Matronics Email Lists :: View topic

Matronics Email Lists
Web Forum Interface to the Matronics Email Lists

Get Email Distribution Too! FAQ Search Memberlist Usergroups Register

Profile Log in to check your private messages Log in

LED Ghosting
Goto page 1, 2, 3, 4, 5 Next

Matronics Email Lists Forum Index -> AeroElectric-List

View previous topic :: View next topic

Author Message

donjohnston

Joined: 13 Dec 2009
Posts: 231

Posted: Sun Jun 15, 2014 9:57 am Post subject: LED Ghosting

I've built a Master Warning Panel using LEDs for things like boost pump on, starter engaged, pitot heat on, etc. I'm using Eric's EGRAVR dimmer to control the brightness.  Everything works great.



http://www.velocity-xl.com/blog/wp-content/uploads/2013/08/2013-08-14-IMG_20130814_191418_518.jpg



Except the engine analyzer (low oil pressure) alarm. 



That line is supposed to go to ground when an alarm condition occurs.  But the LED for that warning is always slightly lit. The engine analyzer people said it was designed for incandescent and to put a 1.5k resistor across the LED. 



That eliminates the ghosting, but it also makes the LED dimmer than the rest. In fact, once I dim the panel to about 25%, the engine warning LED does not illuminate at all. 



I suppose that I could put resistors across all the LEDs which would dim all the LED's.  Then again, I could reduce the inline resistor which would increase the brightness but I'm not sure this is the correct fix. 



Anybody have any suggestions?

sprocket(at)vx-aviation.c
Guest

Posted: Sun Jun 15, 2014 10:55 am Post subject: LED Ghosting

There are many elegant fixes to this problem, but the brute force technique may be the easiest to wire.



Assuming your engine monitor has a 5volt output, install a low power (gold contact) relay with the coil between the alarm output and the 5v.  One pole of the contacts gets connected to ground and the other to the LED.



There are ways to do this with transistors, but this is simple and hard to mess up.



Vern



====================================================

Sent from my iThing. It is responsible for all gramma and typo terrors.



 	  Quote: 	
		   On Jun 15, 2014, at 10:57 AM, "donjohnston" <don(at)velocity-xl.com> wrote:

 

 

 

 I've built a Master Warning Panel using LEDs for things like boost pump on, starter engaged, pitot heat on, etc. I'm using Eric's EGRAVR dimmer to control the brightness.  Everything works great.

 

 http://www.velocity-xl.com/blog/wp-content/uploads/2013/08/2013-08-14-IMG_20130814_191418_518.jpg

 

 Except the engine analyzer (low oil pressure) alarm. 

 

 That line is supposed to go to ground when an alarm condition occurs.  But the LED for that warning is always slightly lit. The engine analyzer people said it was designed for incandescent and to put a 1.5k resistor across the LED. 

 

 That eliminates the ghosting, but it also makes the LED dimmer than the rest. In fact, once I dim the panel to about 25%, the engine warning LED does not illuminate at all. 

 

 I suppose that I could put resistors across all the LEDs which would dim all the LED's.  Then again, I could reduce the inline resistor which would increase the brightness but I'm not sure this is the correct fix. 

 

 Anybody have any suggestions?

 

 

 

 

 Read this topic online here:

 

 http://forums.matronics.com/viewtopic.php?p=424883#424883

donjohnston

Joined: 13 Dec 2009
Posts: 231

Posted: Sun Jun 15, 2014 12:20 pm Post subject: Re: LED Ghosting

 	  sprocket(at)vx-aviation.c wrote: 	
		  There are many elegant fixes to this problem, but the brute force technique may be the easiest to wire.



Assuming your engine monitor has a 5volt output, install a low power (gold contact) relay with the coil between the alarm output and the 5v.  One pole of the contacts gets connected to ground and the other to the LED.



There are ways to do this with transistors, but this is simple and hard to mess up.



Vern 	




It does have a 5v output, but not easy to get to at this point. There's a 12v supply that's easily accessible. Would that work? 



Can you give me a part number for the type of relay?



