alarm for a door not closed this will come up as a Master Caution
if so configured and presented on the EFIS as DOOR WARNING.

Good luck
 

Attachments

  • GEA24.pdf
    227.6 KB · Views: 183
Oil pressure switch

Abe,
The oil pressure LED will probably need a separate pressure switch sender. If you are using the Vans manifold on the firewall then there is a port that allows both a pressure sender and a pressure switch to connect to the oil pressure line.
This set up will give a bit of redundancy for oil pressure warning in case the glass fails completely and it also will remain lit if the master is accidentally left on at the end of a flight.
Similarly, you could also install a second micro switch for the canopy LED to keep things simple.
 
Why not just wire them in parallel? Current will only flow to the LED and you don't have to worry about interfering with the G3X (also, it's a cheap way to have redundant warnings - otherwise if your LED fails, the G3X won't show the alert either).
 
Resistor

Ray/Avanza - If I'm reading the diagram correctly Discrete In 1 - 4 won't need a pull up resistor but GP1 to 7 would. I suspect that's because the GP inputs can be high or low whereas the Discrete Ins can only be low. Is that right?

Paul - I like the redundancy of a seperate pressure switch. It would be a bad day to see the glass go off and the oil pressure light come on! :eek:

Rodrigo - I think that's what the design has come to, wiring the Discrete In to the negative output is in parallel so an LED failure shouldn't affect the EFIS.

The LED's have arrived so I'll be testing this all out in the shop later today.

Thanks everyone for all your advice.
 
Just another way to skin this cat: I went and built a small circuit board to drive my 12V annunciator LED lights from G3X (and other) active low signals:



This board accepts four low-current active-low discrete signals, and uses those signals to drive four independent LEDs (diode/resistor method described above). It also includes a 556 timer that will momentarily test-flash the LEDs when a button is pushed.

Schematic:


Board layout:


The board attaches to the back of my panel, where the LEDs and switch are mounted:



Front view:



I can make the Eagle files available if anyone wants to use this as the basis for their design. Disclaimer: Experimental, amateur built, not flying yet, may not work for your application.
 
Just another way to skin this cat: I went and built a small circuit board to drive my 12V annunciator LED lights from G3X (and other) active low signals:

Just to be clear, the G3X discrete signals are inputs, not outputs, so I assume you are just paralleling the G3X inputs with your module to drive external LED's rather than have an EICAS message displayed on the PFD?

The G3X does have a master caution and warning output that will drive some LED's.
 
This might help. Although it is Dynon (so the required pullups can be configured inside the EMS), the 1n4007 diodes provide the isolation from the 12V'ish levels that the LEDs have.

Dynon will say that you can connect 12V signals directly to their EMS inputs, but if there is an LED, the current will leak through the internal resistors and make the lamps glow, hence the diodes.

Garmin looks like a more conventional input, but the pullups need to be provided externally. You still need the diodes.

http://www.vx-aviation.com/sprocket/photos/panel_elec/schematics-2/EMS.pdf

VV
 
Just to be clear, the G3X discrete signals are inputs, not outputs, so I assume you are just paralleling the G3X inputs with your module to drive external LED's rather than have an EICAS message displayed on the PFD?

The G3X does have a master caution and warning output that will drive some LED's.

That's right, I wasn't clear: I'm using this for two signals from the G3X (master warning and caution), and two from other devices which should also be able to parallel into G3X discrete inputs. I'm not relying on any of those signals to drive LEDs directly, even though they may be able to. The board was designed before I decided what signals to connect.
 
Abe,

Sorry I was traveling yesterday.

I should have taken a little more time to understand the information that you posted from the install manual. It's not totally obvious from the way that it is written, but I think I now understand what they mean. I inserted the information that is important here for the discretes.

When configured as Active Low, the inputs conform to the following specification:
Active: Vin < 3.5 VDC, or Rin < 375 Ω
Inactive: Vin > 8 VDC, or Rin > 100 kΩ


So for the active condition you have your switch to ground so that is no problem you meet the requirements. The inactive statement is more murky.
If we let the supply voltage = 12V for our example then if Rin = 100K and Vin = 8V that works out to 80uA. That also means that there is an internal pullup of 12V-8V/80uA or 50K ohms. Its a strange way to write this spec, but I think I get what they are saying now.

Now it doesn't hurt to have an external pullup but when you are in the active state you will be consuming more current 12V/4.7K = 2.5mA approximately, but give it a try without the pullup.

There was mention by others of needing an external diode. I don't think you will need one. Some experimentation may be needed to confirm all of this. I hope this helps.
 
Back
Top