# LED flasher circuit resistor question

Discussion in 'The Projects Forum' started by danleigh57, May 6, 2010.

1. ### danleigh57 Thread Starter Member

Apr 21, 2010
31
0
I am working on ordering some parts to finish breadboarding this circuit.
The actual arrays I plan to use are approximated in the box at the right.
The part I am concerned with is R10 and it's function. Using the formula given with the schematic - R = .06/I a 33ohm resistor is correct for the red piranha LEDs I am using. If I run the array through a calculator, given 2.8 volt drop and .020 amps I get a value between 100 and 150 ohms given the variables on a 12 volt system.

If any of that makes sense, what is the function of R10? Does it take the place of the normal series resistor? Is it in addition to those resistors? If it is supposed to take the place of them, having looked at the "LEDs, 555s,..." post this doesn't seem like a functional solution.

Also I would appreciate any additional input on this circuit's strengths/faults. And lets just consider this a bench project for now.

2. ### Wendy Moderator

Mar 24, 2008
21,838
3,047
What you are looking at is a constant current source. R10 is a emitter resistor for Q2, the voltage imposed on it sets up a constant current through the collector of Q2. A few more parts than I like, but the design seems sound.

I'm not quite sure what Q3 does. The UJT Q1 seems kinda odd too, since the 555 can make a very decent oscillator all by itself.

3. ### danleigh57 Thread Starter Member

Apr 21, 2010
31
0
So does R10 eliminate the need for individual resistors on the legs of the array?

Is it possible to do away with Q1? (That's the part I forgot to order.)

4. ### Wendy Moderator

Mar 24, 2008
21,838
3,047

The circuit is similar to this.

The emitter resistor is carrying a light load, the transistor is taking the brunt of the wattage. Since there are a lot of heat sinks for transistors (unlike resistors) it makes for a easier circuit, and more stable overall.

What does the UJT do exactly? Vary the speed of the 555 oscillator? This can be done several different ways.

Last edited: May 6, 2010
5. ### danleigh57 Thread Starter Member

Apr 21, 2010
31
0
Here's the rest of the page the post came from. I am learning by doing, so as these questions come up I search around to try to find the answers. On the third or fourth cycle the light usually comes on. Now it's 555's, JFETs and Schmitt triggers. There's an awful lot to absorb.

Parts:

R1,R5____________1K 1/4W Resistors
R2______________33K 1/4W Resistor
R3_____________100K 1/2W Trimmer Cermet
R4_______________1M 1/4W Resistor
R6,R7,R8________10K 1/4W Resistors
R9_______________4K7 1/4W Resistor
R10_____________33R 1/4W Resistor (See Notes)

C1_______________1µF 63V Polyester Capacitor
C2_____________100nF 63V Polyester Capacitor
C3_____________470µF 25V Electrolytic Capacitor

D1______________LED (Any dimension, shape or color)
D2,D3________1N4002 100V 1A Diodes
D4--D15________LEDs High brightness, high efficiency red types (See Notes)

Q1___________2N3819 General-purpose N-Channel FET
Q2____________BC337 45V 800mA NPN Transistor
Q3____________BC547 45V 100mA NPN Transistor

IC1____________7555 or TS555CN CMos Timer IC
IC2,IC3________4094 8-stage shift-and-store bus register IC

SW1____________SPST Brake/Turn Lights Switch

This circuit, designed on request, allows up to 16 LED clusters illuminate in bar-mode sequence. LED sequencing will start when the Brake pedal is pressed (or Turn Lights are switched-on) and, when the last cluster illuminates, all LED clusters will remain steady on until the Brake pedal is released or Turn Lights are switched-off.
Notes:

• R5 and D1 are optional: they could be of some utility in monitoring the sequence frequency set by means of R3.
• Clusters can be formed by up to 12 LEDs as shown in the circuit diagram, right side. Common cluster types usually range from 5 to 10 LEDs.
• Up to 16 of these cluster driver circuits, each formed by the LEDs, two transistors and three resistors can be built and connected to the progressively numbered outputs of IC2 (the first eight clusters) and IC3 (the remaining clusters).
• If a number of clusters up to 8 is required, IC3 can be omitted.
• Constant output current value for the LEDs can be changed by varying R10.
The formula is: R = 0.6/I (I expressed in Amperes).
• Use high brightness, high efficiency red LED types of suitable size and change R10 value to suit LED's Absolute Maximum Ratings.
• The main circuit, formed by the three ICs and related components is permanently wired to the supply (and D1 will flash) but its 12V positive supply can be wired after the main switch of the car, or an on/off switch can be added.

6. ### danleigh57 Thread Starter Member

Apr 21, 2010
31
0
Is it possible that Q1 is a trigger to take any delay out of the initiation of the circuit?

If I am understanding my reading correctly, this could be my circuit.

I set the output timing by R1, (R2 + R3),and C2, which triggers IC2 to cascade and lock its outputs. Next step is to try to understand the 4094's function. I will only be running 3 arrays and will need an off period equal to 1/4 of the cycle.

Since the final circuit will have a period of about a second, perhaps Q1 is required?

