Maker Pro
Maker Pro

Help with an R2-D2 LED Circuit

LrdSatyr8

Jan 25, 2013
9
Joined
Jan 25, 2013
Messages
9
Greets all... new around here and could really use a little help. I'm building a 1/9th scale R2-D2 robot from Star Wars and have been racking my brain trying to figure out a way to get it to light up properly. I'm using some fiber optics in a fine grid pattern to get the "logic" panels to light up in a random-ish sequence and just can't get he circuit to work. I found a program called Yenka and designed something that works fine there but when I try to test the circuit on a breadboard it don't work. Here's the situation...

On the front and rear of the domed head are 4 areas that I have to light up (also a steady on holoprojector light on the front but I'm not too worried about that). There are front and rear rectangular "logic panels" that will be lit with fiber optics driven by 2 different color sets of 3 LEDs a piece (front being "blue/white/blue" and the rear being "red/yellow/green"). Now there are also 2 round "mood" lights on the front and rear that will be driven by 2 different colored LEDs (front will be "blue/red" and the rear will be "red/green"). The logic panel LEDs should pulse in a trianglur wave pattern (fade in/fade out) at a rate of once every second. The mood lights however will be doing the same only at a rate of once every 20-30 seconds or so (timing isn't critical as long as it's within the ballpark area).

The power supply I want to use will be either 4 AA batteries (6V DC) or a Universal AC Adapter. I'm trying to stay away from IC chips and do everything with restistors/capacitors/transistors for lack of space. I've been trying to get this to work for the past month and still no luck and I'm not exactly sure what I'm doing wrong. So here's the basic scehematic I came up with:


r2-circuitfinal.jpg



I know thre's gotta be someone out there that knows what I'm talking about here and could help me get this circuit working. If you set u pthe circuit yourself on a breadboard you'll see. I'm sure I'm missing something here and hope that someone can help me out here because it's driving me crazy to say the least. I've been blogging my entire R2 build here:

Classic MPC R2-D2 Uber Build

I really appreciate any help that anyone can give. Thanks in advance!
 

Attachments

  • r2-circuitfinal.jpg
    r2-circuitfinal.jpg
    126.3 KB · Views: 148

DuctDuck

Jan 26, 2013
52
Joined
Jan 26, 2013
Messages
52
Hmm, neat, yeah...Uber neat. You are going to fill that R2 Unit until it pops!!!

If you really want to avoid ICs and uCUs </sigh>, you will have to settle with astable multivibrator's predictable pattern. In fact the schematic looks to use that on the lower right side (a capacitor connects Q's collector to another Q's base.
http://en.wikipedia.org/wiki/Multivibrator

I don't know what the lower left side is, sorry.
 

LrdSatyr8

Jan 25, 2013
9
Joined
Jan 25, 2013
Messages
9
Hmm, neat, yeah...Uber neat. You are going to fill that R2 Unit until it pops!!!

If you really want to avoid ICs and uCUs </sigh>, you will have to settle with astable multivibrator's predictable pattern. In fact the schematic looks to use that on the lower right side (a capacitor connects Q's collector to another Q's base.
http://en.wikipedia.org/wiki/Multivibrator

I don't know what the lower left side is, sorry.

Yeah... I was using the astable multivibrator principle and expanded it in the lower left to include 3 Qs instead of just 2. Perhaps I am wrong. Would there be an easier way of doing this with a IC chip?
 

DuctDuck

Jan 26, 2013
52
Joined
Jan 26, 2013
Messages
52
What you are using right now is easy, you know, the bjt multivibrator; a clock signal.

Feed the Vsignal to an IC, counting the pulses is what to do next.

To avoid linear IC regs, use the "4000 logic" family as it has a wide DC supply range (e.g.- 4520, dual 4-bit counter, etc). Try not to sink/source too much current with the Leds/optics. An Led bias circuit connects to one output bit. Some will flash fast and some will flash slow.

-p.s. I found Digi-Key still has 4000 logic ICs...and they have great service here in N.America

-p.s.s This is a great uCU application!
 

LrdSatyr8

Jan 25, 2013
9
Joined
Jan 25, 2013
Messages
9
What you are using right now is easy, you know, the bjt multivibrator; a clock signal.

Feed the Vsignal to an IC, counting the pulses is what to do next.

To avoid linear IC regs, use the "4000 logic" family as it has a wide DC supply range (e.g.- 4520, dual 4-bit counter, etc). Try not to sink/source too much current with the Leds/optics. An Led bias circuit connects to one output bit. Some will flash fast and some will flash slow.

-p.s. I found Digi-Key still has 4000 logic ICs...and they have great service here in N.America

-p.s.s This is a great uCU application!

in my online research into trying to figure this out, I saw something about a 4060 chip having multiple out but the problem was that it could only drive 3 LEDs... I thought about using that, but not sure how to go about it. If it can only drive 3 LEDs at a time that would mean I would need 4 of them... I dunno. Any suggestions?
 

duke37

Jan 9, 2011
5,364
Joined
Jan 9, 2011
Messages
5,364
I cannot see anything wrong with your circuit. The two transistor circuit on the right should be fine, I am not sure of the effect of going to the three transistor circuit.

Beware that there is a strict limit to the emitter/base voltage of a transistor. Your circuit generates slightly less than the battery voltage so is probably OK.

A 4060 is an oscillator followed by many divide by two stages. The current output of such CMOS devices is very low but can be used to drive a transistor for more current.

