Ping....... electrical engineers or just really smart people

[beondwacko]

Hi again guys/ladies,

Does anyone in here have any electronic engineering experience? I went to school a long ass time ago for Audio engineering and had to take a bunch of electronic theory classes , but forgot most of what I've learned since then. If there is any lurkers who know their stuff, PM me please. I've got a few questions RE: project LED tail light with paralell wiring and resistors. THanks in advance.

Also , if anyone here on the board knows of a member who isn't too active but knows his stuff could you forward his name so I can contact them ?

Adam

[Hobicus]

Just for the record: You need a giant signature. That one's just not big enough.

PM bartonmd His degree is in electrical engineering.

[beondwacko]

Sorry about the sig. I'll resize tonight when I get home.

[SwampNut]

Why don't you just post the question or diagrams, you'll get an answer.

[Amherst XX]

I know quite a bit about electronics theory and my room-mate is an electronics tech. Shoot out the question...I'm sure we can help. I'm actually making my own LED taillight assy this winter too

[bartonmd]

I'm an Electrical Engineer (Technology)... Post up or PM me... (preferably post up)

Mike

BTW,

Standard LED's want ~20mA of current each (your datasheet will tell you)

Wire them in parallel (all to one positive and all to one negative)

14.3V/20mA = 715 ohms

[Skull]

Like everybody else said, post up. We've got a lot of folks around here that can help on a great variety of topics.

[Skull]

Standard LED's want ~20mA of current each (your datasheet will tell you)

Wire them in parallel (all to one positive and all to one negative)

14.3V/20mA = 715 ohms

Gotta be careful with that. Many LEDs only make a ~1 V drop across them naturally, so unless they have a built-in step-down resistance, run at least one 1kΩ resistor in series with them.

[Animal Mother]

Like everybody else said, post up.  We've got a lot of folks around here that can help on a great variety of topics.

And there are also a lot more who would like to learn something new!

[Suf Daddy]

Like someone said, LED's have a low voltage drop, so most blinkers need a load capacitor or gizmo to make them bulbs "blink" or not set off the electronic car monitoring system that tells you your bulbs out, headlight out etc.........

Otherwise those chimes or symbols will wear on you.

-Suf Daddy

[beondwacko]

I didn't want to clog the pub with techno gibber jabber. I figured it was better to correspond via PM and save all the others , but now on the other hand , it might be cool to have others put their spin on this !

Rossodave asked first if he could offer any help. This is what I wrote to him.

Dave,

YOU'RE A LIFE SAVER !!!!

I'm going to be building a couple of boards with LED's as an insert into a

clear 99'-04' XX tail light. My first question is that I need a formula to come up with the resistance values for the LED's listed just below.

Product Description

Emitted Color : ULTRA RED

Size (mm) : 5mm

Lens Color : Water Clear

Peak Wave Length (nm) : 640 ~ 645

Forward Voltage (V) : 1.8 ~ 2.2

Reverse Current (uA) : <=30

Luminous Intensity Typ Iv (mcd) : Average in 5000

Life Rating : 100,000 Hours

Viewing Angle : 20 ~ 25 Degree

Absolute Maximum Ratings (Ta=25°C)

Max Power Dissipation : 80mw

Max Continuous Forward Current : 30mA

Max Peak Forward Current : 75mA

Reverse Voltage : 5~6V

Lead Soldering Temperature : 240°C (<5Sec)

Operating Temperature Range : -25°C ~ +85°C

Preservative Temperature Range : -30°C ~ +100°C

I'm planning in wiring them in paralell probably in groups of 10, but possibly more. I know that you need to use a resistor to limit the current, and these all come with resistors. The question is I need to know with an input voltage of 12.5v - 13.8v and 10 led's wired in paralell what value resistors I would need. I assume that the same formula could be use to figure more or less LED's in the circuit as well.

Now , below here are the other version of LED's that I'll be installing for the turn signal indicators ( seperate circuit ). They seem to have the same requirements for power.

