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

Using LEDs in Series

Matronics Email Lists Forum Index -> AeroElectric-List

View previous topic :: View next topic

Author Message

Eric M. Jones

Joined: 10 Jan 2006
Posts: 565
Location: Massachusetts

Posted: Mon Jul 30, 2012 7:35 am Post subject: Using LEDs in Series

 	  Quote: 	
		  This raises a good question: is it okay to light a strip of LEDs by just wiring them in series in a quantity such that each LED sees the appropriate voltage drop? This would seem to eliminate the (admittedly small) energy loss from heating up resistors. On the other hand, failure of one LED would darken the whole string. Thoughts?Dan 	




Great questionâ€¦and one that needs a proper answer.



LEDs have a characteristic called Forward Voltage (or Vf). This means that the LED will not light until the voltage across it exceeds the Vf. This voltage is usually small , perhaps 2Vf and depends on the diodeâ€™s materials of construction. The diode itself follows Ohmâ€™s Law V=IR (or E=IR) when the voltage is above the Vf.



Letâ€™s see how this works:



Put a common LED across a 12V source, and the current will be extremely large, since 12V/~0 Ohms=Huge current. Now we have to understand that the battery or source of power and the wires, etc. have some small resistance, so the current is not quite infinite, but some very large value. (Ka-blamâ€¦.) But the LED lowers the voltage by its forward voltage (here 2 Vf). Even so 12V-2 Vf =10V and 10V/~0 Ohms is still huge.



So we have the bright idea that we can simply use 6 x 2 Vf LEDs, we can get:



12V-2Vf -2Vf -2 Vf -2Vf -2 Vf -2Vf = 0 volts/~0=hmmmmm. The LEDs in this case are each at the voltage where they can almost light. But this is no good for anybody. If we use 5 x 2V LEDs, then 12V-2V-2V-2V-2V-2V= 2V/~0=Current (probably excessive) that is dependent only upon the residual resistance and true source voltage of the circuit. Furthermore it depends on the LEDsâ€™ Vf being exact and independent of all outside influencesâ€¦which it is most certainly not. Additionally if you build a circuit where the LEDs are on the hairy edge of lighting, dimming them is problematic.



What to do, what to do? â€”Add a resistor to the string of LEDs If we use 5 x 2Vf, and want a current of 30 mA, then 12V-2Vf -2Vf -2Vf -2Vf-2Vf =2V/.030A=67 Ohms. This 67 Ohm resistor swamps any likely source resistance (which is usually just a few Ohms), and doesnâ€™t allow for much slump in the voltage source (10V and these LEDs are off) The resistor needs to be a wattage of at least I^2 X R=0.030A X 0.030A X 67 Ohms=0.06 Wattsâ€¦very small in this case but the wattage calculation of LED current-limiting resistors is important and frequently neglected.



Additional note: The voltage source can be as large as one likesâ€¦even kilovoltsâ€¦but the resistor must be suitable for the voltage.



Additional note: Vf is not a parallel-is-lower type of characteristic. Vf  values sum in series, but do not sum in parallel, exactly like batteries! This makes putting banks of LEDs in series and parallel together a practical approach. 



Conclusion: When using LEDs in a string, the current limiting resistor must have a high enough value to make ignoring the source resistance reasonable, and allow for dimming capability headroom as desired.



Eric M. Jones, Perihelion Design, 07/30/12

_________________
Eric M. Jones
www.PerihelionDesign.com
113 Brentwood Drive
Southbridge, MA 01550
(508) 764-2072
emjones(at)charter.net

rv8iator(at)gmail.com
Guest

Posted: Mon Jul 30, 2012 8:35 am Post subject: Using LEDs in Series

 	  Quote: 	
		  LEDs in series is not as potentially problematic as incandescents in series.  Incandescent bulbs are relatively delicate and short lived as they are a heater that emits light.   The life of an LED that is driven at it's rated current (not over driven) and properly heat sinked (in the case of high brightness LEDs >1W) is tens of thousands of hours.  In the developing industry of SSL (solid state lighting) it is common practice to series connect LEDs.  In most cases this is preferred as the current is held constant and the voltage is varied depending on the number of LEDs in the string (as Eric has described).  If you are driving high output devices >1W you will need to use a constant current driver.  You will find that as you limit current on these higher output devices a limiting resistor will dissipate considerable power, much more than the LED.