A 4017 counter has one of 10 outputs activated.

A 4093 can be made to generate four outputs but if you wish to swap from one set of leds to another as your original circuit, then two outputs only are possible.

If you wish to stick with your original circuit, then try a simplified version and then expand it. I do not have a breadboard or the time to do it for you.

I have just had a thought, it hapens sometimes!
If the leds are acting like zener diodes, the collector voltage will not rise (much) when the transistor turns off, thus there will be little drive to the base of the other transistor. Try a 1k resistor from collector to the supply.
 

LrdSatyr8

Jan 25, 2013
9
Joined
Jan 25, 2013
Messages
9
I cannot see anything wrong with your circuit. The two transistor circuit on the right should be fine, I am not sure of the effect of going to the three transistor circuit.

The 3 transister circuit will basically just light each LED in sequence.

If you wish to stick with your original circuit, then try a simplified version and then expand it. I do not have a breadboard or the time to do it for you.

I've been working on that... I've stripped it down to just the 3 LED blue/white/blue circuit... I can get red/yellow/green to work, but not the blue/white combo.

I have just had a thought, it hapens sometimes!
If the leds are acting like zener diodes, the collector voltage will not rise (much) when the transistor turns off, thus there will be little drive to the base of the other transistor. Try a 1k resistor from collector to the supply.

I just tried the 1K resister between the collector and the neg led of the capacitor and nada... same problem. The collector doesn't go thru to the power supply. The emitter goes to the neg lead of the power supply thou... is that what you were suggesting?
 

duke37

Jan 9, 2011
5,364
Joined
Jan 9, 2011
Messages
5,364
I was suggesting a resistor from the collector to the positive of the supply to pull up the collector when the transistor is turned off.

I do not see any part of the circuit which has more than two leds on a transistor.If the leds are in series there will not be enough voltage to turn three on.

Q4 and Q5 have 56k base resistors (R15,R16) which will give very little base and collector current, go down to 22k like the other transistors
 

LrdSatyr8

Jan 25, 2013
9
Joined
Jan 25, 2013
Messages
9
I was suggesting a resistor from the collector to the positive of the supply to pull up the collector when the transistor is turned off.

I do not see any part of the circuit which has more than two leds on a transistor.If the leds are in series there will not be enough voltage to turn three on.

Q4 and Q5 have 56k base resistors (R15,R16) which will give very little base and collector current, go down to 22k like the other transistors

Well I tried the 1K resistor from the collector of all the Qs to the positive and nothing... they all remained lit. Then I attempted just one 1K at a time and all it did was cause the LED that the Q was attached to to go out. Still no flashing. I'm just doing one set of 3 here (blue/white/blue). Any more suggestions?
 

duke37

Jan 9, 2011
5,364
Joined
Jan 9, 2011
Messages
5,364
It is difficult to help at a distance. It seems you are still going with the three transistor cicuit, prhaps we now know what happens. try reducing to two transistors.

You do not say what transistors you have. You could try going to Darlingtons (I use MPSA13) whih have a colossal gain but a bigger voltage drop.

I would separate the oscillator from the driver by using a 4093 chip (four oscillators) or use an oscillator and a Johnson counter (up to ten outputs).
 

LrdSatyr8

Jan 25, 2013
9
Joined
Jan 25, 2013
Messages
9
The transisters I am using are the 2N2222As ... I do have some 2N4401s or 2N3904s... think changing out the transisters might help?
 

LrdSatyr8

Jan 25, 2013
9
Joined
Jan 25, 2013
Messages
9
I tried swapping out the transisters and no luck. Get the same results. No flashing or just 2 will flash a couple of times and then stay on. Like i said before, I'm just testing out the 3 LED flashing segment right now and trying to get it to work. The 2 LED segment works fine but still too fast, but I can't get the 3 to work at all.
 

duke37

Jan 9, 2011
5,364
Joined
Jan 9, 2011
Messages
5,364
The resistors need to be selected to get the DC conditions right so you will need to increase the capacity to lower the frequency. Doubling the capacity, halves the frequency.
 

BobK

Jan 5, 2010
7,682
Joined
Jan 5, 2010
Messages
7,682
Adding a third stage to a bistable mutivibrator does not make it a tristable multivibrator. I tried this in simulation, just to see what would happen. With 2n2222 transistors, 1K collector resistor and 50K / 0.1uF timing compoents, I got a bistable to work as expected. When I add at third stage, it oscillates at 10MHz! Why?

Basically you have 3 inverters with the ouput of one connected to the input of the next and the third one looping back to the first. This is also know as a ring-oscillator and oscillates at about the switching frequency of the inverters / 3.

Bob
 

LrdSatyr8

Jan 25, 2013
9
Joined
Jan 25, 2013
Messages
9
So, even thou the 3 LED oscillation circuit works in simulation, how do I get it to work in reality?
 

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
Jan 21, 2010
25,510
Joined
Jan 21, 2010
Messages
25,510
It may appear to work in simulation, but check the timing, it may be telling you that it's running at a very high frequency.
 

duke37

Jan 9, 2011
5,364
Joined
Jan 9, 2011
Messages
5,364
It did not work in simulation, it oscillated at 10MHz.
I wonder what would happen if an extra stage were added bringing the phase back to the two transistor circuit.

If you want a walking output, use a 4017 Johnson counter
 
Top