Last edited: May 7, 2010
7. ### Wendy Moderator

Mar 24, 2008
21,838
3,047
It is possible. I'd do it differently, but one of the fun things about electronics is there is usually more than one good way to do something.

I've been thinking of trying a different idea to eliminate the startup delay.

I figure the base voltage should be 1/3Vcc + 0.5V, and select the resistors accordingly. I've not built or tested this idea yet.

• ###### temp.GIF
File size:
1.9 KB
Views:
164
Last edited: May 7, 2010
8. ### danleigh57 Thread Starter Member

Apr 21, 2010
31
0
I threw together the circuit I drew, and failed miserably. So I pulled out the 4094 and all related, and when that didn't work I backed up again and tried to breadboard a rough version of your long version flasher circuit, using what I had on hand with some calculations from another 555 tutorial. (I didn't like natural logs the first time around...)

I threw out the trimmer, used a1K resistors for R1 and R2 and the .1uf capacitor I had for the circuit. My math said this would give me a period of just over .2 seconds. The LED lights, very dimly, but it doesn't flash. I'm thinking maybe my math is wrong and I've pt together a pwm circuit.

The thing that is really throwing me is that I am running 12V to the board But as soon as u plug it in (as you show, with + on one runner, - one the other side and the positive jacked across), I get 1.96 volts. Various measurements around the board give me values of around 1 or around 2V.

Since I absolutely don't understand what is causing this, I don't have a prayer of figuring out the rest of it.

9. ### Wendy Moderator

Mar 24, 2008
21,838
3,047
OK, you have R1 as 1KΩ, R2/R3 as 133KΩ, C2 is 1µF. The flash rate is F=1.44/(1E-6(1KΩ+2*133KΩ))=5 Hz.

This works, fast but visible. If D1 is not flashing there is likely a wiring error. You are using this schematic, right?

Double check your pot setting, I'm assuming it is maxed out. Around 10 Hz it gets hard to see, your design has a max of 21 Hz. If you are using a protoboard try taking a digital picture and attach it.

*****************

As a side thought, try upping C1 to 10µF.

Last edited: May 8, 2010
10. ### SgtWookie Expert

Jul 17, 2007
22,201
1,809
This is why we stopped helping people who wanted to do projects regarding automotive lighting.

Vstar 650, eh? Not bad. You should've saved your money and bought a Harley though.

11. ### Ron H AAC Fanatic!

Apr 14, 2005
7,012
681
The Vbe of Q3 is the reference for the current sink. Ie2=Vbe3/R10. Q2 and Q3 form a feedback circuit.
The value of the collector current of Q3 is a second order effect, as it only affects Vbe3 slightly. R9 is redundant.
Q1, as you may realize by now, is a JFET. It is on whenever SW1 is closed, and it puts C1 in parallel with C2. I don't understand the function of the 555 oscillator when SW1 is open. The shift register outputs are disabled, so the LEDs will not light. Maybe it's a test mode?
Unless the LEDs are matched, each leg should have a ballast resistor in series with it, which will tend to equalize the brightness of the LEDs.

EDIT: Disclaimer of Responsibility:
With respect to the above information, neither myself nor Barack Obama, makes any warranty, express or implied, including the warranties of merchantability and fitness for a particular purpose, or assumes any legal liability or responsibility for the accuracy, completeness, or usefulness of any information, apparatus, product, or process disclosed, or represents that its use would not infringe privately owned rights.
Having said that, if you crash and burn as a result of the tail lamps not working, not to worry about your hospital bill. President Obama will have your butt covered. That is, as long as you have purchased insurance. If you haven't, you will be fined.

Last edited: May 8, 2010
12. ### SgtWookie Expert

Jul 17, 2007
22,201
1,809
Ron_H,
This poster is trying to make a circuit for tail lamps on a motorcycle.

Please include the standard disclaimers, as the OP's survivors may try to sue you for his inability to make the circuit work.

13. ### Ron H AAC Fanatic!

Apr 14, 2005
7,012
681
Thanks for the reminder. See edit above.

14. ### danleigh57 Thread Starter Member

Apr 21, 2010
31
0
I tossed the pot. Looks like I have a math error. I'll check using your formula once I get some coffee in me. I think I will also take your advice and slow the blink rate down.

Can you tell me why the voltage drop on the board? That confuses the heck out of me.

Also what is the technique for quoting other posts - copy and paste didn't work.

Sgt Wookie- You stopped helping people with these circuits because they get confused?

Riding a Roadie now. I didn't have enough time to save to buy a Harley...

15. ### retched AAC Fanatic!

Dec 5, 2009
5,197
315
I think this is why we stopped.

No matter how hard me try to get points across, people dont listen.

A breaklight failing on a car is a fender bender, on a motorcycle it can be a funeral.

Just because you see light coming from your leds, and hook them up, doesn't mean that after 3 uses the LEDs wont burn up.

Cheap LEDs and people not understanding current limiting can end up with NO break lights at 10pm on a Saturday night.