 	




Chris Stone

RV-8

Newberg, OR

 

 

[quote]  

 > This raises a good question: is it okay to light a strip of LEDs by just wiring them in series in a quantity such that each LED sees the appropriate voltage drop? This would seem to eliminate the (admittedly small) energy loss from heating up resistors. On the other hand, failure of one LED would darken the whole string. Thoughts?Dan

  

 <snip>

[b]

Eric M. Jones

Joined: 10 Jan 2006
Posts: 565
Location: Massachusetts

Posted: Mon Jul 30, 2012 10:06 am Post subject: Re: Using LEDs in Series

One last thing:



In many cases a simple current-regulated power supply is all you need to drive series/parallel LEDs. This also has the advantage of being essentially noise free. I sell an LED tail light where I used a tiny switching power supply but never could stop the noise (Bob N. has a solution to this...but it is a little complicated) The problem was particularly bad because the LED is in the tail...and frequently so are the antennas. So I changed to a simple current-regulated supply that is a little inefficient, but stone-silent.



I have no doubt that there are switching supplies that are very quiet, too.

_________________
Eric M. Jones
www.PerihelionDesign.com
113 Brentwood Drive
Southbridge, MA 01550
(508) 764-2072
emjones(at)charter.net

henador_titzoff(at)yahoo.
Guest

Posted: Mon Jul 30, 2012 12:01 pm Post subject: Using LEDs in Series

Eric, a "simple current regulated power supply is nothing more than an active circuit providing variable resistance to keep the current constant while the applied voltage fluctuates.  This means that power is dissipated in its "variable resistance circuit, usually a transistor in its active state.



A switched current regulator, on the other hand, keeps the switching transistors in their on-off states with power dissipated mostly during the switching time when they're in the active state.  This has the disadvantage of generating transients and harmonics which may bleed into the wrong places.

Do you have a magical circuit that somehow doesn't generate RF interference and also doesn't dissipate very much power?

Thanks.



Henador Titzoff



--- On Mon, 7/30/12, Eric M. Jones  <emjones(at)charter.net> wrote:

[quote]

From: Eric M. Jones <emjones(at)charter.net>

Subject: AeroElectric-List: Re: Using LEDs in Series

To: aeroelectric-list(at)matronics.com

Date: Monday, July 30, 2012, 11:06 AM



--> AeroElectric-List message posted by: "Eric M. Jones" <[url=/mc/compose?to=emjones(at)charter.net]emjones(at)charter.net[/url]>



One last thing:



In many cases a simple current-regulated power supply is all you need to drive series/parallel LEDs. This also has the advantage of being essentially noise free. I sell an LED tail light where I used a tiny switching power supply but never could stop the noise (Bob N. has a solution to this...but it is a little complicated) The problem was particularly bad because the LED is in the tail...and  frequently so are the antennas. So I changed to a simple current-regulated supply that is a little inefficient, but stone-silent.



I have no doubt that there are switching supplies that are very quiet, too.



--------

Eric M. Jones

www.PerihelionDesign.com

113 Brentwood Drive

Southbridge, MA 01550

(508) 764-2072

emjones(at)charter.net





Read this topic online here:



http://forums.matronics.com/viewtopic.php?p=379610#379610



http://www.matroni/forums.matronics.com" target="_blank">http://forums.matronics.com [quote][b]

Eric M. Jones

Joined: 10 Jan 2006
Posts: 565
Location: Massachusetts

Posted: Mon Jul 30, 2012 12:42 pm Post subject: Re: Using LEDs in Series

 	  Quote: 	
		  Do you have a magical circuit that somehow doesn't generate RF interference and also doesn't dissipate very much power? Thanks. Henador Titzoff 	




If I did, I wouldn't be posting here. I'd have my jet parked on my private Caribbean Island. But I'd invite everybody over for a Margarita. 



Seriously, there are two kinds of power supplies...Digital and Analog. ALL the digital stuff makes noise, ALL the analog stuff dissipates power. Is there some magical in-between*? Probably not, but the digital stuff can be made arbitrarily close to noise-free with really good design. This usually involves knowing what frequency spectrum you can use for switching...something that won't interfere.