Thanks,

Don

edpav8r(at)yahoo.com
Guest

Posted: Sun Jun 15, 2014 5:27 pm Post subject: LED Ghosting

On Jun 15, 2014, at 4:20 PM, "donjohnston" <don(at)velocity-xl.com> wrote:

 	  Quote: 	
		   It does have a 5v output, but not easy to get to at this point. There's a 12v supply that's easily accessible. Would that work? 

 

 Can you give me a part number for the type of relay?


	




A 12V supply is fine.  It doesn't even have to be sourced from the engine analyzer.  12V from the supply side of your dimmer would work, or from anywhere else that's convenient.



A relay like this...



http://bit.ly/1voFDLC



..should work a treat.  This relay's coil current is only 12.5 milliamps, so it will be safe to connect directly to the engine analyzer's alarm output.



Connect 12V to pin 2, the engine analyzer's alarm output to pin 9, your LED indicator's ground line to pin 5, and ground to pin 10.  Ignore pins 1 and 5 (or snip them off).



I think it's pretty unlikely that the "leakage" you're seeing into the alarm output pin under alarm off conditions would be sufficient to close the relay.



Solder directly to the relay's pins, then encapsulate the connections in hot glue, Shoe Goo or similar.  In this service, the relay will probably outlast us all.

Alternatively, you could eliminate moving parts and more than halve the cost by using a P-channel MOSFET like this...



http://bit.ly/UEp2qv



Connect your LED indicator's ground line to the source (S) pin, ground to the drain (D) pin, and the engine analyzer's alarm output to the gate (G) pin.  You may find that it works fine like this, or it may still turn on the LED (it will be full bright now) with the alarm off.  It depends what exactly is happening on the engine analyzer's alarm output pin; a MOSFET's gate is much more sensitive than a relay coil.



If it does turn on with no alarm, just connect the 1.5k ohm resistor you already have from the MOSFET's gate to 12V.  This will pull up on the gate during the alarm off condition, keeping the MOSFET -- and your indicator -- turned off.  When the alarm output is on, it will pull the gate to ground, turning on the MOSFET, which in turn connects the LED to ground.



Heat shrink each pin connection as you assemble, then the whole thing when it's complete.

Very nice indicator panel, Don.  Beautifully made.  How did you make the legend transparency?



Eric

donjohnston

Joined: 13 Dec 2009
Posts: 231

Posted: Sun Jun 15, 2014 5:50 pm Post subject: Re: LED Ghosting

Eric,



That sounds perfect. Is there any need to physically mount the MOSFET? Like with a heatsink? Or can I RTV it someplace once it's all assembled and heatshinked?



I had the mask made by trophy shop... for $20!  I got a tip that they have this black material that they hit with a laser printer. Different fonts, sizes, the works. All I did was email the a PDF of what I wanted and a couple hours later I picked it up.   



Here's some construction details for anyone that's interested. 



http://www.velocity-xl.com/blog/2013/08/03/13-99-annunicator-panel/



http://www.velocity-xl.com/blog/2013/08/15/13-99-annunicator-panel-3/

sprocket(at)vx-aviation.c
Guest

Posted: Sun Jun 15, 2014 6:08 pm Post subject: LED Ghosting

..I think it's pretty unlikely that the "leakage" you're seeing into the 

alarm output pin under alarm off conditions would be sufficient to close the 

relay.



Eric has provided good advice.  I am concerned that you will energize the 

relay due to the output protection structure of the EMS.  If you can use the 

5V this won't be a problem (but you would need a 5V relay).  If it is, we'll 

refund the consulting fee.



V

edpav8r(at)yahoo.com
Guest

Posted: Sun Jun 15, 2014 7:37 pm Post subject: LED Ghosting

On Jun 15, 2014, at 9:50 PM, "donjohnston" <don(at)velocity-xl.com> wrote:

 	  Quote: 	
		   Is there any need to physically mount the MOSFET? Like with a heatsink? Or can I RTV it someplace once it's all assembled and heatshrinked?


