traffic light control: hardware driven

[schmitt trigger]

Here in Belgium it goes in straight sequences: green, then yellow a few seconds (short enough for people trying to get through orange to get caught by the thousands of traffic light camera's to get a fine) and then red.
.

It is called tax- revenue enhancement

 

[dr pepper]

I misunderstood #1, I thought you meant 2 inputs, you meant 1 input 2 states, doh.
In the Uk its:
Red
Red & Yellow
Green
Yellow
Then back to Red.
I didnt think camera's got you on yellow.

 

[mbarazeen]

There was a discussion long ago, i posted a circuit to control lights using basic gates, have a look
https://www.electro-tech-online.com/threads/traffic-light-help-sos.99450/

 

[earckens]

I think this fulfills your explanation.
However your explanation may not be what your traffic lights want to do.
Here in the Uk this is not how ours work.
Anyway:

Hi dr pepper,
there is an issue with the red and yellow LED: they keep blowing because as soon as the npn from digital1 is conducting, all current flows through red an yellow led, no resistance is met.
Can you suggest an alternative for the alternate sinking (red) and sourcing (yellow) of output 3 of the 555?
I tried the design below but both red and yellow turn on when green is off.

 

[rjenkinsgb]

With your attached circuit, the logic input will need to be actively driven (or connected) to either 0V or the 5V supply; eg. a switch or a CMOS logic output.
An open input will give strange results from the look of the circuit.

 

[dr pepper]

Thats what can happen if you design a circuit in your head without testing.
On my hand drawn circuit insert a 1k in series with the Red Led's anode, this might introduce some brightness differences, but maybe not as green Leds tend to be more efficient.

I dont know why the green Led would blow in the circuit you just showed us, there is a 1k permanently in series with it, so it should never see more than a few ma, for that matter cathode current for red has to go through R2 or R7 on your circuit, so how enough ma's gets through to blow the Led's isnt obvious to me, unless maybe the polarity reverses and that takes out the Led, though thats unlikely.

 

[earckens]

Thats what can happen if you design a circuit in your head without testing.
On my hand drawn circuit insert a 1k in series with the Red Led's anode, this might introduce some brightness differences, but maybe not as green Leds tend to be more efficient.

I dont know why the green Led would blow in the circuit you just showed us, there is a 1k permanently in series with it, so it should never see more than a few ma, for that matter cathode current for red has to go through R2 or R7 on your circuit, so how enough ma's gets through to blow the Led's isnt obvious to me, unless maybe the polarity reverses and that takes out the Led, though thats unlikely.

Hi dr pepper, the green led is fine. But when the digital input is high, and green is out, then both red and yellow light up strongly and then the red is bust: so the current -instead of being sinked through pin 3- goes straight through red and yellow led in you intial design.
By the way, I think you did a fantastic job! This is just some fine tuning here.

 

[earckens]

With your attached circuit, the logic input will need to be actively driven (or connected) to either 0V or the 5V supply; eg. a switch or a CMOS logic output.
An open input will give strange results from the look of the circuit.

The input drives the led's fine, except that when pin 3 should be sinking the red LED, the current flows through the yellow LED.

 

[rjenkinsgb]

current flows through the yellow LED

OK.. I'd try moving the base connection of Q4 directly to pin 3 of the 555, so R2 is only in series with the red led.

 

[dr pepper]

I agree, hence the 1 k in series with red between its anode and the emitter.

 

[earckens]

I agree, hence the 1 k in series with red between its anode and the emitter.

Done. But there are a few more issues:
1. in order to trigger pin 3 input 2 must be connected to digital input, not the collector of Q3?
2. pin 2 needs to be high and stay high for pin 3 to go first low then after time=RC go high: so capacitor C1 may be omitted?
3. when the digital input is simulated by a pull down resistor of 1k and a switch is used to pull high then low stays about 0.15V, enough to keep Q2 conducting hence yellow stays on when green is on. Even when forcing digital in to 0,00V the emitter of Q2 measures 4V while it should be 0V (Q2 not conducting): replace with mosfet?

 

[dr pepper]

1) the 555 triggers on a falling edge, Q3 is to invert this and trigger on a positive.
2) for pin 3 to go high as above pin 2 needs to momentarily go low (C1 does this)
3) hmm not sure about that one the circuit initially had 2 inputs, some of the hold on current might be comming through r6, a mosfet might work.

 

[earckens]

1) the 555 triggers on a falling edge, Q3 is to invert this and trigger on a positive.
2) for pin 3 to go high as above pin 2 needs to momentarily go low (C1 does this)
3) hmm not sure about that one the circuit initially had 2 inputs, some of the hold on current might be comming through r6, a mosfet might work.

Hi dr pepper, thanks for your responses! I reworked a bit the circuit, see attached schematic. Never mind the second order input filter, this is to be removed and only a capacitor on the basis is needed (I did this for additional testing for another application).

 

[dr pepper]

Simulate it and see what happens.
what fets 2n7000?

 

[earckens]

Simulate it and see what happens.
what fets 2n7000?

It's been built and it works.
I had these lying around in TO-92; what is wrong with them?

 

[dr pepper]

Dunno.
When you say 'these' I'm not sure what you mean, theres no mention on the schem of part numbers for the fets.
Are you using 2n7000's?
I was just wondering whether you were going to use mosfets or jfets as one inverts and the other does not, but then you did specify mosfets.
Glad to see you got it working.

 

[earckens]

Dunno.
When you say 'these' I'm not sure what you mean, theres no mention on the schem of part numbers for the fets.
Are you using 2n7000's?
I was just wondering whether you were going to use mosfets or jfets as one inverts and the other does not, but then you did specify mosfets.
Glad to see you got it working.

Oh I understand. The mosfets used are 2N7002, N-channel.
Thanks for your help, without your initial drawing I would not have found this.

 

[AnalogKid]

Late to the party. In case anyone is interested in an alternate approach, the requirements in posts #1 and #16 can be met with a single ULN2803 darlington transistor driver plus a few Rs and one C. This includes the LED drivers. The part is basically an 8 section open collector inverter, with each section rated for 50 V and 500 mA. I'll try to post the schematic later.

ak

 

[earckens]

Late to the party. In case anyone is interested in an alternate approach, the requirements in posts #1 and #16 can be met with a single ULN2803 darlington transistor driver plus a few Rs and one C. This includes the LED drivers. The part is basically an 8 section open collector inverter, with each section rated for 50 V and 500 mA. I'll try to post the schematic later.

ak

Interesting, I look forward to that. I have been impressed with your circuits in the past, among them this moisture sensor based on a 74HC132 oscillator, very usefull design!