(*Filters are almost all analog...but here I mean the active regulation element)

_________________
Eric M. Jones
www.PerihelionDesign.com
113 Brentwood Drive
Southbridge, MA 01550
(508) 764-2072
emjones(at)charter.net

henador_titzoff(at)yahoo.
Guest

Posted: Mon Jul 30, 2012 6:58 pm Post subject: Using LEDs in Series

Therein lies another problem, Eric. For a switching regulator (digital) that is "noise free with really good design," you say that it involves knowing what frequency spectrum you can use for switching ... something that won't interfere.  As diverse as our airplanes are, with several systems and with ubiquitous installation techniques and paths, how does one design a noise free switching regulator?  The only way I can think of is to design it with passive filtering and packaging methods such that radiated and conducted EMI is reduced to something extremely low. This would also involve extensive testing to verify the low EMI. That kind of design and testing would cost a pretty peso.



Seems to me the safest way to go is an active linear regulator with a sufficient back plate to dissipate the heat.



Henador Titzoff



--- On  Mon, 7/30/12, Eric M. Jones <emjones(at)charter.net> wrote:

 	  Quote: 	
		  

From: Eric M. Jones <emjones(at)charter.net>

Subject: Re: Using LEDs in Series

To: aeroelectric-list(at)matronics.com

Date: Monday, July 30, 2012, 1:42 PM



--> AeroElectric-List message posted by: "Eric M. Jones" <[url=/mc/compose?to=emjones(at)charter.net]emjones(at)charter.net[/url]>

 	  Quote: 	
		   Do you have a magical circuit that somehow doesn't generate RF interference and also doesn't dissipate very much power? Thanks. Henador Titzoff


	




If I did, I wouldn't be posting here. I'd have my jet parked on my private Caribbean Island. But I'd invite everybody over for a Margarita. 



Seriously, there are two kinds of power supplies...Digital and Analog. ALL the  digital stuff makes noise, ALL the analog stuff dissipates power. Is there some magical in-between*? Probably not, but the digital stuff can be made arbitrarily close to noise-free with really good design. This usually involves knowing what frequency spectrum you can use for switching...something that won't interfere.



(*Filters are almost all analog...but here I mean the active regulation element)



--------

Eric M. Jones

www.PerihelionDesign.com

113 Brentwood Drive

Southbridge, MA 01550

(508) 764-2072

emjones(at)charter.net





Read this topic online here:



http://forums.matronics.com/viewtopic.php?psp; --> http://www.matronics.com/Navigator?AeroElectric-Lis   --> http://www.matronics===================





	


 [quote][b]

Eric M. Jones

Joined: 10 Jan 2006
Posts: 565
Location: Massachusetts

Posted: Tue Jul 31, 2012 5:59 am Post subject: Re: Switching Power SUpplies

 	  Quote: 	
		  ... For a switching regulator (digital) that is "noise free with really good design," you say that it involves knowing what frequency spectrum you can use for switching ... something that won't interfere. As diverse as our airplanes are, with several systems and with ubiquitous installation techniques and paths, how does one design a noise free switching regulator? The only way I can think of is to design it with passive filtering and packaging methods such that radiated and conducted EMI is reduced to something extremely low. This would also involve extensive testing to verify the low EMI. That kind of design and testing would cost a pretty peso. 



Seems to me the safest way to go is an active linear regulator with a sufficient back plate to dissipate the heat.  Henador Titzoff 	




Henador, 



Yes, yes and yes. But this is drifting off the AeroElectric List interest level, so let's not post here.



I carefully removed the potting from the Buck-Puck that Bob N. designed the filter for and discovered that it was (IMHO) a very poor design. It could have been much quieter with 1) A clock line that didn't wander around the board like a drunken sailor and 2) a four-layer board which would have cost a dime more.  Several of the semiconductor companies (like National and TI) have design programs where you just punch in the parameters and they give you a well-designed PCB and parts list. Also Vicor, Astrodyne and other higher-end companies have good switcher designs that are used in aircraft all the time.



I once had a 175W switcher design tested by an RFI/EMI lab and the test guy said he thought his equipment had broken because he couldn't tell if the unit was ON or OFF by looking at his spectrum analyzer. High praise indeed! It was remarkable, quite simple  and used virtually no line filtering. The designer was brilliant...but he said he didn't really like designing power supplies and went to work for a specialty custom IC house.



do not archive

_________________
Eric M. Jones
www.PerihelionDesign.com
113 Brentwood Drive
Southbridge, MA 01550
(508) 764-2072
emjones(at)charter.net

dfritzj(at)yahoo.com
Guest

Posted: Tue Jul 31, 2012 9:31 am Post subject: Using LEDs in Series

Thanks for the in-depth answer, Eric. It sounds like I should stick to a more classic approach. I didn't like the idea of a whole set of lights going out for one failure anyway (too many bad Christmas-tree memories...)