	


No heatsink required.  It's a pretty stout device, capable of carrying several amps if heatsinked, but in this application it will only see the 20-30mA that passes through your LED.  If we assume 30mA, the MOSFET will dissipate 0.03A x 0.03A x 0.185 ohm = 0.00017 watts of heat.



MOSFET power dissipation is calculated by multiplying the square of current flow by the resistance between source and drain when the device is turned on: Rds(on) in the datasheet.



You can glue or zip-tie it to whatever mounting point is convenient.

Thanks for the details on your annunciator panel; I'm saving that for future reference!



Eric

nuckolls.bob(at)aeroelect
Guest

Posted: Mon Jun 16, 2014 5:06 am Post subject: LED Ghosting

At 12:57 PM 6/15/2014, you wrote:

  	  Quote: 	
		  --> AeroElectric-List message posted by: "donjohnston" <don(at)velocity-xl.com>



 I've built a Master Warning Panel using LEDs for things like boost pump on, starter engaged, pitot heat on, etc. I'm using Eric's EGRAVR dimmer to control the brightness.  Everything works great.



  http://www.velocity-xl.com/blog/wp-content/uploads/2013/08/2013-08-14-IMG_20130814_191418_518.jpg 



 Except the engine analyzer (low oil pressure) alarm. 



 That line is supposed to go to ground when an alarm condition occurs.  But the LED for that warning is always slightly lit. The engine analyzer people said it was designed for incandescent and to put a 1.5k resistor across the LED. 	


    Wrong place for the resistor. There is a resistor X in

    series with the LED to bias it up to run on the

    14V supply. Resistor Z is used to make the diode/X

    string emulate an incandescent bulb. Try 1000 ohms

    at Z.



 [img]cid:7.1.0.9.0.20140616075849.01f3d938(at)aeroelectric.com.0[/img] 



 

 

   Bob . . .

172e08cb.jpg

Description:

Filesize:

22.29 KB

Viewed:

18667 Time(s)

donjohnston

Joined: 13 Dec 2009
Posts: 231

Posted: Mon Jun 16, 2014 7:42 am Post subject: Re: LED Ghosting

 	  nuckolls.bob(at)aeroelect wrote: 	
		  

    Wrong place for the resistor. There is a resistor X in

    series with the LED to bias it up to run on the

    14V supply. Resistor Z is used to make the diode/X

    string emulate an incandescent bulb. Try 1000 ohms

    at Z.



 [img]cid:7.1.0.9.0.20140616075849.01f3d938(at)aeroelectric.com.0[/img] 



 

 

   Bob . . . 	




Well that would make my life a gazillion times easier!



It makes sense too. I'll give it a try. 



Thanks!

donjohnston

Joined: 13 Dec 2009
Posts: 231

Posted: Mon Jun 16, 2014 11:41 am Post subject: Re: LED Ghosting

Sadly, it did not help.  It does eliminate the ghosting.  But just like when I put the resistor across the LED, it still results in a dimmer LED overall.

nuckolls.bob(at)aeroelect
Guest

Posted: Mon Jun 16, 2014 5:02 pm Post subject: LED Ghosting

At 02:41 PM 6/16/2014, you wrote:

 	  Quote: 	
		  



Sadly, it did not help.  It does eliminate the ghosting.  But just 

like when I put the resistor across the LED, it still results in a 

dimmer LED overall.


	


  Okay, was this 'output' intended to drive

  an incandescent lamp in the first place?

  The symptoms you describe suggest that it's

  an LED specific output and may have a resistor

  already built into the output.



   Bob . . .

donjohnston

Joined: 13 Dec 2009
Posts: 231

Posted: Tue Jun 17, 2014 12:04 am Post subject: Re: LED Ghosting

Yep. 



"The engine analyzer people said it was designed for incandescent and to put a 1.5k resistor across the LED. "

user9253

Joined: 28 Mar 2008
Posts: 1921
Location: Riley TWP Michigan

Posted: Tue Jun 17, 2014 2:56 am Post subject: Re: LED Ghosting

Does the LED have a series resistor?  If so, what is its value?