Product Description

Emitted Color : ORANGE

Size (mm) : 8mm

Lens Color : Water Clear

Peak Wave Length (nm) : 615 ~ 620

Forward Voltage (V) : 1.8 ~ 2.2

Reverse Current (uA) : <=30

Luminous Intensity Typ Iv (mcd) : Average in 5000

Life Rating : 100,000 Hours

Viewing Angle : 20 ~ 25 Degree

Absolute Maximum Ratings (Ta=25°C)

Max Power Dissipation : 80mw

Max Continuous Forward Current : 30mA

Max Peak Forward Current : 75mA

Reverse Voltage : 5~6V

Lead Soldering Temperature : 240°C (<5Sec)

Operating Temperature Range : -25°C ~ +85°C

Preservative Temperature Range : -30°C ~ +100°C

Dave , I'm also looking into building a seperate control board that will function as a brake light modulator. What I'm looking for it to do is to for the first 2 seconds of brake light circuit activation, I'd like it to pulse for 250ms on / 250ms off......then after the first two seconds of pulsing , go to full on. This would be a nice add on , but it won't break the design if it's not incorporated into the final result.

Now , I'm not sure if I am to add up all the led's current draw to come up with a total, or if there is another formula to figure that one out. The only reason I would think that I would need to do that is for the Modulator circuit. 3.6 amps total draw based on 120 led's with a individual draw of 30ma just doesn't sound right. The load sounds high for just led's.

If you have the patience for answering these and a few more questions that I'll have along the way , it would be really great. I'll of course post a ton of pics and drop names at the end of all the folks who deserve praise for advising and helping me.

BTW, the design plan is as follows.

Gut a factory housing of reflector ( you can seperate the lens from the housing by putting it into nearly boiling water ) and remove the reflector assembly. I sourced out 2 different types of circuit boards at radio shack both of different sizes, 1 has multipbe busses , the other is plain w/780 mounting holes. I'm going to have to measure up which one will work out better. I'll plan a division of the led's for running light and brake light probably a 30/70 split. Then along the left and right sides of the boards, there will be 2 vertical rows of amber leds ( 10 per side ) that will function as indicators. I'm also playing with the idea of having a symbol of sorts for the running light. Maybe an X in the lower portion of the tail light. How does all this sound? Do-able easily in it's construction , but I need a little help with the planning.

Thanks Dave. Sorry this ran so long

This make any sense yet ????

[beondwacko]

bartonmd wrote:

Standard LED's want ~20mA of current each (your datasheet will tell you)

Wire them in parallel (all to one positive and all to one negative)

14.3V/20mA = 715 ohms

So , I'm going to be using 30ma peices, and at 70 units on one circuit,

14.3/30ma = 476 ohms ? Now is that per unit , or how does that apply to paraell as well ?

[XXL-XX]

bartonmd wrote:  

Standard LED's want ~20mA of current each (your datasheet will tell you)  

Wire them in parallel (all to one positive and all to one negative)  

14.3V/20mA = 715 ohms  

So , I'm going to be using 30ma peices, and at 70 units on one circuit,  

14.3/30ma = 476 ohms ? Now is that per unit , or how does that apply to paraell as well ?

Above is a circuit diagram for each bank of 10 LEDs from the taillight specs. remember to repeat the ciruit for each 10 LED bank. ADD up the number of 10-LED banks multiply by 300ma multiply by 1.5 to get the fuse value to use from the battery terminal.

[Skull]

[snipped excellent circuit diagram]

Above is a circuit diagram for each bank of 10 LEDs from the taillight specs. remember to repeat the ciruit for each 10 LED bank. ADD up the number of 10-LED banks multiply by 300ma multiply by 1.5 to get the fuse value to use from the battery terminal.

This is exactly right.

One thing to keep in mind, though, is that the 30mA values listed in the datasheet posted are MAX values. You probably don't want to run them quite at the ragged edge like that. Backing off ~10%-20% in the current value most likely won't be significantly less bright, and might increase the life of the LEDs quite a bit.

[rossodave]

You don't typically want to run an LED at its maximum rating. I would run them at 15ma each or less. This should give you good light intensity without sacrificing the life of the LED's. I would try a common (easy to get value) of 100 ohms to start with. If you want to see how bright this is just hook up one LED with a 1K resistor. If that is not bright enough step down to another value of like 750 ohms. When you reach the intensity that you would like with a single LED just divide that resistor value by ten and you will have the value for your group of ten. I would definitely keep in mind though, that you do not want to run them above 20ma each. Running higher than this will give you shorter than desireable LED life.