Dan

nuckolls.bob(at)aeroelect
Guest

Posted: Wed Aug 01, 2012 5:27 am Post subject: Using LEDs in Series

At 09:58 PM 7/30/2012, you wrote:

  	  Quote: 	
		  Therein lies another problem, Eric. For a switching regulator (digital) that is "noise free with really good design," you say that it involves knowing what frequency spectrum you can use for switching ... something that won't interfere. 	


   Switching regulators are ubiquitous and definitely

   here to stay. ALL of the accessories that trumpet

   10-30 volt supply inputs are fitted with such

   power supplies.



   Operating frequency for such supplies is never

   'selected' to be compatible with the operating

   environment. It's selected as a trade-off for

   for the components selected to achieve maximum

   efficiency at rated load.



   After design goals for efficiency are achieved, then

   careful board layout for controlling aperture

   area (radiation concerns) and good management

   of RF grounds (conduction concerns) and

   dv/dt limiting (both concerns) will minimize

   any necessary filtering and shielding to

   meet DO-160 emissions limits.



   Anyone who has done this exercise a couple

   of times will never again be surprised in the lab

   with failure of a new design to meet emissions

   limits. It's like learning to bake cakes.

   After my first few passes through the vibration

   immunity testing, I never again shook a component

   of a board . . . it's that 'learning curve' thingy.



   The filter I crafted for the BuckPucks was

   a brute force endeavor to fix a problem in

   systems that were already under construction

   and parts were on hand. The filter may be

   more than was really necessary . . . but it

   proved to be adequate to the task. A detailed

   study in the RFI lab might show that a simpler

   filter would do . . . but given the small

   volume of sales for this particular LED drive

   solution, it hardly justifies the expense of

   a finely tuned development effort.



   The BuckPuck was never intended for use in

   an aircraft EMC environment else it would not

   have been built that way. Some of the HID lamp

   supplies I've played with are similarly burdened

   with EMC issues for aircraft. That doesn't mean

   that any of these devices were "poorly designed",

   it only means that the customer base for thousands

   of units don't care that the device going into their

   boat, motorcycle or hot-rod is unsuited to airplanes.



   In the OBAM aviation world, we are a primo filter

   for searching out and identifying the best that

   folks know how to do. But it's a fine demonstration

   of the need for caveat emptor . . . give it a try

   in low risk experiments and report findings to

   the group. If a short-fall in performance is

   noted, then try something else . . . or modify

   the product (such as the filter board for the

   BuckPucks). In any case, the ultimate goodness

   of the device and the system in fits is up to

   us and NOT the designer of the device who probably

   never envisioned his work product going into

   an airplane . . . and would probably rather that

   it did not.



   It used to be that a phone call to Gates about

   v-belts had to be made in 'disguise' . . . if

   you mentioned that you were working on an airplane,

   the standard vociferous response was "our products

   are not intended for use on aircraft" . . . i.e.

   "you're on your own my friend". 



 

   Bob . . .    [quote][b]

nuckolls.bob(at)aeroelect
Guest

Posted: Wed Aug 01, 2012 11:30 am Post subject: Using LEDs in Series

I have been experimenting with a product available

 off eBay that offers 300, .05 x .05 LED's mounted

 to a 3/8" wide x 16 foot flexible etched circuit

 strip.



  http://tinyurl.com/c33w25o



 The leds are wired in paralleled series-clusters of

 3 devices and a 150 ohm resistor. Here's a 2-cluster

 segment of this product . . .



 [img]cid:.0[/img]



 You can cut the strip every three devices and attach new

 power leads at the solder pads visible in the photo.



 I'm going to be using about 80 feet of this product in

 my kitchen to put double rows of ceiling accent lighting high up,

 double rows to replace recessed canister lights over the

 counters and a third twin-row installation under the cabinets

 for work lighting. The double-rows will be wired in series to run

 off 24 volt supplies in the walls. My present lighting

 is fraught with hot-spots from 11 different fixtures with

 no accent lighting on a budget of about 700 watts. The working

 end of my kitchen will be nicely lighted for both appearance

 and working illumination on a power budget of about 175

 watts total. This technique produces very wide-angle

 illumination that is virtually shadow free.



 The strips are fitted with peel and stick adhesive. I'm

 going to use 3/4" wide strips of Formica to assemble one-

 piece lighting assemblies that make it easy to get the

 rows installed straight while laying on my back under

 a cabinet.