Joe

_________________
Joe Gores

ceengland7(at)gmail.com
Guest

Posted: Tue Jun 17, 2014 5:38 am Post subject: LED Ghosting

On 6/17/2014 3:04 AM, donjohnston wrote:

 	  Quote: 	
		   



 Yep.



 "The engine analyzer people said it was designed for incandescent and to put a 1.5k resistor across the LED."

What kind of LED are you using? Is it one designed to run directly on 
	


12V (meaning it already has an internal current limit resistor)?

donjohnston

Joined: 13 Dec 2009
Posts: 231

Posted: Tue Jun 17, 2014 6:55 am Post subject: Re: LED Ghosting

 	  Quote: 	
		  Does the LED have a series resistor? If so, what is its value?

Joe 	


Yes. 390ohm (see attached).



What kind of LED are you using? Is it one designed to run directly on 12V (meaning it already has an internal current limit resistor)?

No. See above. 



There's nothing wrong with any of the other LEDs. It's just the one for the engine analyzer.  The analyzer is allowing some current through when it's off. With an incandescent bulb, this is not a problem. With an LED, it is.  Putting a 1k resistor solves the dimly lit LED but it make the LED dimmer than the rest. So I'm trying to solve two possible problems. 1) Keep the engine analyzer alarm LED from being partially lit when it should be off, and 2) have the LED at the same brightness as the other LEDs.



Now it's possible (more like probable) that I'm going to have the exact same problem with the Low Voltage warning since the LR3C voltage regulator is designed for an incandescent bulb also. But I haven't gotten to the point where I can start the engine so I don't know yet.

Warning Circuit.pdf

Description:

Download

Filename:

Warning Circuit.pdf

Filesize:

57.89 KB

Downloaded:

592 Time(s)

user9253

Joined: 28 Mar 2008
Posts: 1921
Location: Riley TWP Michigan

Posted: Tue Jun 17, 2014 9:06 am Post subject: Re: LED Ghosting

I suggest that you put a 1.5k resistor across the LED per the engine analyzer people.  Then replace the 390 ohm series resistor with a 330 ohm resistor.  Or experiment with different values (all less than 390) to get the desired brightness.

Or put a pot in series with the dim LED and turn the pot until the current though the dim LED is equal to the current through one of the other LEDs.  Then substitute a fixed resistor for the pot.

40 years ago I could have calculated the resistor value.  Now trial and error is easier.  

Joe

_________________
Joe Gores

nuckolls.bob(at)aeroelect
Guest

Posted: Tue Jun 17, 2014 9:08 am Post subject: LED Ghosting

At 09:55 AM 6/17/2014, you wrote:

  	  Quote: 	
		  --> AeroElectric-List message posted by: "donjohnston" <don(at)velocity-xl.com>



 

 > Does the LED have a series resistor? If so, what is its value?

 > Joe



 Yes. 390ohm (see attached).\ 	


    Ooops, two LED's in series . . . should have asked about

    that. Okay, I suggest we pay homage to our teachers one

    of whom once opined: ". . . when you can measure what you are

    speaking about, and express it in numbers, you know something

    about it; but when you cannot measure it, when you cannot

    express it in numbers, your knowledge is of a meagre and

    unsatisfactory kind; it may be the beginning of knowledge,

    but you have scarcely in your thoughts advanced to the state

    of Science  . . ." -Lord Kelvin-



    Consider the array of sketches at:



  http://tinyurl.com/oxrwae7



     . . . it would be exceedingly helpful to conduct three

    experiments to measure V1 (voltage across the bare-foot

    resistor/led array; V2 (voltage across incandescent

    lamp); V3 voltage across resistor/led array with anti-

    ghosting resistor added.



 

  	  Quote: 	
		  There's nothing wrong with any of the other LEDs. It's just the one for the engine analyzer.  The analyzer is allowing some current through when it's off. With an incandescent bulb, this is not a problem. With an LED, it is.  Putting a 1k resistor solves the dimly lit LED but it make the LED dimmer than the rest. So I'm trying to solve two possible problems. 1) Keep the engine analyzer alarm LED from being partially lit when it should be off, and 2) have the LED at the same brightness as the other LEDs.