[Helvet]

Sounds interesting. Do you have an idea about a design, or will you just full up the whole space in the reflector?

[beondwacko]

Helvet ,

I'd like to fill the entire space in the reflector area with minimal clearances between the LED's. Really pack em' in there tight if I can. And for the running light in the lower section , some how draw up a double X for a pattern.

Rossodave,

I like your idea of not running them at full intensity. Possibly 2/3 or 20ma will give a decent result especially because they will be grouped on the board so tightly.

XXL ,

Isn't there a law or formula that states when you have common static loads connected in paralell that the resistance drops with the addition of more devices connected across the whole circuit ?? Also , Duh....................

I swear , if these were just light bulbs , something in my head thinks it would be more simple, but with the LED's needing to have both voltage and amperage limited, it's just different.

[beondwacko]

I also came to the conclusion that the larger ( same electrical specs ) amber LED's that are going to be used for the turn signals will be ran at 30ma ( full power ). The reason I'm doing this is because they will only be on intermitantly, and with them not staying on full time, they should be just fine.

[Skull]

I also came to the conclusion that the larger ( same electrical specs ) amber LED's that are going to be used for the turn signals will be ran at 30ma ( full power ). The reason I'm doing this is because they will only be on intermitantly, and with them not staying on full time, they should be just fine.

Seriously, man, test them out first to see if you really need to run them at full power. It's entirely likely that they achieve full brightness significantly lower than max current.

[XXL-XX]

XXL ,

Isn't there a law or formula that states when you have common static loads connected in paralell that the resistance drops with the addition of more devices connected across the whole circuit ?? Also , Duh....................

I swear , if these were just light bulbs , something in my head thinks it would be more simple, but with the LED's needing to have both voltage and amperage limited, it's just different.

The voltage drop across a semi-conductor like an LED is determine by the PN junction and its dopping level, it's design. By design the voltage drop will always be 1.8 ~ 2.2 as specified and will not deviate. This is why semi-conductors are so reliable. All the diodes in parallel will draw approximately the same current from the load providing that they are soldered properly. These jumbo LEDs will have no trouble carrying 30mA contineously for years.

[beondwacko]

All the diodes in parallel will draw approximately the same current from the load providing that they are soldered properly. These jumbo LEDs will have no trouble carrying 30mA contineously for years.  

XXL ,

All the diodes will draw approx the same current from the load? As if there were just one or multiple connected in Paralell? That was the big question. When you connect multiple LED is Paralell , the resistance can't be the same as if you were wiring for just one. There are more paths for the current to return to the source, so as opposed to a series curcuirt, wouldn't the required reststance decrease ?? I know that it sounds like the logic is flawed, and I also know that by that reasoning because there are more loads ( work to be done ) the resistance should increase. That's where the big question lies for me. I should be getting the LED's late this week , and start construction shortly afterwards.

[XXL-XX]

Diodes have very little forward bias resistance and they do not behave like resistors. Diodes will draw a lot of current untill they burn up which is why we limit the available current with a resistor. Now, the voltage drop across the diode is determine by the design of the diode and is constant. When you connect multiple diodes in parrallel, the one with the largest voltage drop across it will regulate the voltages for the others. They will all draw approximately the same current. Providing they are closely match. Your diodes are closely match based on the specs. There are other factors involved base on the design of the diodes but you need not worry about them in this simple circuit. Remember to protect the circuit and your bike's electrical system with fuses. If you have any question on values for the fuses give me a call.

[beondwacko]

Thank you XXl,

We'll be in touch almost for sure.

[bartonmd]

Just FYI... check out this site... it does a good job of explaining how to hook them up... http://www.kpsec.freeuk.com/components/led...d.htm#calculate

Seems that series is a better way to hook LED's up than parallel, because of the slightly different voltages that they can drop...

Mike

Edit: And here... http://wolfstone.halloweenhost.com/TechBas...eLEDsInParallel

[XXL-XX]

Edit: And here...http://wolfstone.halloweenhost.com/TechBase/litlpo_PoweringLEDs.html#WiringMultipleLEDsInParallel

This is an exellent explanation and the circuit you should use.