555 astable timer to h-bridge for 2 channel christmas led lights

[Digidank]

I have a set of christmas lights that only has 2 wires. When i hook up 4.5 volts to it only half the lights light up. When I reverse the polarity then it switches and the other half light up. I am simply trying to create a circuit to keep both sets to appear lit at all times. Based on some reading I believe I just need to alternate the polarity at a frequency around 50hz.

I have been scouring the internet and trying to build the circuit with circuitlab.com's tool, but cannot seem to figure out how to take the single control square wave signal from the 555 astable timer to trigger the h-bridge properly. I have built the 555 astable timer independently and see i get a nice square wave on the control output. I then built the h-bridge independent of the 555 timer and i get it to alternate the polarity when i switch the polarity on the 2 drive points. Can someone please help me figure this out? Sorry, I am a noob with this stuff and trying to teach myself.

Attached I have my circuitlab.com circuit layouts.

555-astable.jpg shows how i configured the 555 timer and it generates a nice square wave as expected.

555-and-hbridge.jpg shows the 555 and h-bridge laid out. But, I do not know how to appropriately trigger the drive1 and drive2 from the 555's control output. I have tried dabbling with transistors but i just get very wacky voltage outputs at the LEDs.

Any help would be much appreciated! Thanks

 

[AnalogKid]

First, you can do this with nothing but a small wall-wart transformer with an AC output. You might have to add one series resistor depending on the transformer's output voltage.

Second, R12 and R13 do not need to be such a low value. But ...

What is the current value you want to push through the LED string?

The bipolar 555 output stage is rated for +/-200 mA. If the LED current is less than 150 mA, then you can do what you want with only one bipolar555.

Or, you can use two 555's to form a full h-bridge with no external transistors, by having one 555 drive the other as a power inverter.

BTW, the oscillator frequency does not have to be 50 or 60 Hz. The LEDs probably were intended to be powered by a simple power transformer with no rectification or filtering, hence the power line frequency value. But LEDs are fast; you could run them at 100 kHz with no problems other than some radiated interference with other electronic devices. So there is a large range of acceptable values. 1 kHz is a nice round number then won't flicker or cause significant interference.

Update: AND - look up the datasheet for the LMC555, a CMOS version. It has an alternate astable circuit that guarantees an almost perfect 50/50 output duty cycle. This circuit works well with the original bipolar 555, reduces parts count, makes changing the output frequency easier, etc.

ak

 

[Digidank]

Thanks for the great reply! Sorry for my ignorance, but is a bipolar 555 different than a normal 555 or is that just what it’s called? 120mA is fine so I can use a single bipolar 555’s output.

do you by chance have a diagram I could follow to create this?

 

[AnalogKid]

Original 555: NE555 by Signetics. All bipolar circuit (35 NPN and PNP transistors).
Many copies (it never was patented): MC555, LM555, etc.

CMOS 555: lower power operation, lower output currents, higher top speed, etc. Overall, better operation if you don't need the super-fat bipolar output stage. National's is the LMC555. The "50% duty cycle oscillator" circuit is on page 7 of the datasheet.

I try to whip up a schematic tomorrow.

Where are you located?

ak

 

[Digidank]

I believe I may have got it using the 2 555's like you mentioned. Could you critique my diagram here?

I added the 1k resisters on the output (pin 3s) of both because it eliminated some spikes in voltage. Now it looks like I have very close to an alternating circuit back and forth.

Btw, I live down near mickey in Orlando, FL.

 

[bertus]

Hello,

You could make the first 555 circuit simpler by using a hysteretic 555 circuit:
https://www.allaboutcircuits.com/textbook/experiments/chpt-8/555-hysteretic-oscillator/
That will have an almost 50 % duty cycle.

Bertus

 

[AnalogKid]

I believe I may have got it using the 2 555's like you mentioned.

Getting there. You need only one current limiting resistor in series with the LEDs. Also, if you increase R2 and decrease C1 you will have an output closer to 50/50.

As bertus pointed out, there is a circuit that gives almost exactly 50/50. It is the one I mentioned in post #2. I'll post it next.

 

[AnalogKid]

This is a first-pass (((concept))) schematic that shown a half-bridge drive for the LEDs. It is not as efficient as the dual-555 approach, but it is a minimum parts-count solution. Note that some of the components need adjusting; the ones shown are what is already in my design libraries. For example, R2 and R3 are the correct value for 120 mA LED current, but should be 2 W each.

The oscillator frequency is around 150 Hz, a consequence of using parts I already own. The normal value for a decoupling cap (C2) is 0.1 uF, but for a circuit with high output current I use a larger value. Almost anything is fine as long as it is not an electrolytic; ceramic and metal film have a much better frequency response.

ak

 

[bertus]

Hello,

@AnalogKid , Will there be a constant current flowing through R2 and R3, draining the battery faster?

Bertus

 

[Martaine2005]

Btw, I live down near mickey in Orlando, FL.

Is that near Mini too?

Martin

 

[AnalogKid]

Will there be a constant current flowing through R2 and R3, draining the battery faster?

Yes, but it's not as much as you might think.

Without the LEDs connected, the static current through R2 and R3 is 66 mA. But ...

The 555 output is always either high or low, so one resistor always is "shorted out" by an LED. A typical red or green LED has a Vf of around 2 V. Without the extra resistor, the LED current is 100 mA. With the extra resistor in parallel with the LED, there would be 30 mA through the bypassed resistor, leaving 70 mA through the bypassing LED. -ish. Of course, all of this can change depending on the characteristics of the LEDs being used.

As I said, not very efficient, more to illustrate the concept.

A full H-bridge can be more efficient, depending on how the transistors are driven. The dual-555 circuit is very convenient, but when running on 4.5 V I doubt that there is enough differential output voltage between two parts to meet the LED's minimum Vf requirement.

ak

 

[AnalogKid]

Here is the dual-555 version. I still show this as a 9 V design.

By the bipolar LM555 datasheet, when running on 5 V, the minimum output voltage is about +1 V and the max. is about +3.3 V. That leaves about 2.3 V across the LEDs. That is enough for normal little discrete red or green parts, but probably not enough for your Christmas lights.

ak

 

[Digidank]

Is that near Mini too?

Martin

Unfortunately, her too lol

 

[Digidank]

Thank you all so much for the help! I have a project I am working on, but also want to tinker around with all of this. I am going to order some 555's and an LM555 to try out multiple methods. I really wish CircuitLab had an LM555 so I could toy around with it before actually building it. I will let you all know how it goes!

 

[Harald Kapp]

I really wish CircuitLab had an LM555

But it has, see here.

 

[AnalogKid]

To finish up, here is another (((concept))) schematic of a full H bridge driven by one clock source. For this one, a CMOS LMC555 is a much better choice because its output swings much closer to the rails, ensuring that the off transistors are really off without added base pull (up/down) resistors.

I'm not happy with this yet - too high a body count. But it does address cross-conduction during the transitions. For example, when the 555 output is transitioning from low to high, Dr assures that Q2 turns off before Q1 turns on, thus preventing a Q1-Q2 direct short across the rails. Intersil (or whoever owns their semiconductor line these days) was big in bridge driver chips that have this protection built-in. Another option is stepper motor driver chips.

The transistors are rated for 600 mA when saturated, but will fry if they spend too much time in the transition between saturation and off. Specifically, if the square-wave drive from the 555 is disconnected and the entire output stage "floats", it won't float for long.

If Q1-Q4 are replaced with logic-level MOSFETs, then C3-C6 can be scaled to much smaller caps with a correspondingly larger value for R3-R6, such as 0.047 uF and 10 K.

ak