 The light output from these devices is very good. I suspect

 that one could put a single strip of this product on the

 under side of a glare shield and get more than enough

 light . . . with devices easily dimmed. Due to the

 series wiring of three, 3-volt devices, you would want

 to control dimming with a series resistance or an

 adjustable current supply like a Buck-Puck. Ten of

 these clusters would give you about 30 leds and a max

 bright current draw of 300 mA. A constant voltage

 dimmer like B&C sells COULD be modified to have minimum

 brightness occur at 9 volts instead of the present

 4 volts. In any case, This product offers a host of

 illumination opportunities not the least of which is

 your airplane.



 

   Bob . . .

1839fa0.jpg

Description:

Filesize:

347.67 KB

Viewed:

7889 Time(s)

mapratherid(at)gmail.com
Guest

Posted: Wed Aug 01, 2012 5:14 pm Post subject: Using LEDs in Series

This looks pretty cool Bob. But I'm picky about color temp. Any idea what Kelvin color you'll end up with?





Regards,

Matt-





On Aug 1, 2012, at 1:29 PM, "Robert L. Nuckolls, III" <nuckolls.bob(at)aeroelectric.com (nuckolls.bob(at)aeroelectric.com)> wrote:



[quote]  I have been experimenting with a product available

 off eBay that offers 300, .05 x .05 LED's mounted

 to a 3/8" wide x 16 foot flexible etched circuit

 strip.



  http://tinyurl.com/c33w25o



 The leds are wired in paralleled series-clusters of

 3 devices and a 150 ohm resistor. Here's a 2-cluster

 segment of this product . . .



 <1839fa0.jpg>



 You can cut the strip every three devices and attach new

 power leads at the solder pads visible in the photo.



 I'm going to be using about 80 feet of this product in

 my kitchen to put double rows of ceiling accent lighting high up,

 double rows to replace recessed canister lights over the

 counters and a third twin-row installation under the cabinets

 for work lighting. The double-rows will be wired in series to run

 off 24 volt supplies in the walls. My present lighting

 is fraught with hot-spots from 11 different fixtures with

 no accent lighting on a budget of about 700 watts. The working

 end of my kitchen will be nicely lighted for both appearance

 and working illumination on a power budget of about 175

 watts total. This technique produces very wide-angle

 illumination that is virtually shadow free.



 The strips are fitted with peel and stick adhesive. I'm

 going to use 3/4" wide strips of Formica to assemble one-

 piece lighting assemblies that make it easy to get the

 rows installed straight while laying on my back under

 a cabinet.



 The light output from these devices is very good. I suspect

 that one could put a single strip of this product on the

 under side of a glare shield and get more than enough

 light . . . with devices easily dimmed. Due to the

 series wiring of three, 3-volt devices, you would want

 to control dimming with a series resistance or an

 adjustable current supply like a Buck-Puck. Ten of

 these clusters would give you about 30 leds and a max

 bright current draw of 300 mA. A constant voltage

 dimmer like B&C sells COULD be modified to have minimum

 brightness occur at 9 volts instead of the present

 4 volts. In any case, This product offers a host of

 illumination opportunities not the least of which is

 your airplane.



 

   Bob . . .  

[b]

nuckolls.bob(at)aeroelect
Guest

Posted: Wed Aug 01, 2012 6:44 pm Post subject: Using LEDs in Series

At 08:13 PM 8/1/2012, you wrote:



 	  Quote: 	
		  This looks pretty cool Bob. But I'm picky about color temp. Any idea 

what Kelvin color you'll end up with?


	


  You can select for cool, pure, or warm white colors

  when you order. Given the cheapness of excess

  brightness for you can probably do some shading with

  theater lighting gels.



  I had to match some lighting colors in incandescent

  lamps about 25 years ago at Videmation. A local

  theater supply store had every color hue and density

  imaginable in gels for about $2 a sheet. We fiddled

  with the filters until we got it right.



  In my kitchen I'll be using the pure whites and

  letting colors reflected from surfaces set the

  mood. At Cessna, blue-white post lighting was a

  high dollar option for panel illumination. You

  could get the same thing by putting the appropriate

  gel over the lamps.



   Bob . . .

Display posts from previous:

	Matronics Email Lists Forum Index -> AeroElectric-List	All times are GMT - 8 Hours
Page 1 of 1

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