Not wiring them in parallel so others will work WHEN one fails. Is also a problem.
If one fails, the voltage that the rows are designed for just jumped up causing increased current to the rest of the LEDs and continual POPing.

Plus getting yelled at because someone got a ticket the first day with there "NEW LED BLINKERS or TAIL LIGHTS" because they failed is not as bad as a court appearance because that person FAILED to survive the collision.

Still, both suck.

You want to be sure, that your LEDs will work in 'running mode dim' and 'breaking bright' for long periods of time with out over heating and/or burning out.

So AAC didnt stop helping people with vehicle alterations because they got confused, but because they would still make the changes WHILE confused.

Not figuring in water resistance, vibration, heat, voltage spikes, etc. Vehicle are VERY difficult to work with. If you design something on your bench to work with 12v, put it in a vehicle, and It dies because you hit a bump, the engine heat killed it, or the 60+ Volt spikes nailed it when it wasnt properly designed to handle it. And vehicles are rarely at 12v. The alternators are dumping 14+ volts into the system while the vehicle is running. If you are holding the break while starting the vehicle, when you disengage the starter, a huge spike will SLAM into the LED circuit. Same with blinkers blinking and turning headlight(s) on and off.

Vehicles typically use a chassis ground. If the circuit you design fails and dumps power to ground, the 0v reference to the break by wire, or throttle by wire, etc, can be affected.

SgtWookie is a kick-ass guy, and he enjoys All About Circuits as a pastime. He is doing what he has been trained to do. And that is to protect what he holds dear.

I would be upset if a circuit someone built, of my or anyones design, hurt or killed someone and caused this place to be sued and shut down, and someone fined and/or jailed.

Remember it isn't just you as a designer or mechanic. It is also the manufactures. Look at Toyota. Imagine if an LED light system dumped juice to ground causing rapid unintended acceleration.

Thats Toyota designers who overlooked a possibility, but our actions that could kick it in.

So, ask questions and be safe. And remember that what you think of as 'SAFE' changes with what you learn. You may have had no idea what the breaklight circuit could be receiving in terms of voltages. Now you do. So before, you would have thought you were building for a safe 15v just to find out after some learning, you need much more to be safe.

16. ### danleigh57 Thread Starter Member

Apr 21, 2010
31
0
As I've said before, I appreciate your concerns. I also applaud you for your efforts here, and for your cautioning persons to be careful. Truly I do.

Again, these lights, if I get them to work, will be used in addition to the lights I already have on the bike to increase visibility and safety. Your comments have already caused me to notice that automotive standards for components are stated on at least some data sheets, and to put more thought into how components can be protected and cooled by placement on the vehicle. I get virtually no tickets and do not plan to start now, and am not inclined to blame other people for my actions. Oh, and I tend to run my grounds back to the vehicle wiring. Safety is incredibly important on two wheels, and I keep it in mind with all the modifications I do.

This has proved to be an intellectually stimulating endeavor, perhaps the most learning I have done in years, and the assistance provided by this forum has been invaluable. It may never get off the bench, and that will be fine, if not optimal.

Taking into account the imprecision of internet communication, I should have used a smiley in my communication with SgtWookie. In addition to requesting clarification, I was poking gentle fun, as I assume he was doing when he questioned my choice of motorcycle.

17. ### retched AAC Fanatic!

Dec 5, 2009
5,197
315
Good. Its nice to see you have a dollar worth of sense about you.

The drastic voltage drop is either a bad power supply, or It requires a load before it starts.

But of you measure 12v before the bread board and <2v after, it seems like its bad.

Is there a voltage divider circuit it is going through?

As for quoting posts, there is a Quote button on each and every post at the lower right.
Clicking that will start a response post for you with the quoted post inside, ready to go.

If you want to grab a certain part or line of a post, highlight it, and click copy, or press [CTRL-C] on the keyboard after the selection is highlighted.

In your post, click the button in your reply and press [CTRL-V] or PASTE to paste the quote into the (QUOTE) (/QUOTE) tags. (Replace the '()' with '[]')

Last edited: May 8, 2010
18. ### danleigh57 Thread Starter Member

Apr 21, 2010
31
0
I have used a wall charger rated at 12 volts but putting out roughly 16V. I ganged a couple of resistors and tapped into the middle (a voltage divider I assume?) as I had no other immediate fix. The wires I am plugging into the proto board read 11.89 to 12.02 volts.

I guessed this was not a perfect solution, but it seemed to work and it was interesting. This circuit is on another board and then run to this one.

19. ### retched AAC Fanatic!

Dec 5, 2009
5,197
315

No good.

The wall wart is an unregulated power supply. You can and should add a LM7812 Voltage regulator. This will turn your wall wart power into a regulated power supply.

You will want to use this on your circuit anyway you go with it.

20. ### danleigh57 Thread Starter Member

Apr 21, 2010
31
0
Are you saying that this won't work on this prototype? I believe both the 555 and 4094 are rated to 15V.

I imagine that just for the 555 timer I could use the 7805 I have, I just was trying to stay consistent.

And yes the 7812 is on the list as soon as i figure out what else I need to order. The electronic components are affordable, bu the freight is killing me.