 

 Now it's possible (more like probable) that I'm going to have the exact same problem with the Low Voltage warning since the LR3C voltage regulator is designed for an incandescent bulb also. But I haven't gotten to the point where I can start the engine so I don't know yet. 	


    Can't speak to internals of the engine analyzer but

    the LR3 regulator has a lamp output driver



  http://tinyurl.com/cgnwr5k



    INTENDED to exhibit some pull-down current on the

    lamp even when the LR-3 internal electronics are UN

    powered . . . The circuit above was suggested to soak

    off the effects of this small 'leakage' current that

    was insufficient to light an incandescent lamp but

    enough to tickle an LED into a small light output.



    I suspect a similar thing is happening with the engine

    analyzer.



    You can test the LV warn from the LR3 without running

    the engine. The light starts to flash as soon as you

    turn on the master switch even if the alternator is

    off and/or the engine is not running.



    Get those voltages and we can begin to sort out your

    current difficulty.



    Lord Kelven was also known to have said things like

    "I believe x-rays will prove to be a hoax" . . . and

    “There is nothing new to be discovered in physics

    now. All that remains is more and more precise measurement.”



    These blunders by the venerable gentleman serve to remind

    us that the good student questions everything and seeks

    to sort un-substantiated opinion from demonstrable fact.



    Let us go to the workbench and demonstrate . . . 



 

   Bob . . .    [quote][b]

rnjcurtis(at)charter.net
Guest

Posted: Tue Jun 17, 2014 10:21 am Post subject: LED Ghosting

> Does the LED have a series resistor? If so, what is its value?

 	  Quote: 	
		    Joe


	


Yes. 390ohm (see attached).\

  

  

  

         Perhaps you  could use a zener with a voltage just slitely 

         higher than  the open circuit off voltage output. Put that in 

         series with  the LED, to block the small output, and reduce t

         he series  resistor to bring the brightness up.

  

          Roger

   [quote][b]

nuckolls.bob(at)aeroelect
Guest

Posted: Tue Jun 17, 2014 10:24 am Post subject: LED Ghosting

 	  Quote: 	
		  



    You can test the LV warn from the LR3 without running

    the engine. The light starts to flash as soon as you

    turn on the master switch even if the alternator is

    off and/or the engine is not running.


	


   Stubbed my toe here . . . of course without the

   engine running, you'll have to boost bus voltage

   with an ac mains power supply to get the light to

   go out . . . hence, without that power supply you

   are correct, exercising the LR-3's lv warning is not

   yet possible.



   However, the LR-3 is intended to drive lamps with

   over 100 mA of operating current. So the schematic

   I suggested for one led MIGHT still be good for two

   but in any case, the configuration shown for the

   V3 measurement is good . . . the 1K resistor is

   simply reduced in value until the ghost-light

   goes out. So hooking your annunciator to the un-powered

   LR-3 should offer a platform for validating the

   value of the anti-ghosting resistor.

   Bob . . .

donjohnston

Joined: 13 Dec 2009
Posts: 231

Posted: Tue Jun 17, 2014 11:06 am Post subject: Re: LED Ghosting

 	  user9253 wrote: 	
		  I suggest that you put a 1.5k resistor across the LED per the engine analyzer people.  Then replace the 390 ohm series resistor with a 330 ohm resistor.  Or experiment with different values (all less than 390) to get the desired brightness.

Or put a pot in series with the dim LED and turn the pot until the current though the dim LED is equal to the current through one of the other LEDs.  Then substitute a fixed resistor for the pot.

40 years ago I could have calculated the resistor value.  Now trial and error is easier.  

Joe 	




I tried that. I ended up with no series resistor and it was still dimmer than the others... At least at the bottom end.  At the top end they seemed brighter.

Display posts from previous:

	Matronics Email Lists Forum Index -> AeroElectric-List	All times are GMT - 8 Hours Goto page 1, 2, 3, 4, 5 Next
Page 1 of 5

Jump to:

You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum
You cannot attach files in this forum
You can download files in this forum