Can a LDR/LED be constrained to only output resistance in specific range?

[moid]

so I'll try to make this one and see what happens!

OK well this one doesn't work very well - just finished the circuit, added a battery to test the LED in the above layout and it turns on turns on and then dies (fried LED). As far as I can see I have followed the schematic into my layout (layout just above, the schematic is a couple of pasts higher). Can anyone see what I'm doing wrong here?


OK here's a thought - the original schematic uses 5V, but I just put a 9V into it... could it be as simple as that? In which case I presume I need something like a 78L05 circuit between my 9V socket and the power input of the veroboard?

I've got a couple of 7805's and 78L05's - the difference between them is one is rated for more heat and milliamps than the other. I have no idea how many milliamps this circuit requires, so I guess it is better to pick the larger 7805 to ensure that things are stable?

[ElectricDruid]

Do make sure to get hold of some 7555's instead of the original 555.

Honestly, you can mess around for days and you still won't stop it making everything tick. The 555 is absolutely infamous for crowbarring the supply voltage when it switches, and the low-power variant is much better in this respect. That means the problem is much much reduced at source and you don't have to spend so much time messing about trying to get rid of it.

[moid]

Thanks for reminding me! I got a few 7555s in the post on Friday! Right I'll switch that over now - would the use of the 555 have caused my LED to die? Or is it just a power issue?

[moid]

I've made this layout from this schematic - would anyone mind just checking I did it right? Thanks!

[bluebunny]

The layout looks OK if the part is actually a 78L05 - which is as you've drawn it, but not as you've labelled it.  The higher-current TO-220 part would be the other way round.

Both 555s (bipolar and CMOS) will be fine with a 9V supply, btw.

[moid]

Thanks for noticing this! Sorry about the slow reply, this week has been challenging at work. I hope to get some time away from work tomorrow to have a play with it. If the 7555 can cope with 9V then presumably the LED dying is just too much power getting to it, perhaps I should just add a CLR of some type? 4.7K or 2k2? and therefore not need the power board circuit?

[ElectricDruid]

Dunno. What's the series resistor you've got at the moment? If you haven't got one, then *yes* that's your problem.

9V (even 5V) will fry any LED. You need a series resistor. If you find the forward voltage for the one you've got (so you need either the datasheet or some basic manufacturer's data) there are a ton of calculators online that will help you find a suitable resistor. I generally use a fraction of the "maximum" suggested forward current (If). Most modern LEDs are really bright, and I find a quarter or less of the max current is plenty for me to see if it's on or off. And like that, I don't have spots in front of my eyes for hours after looking at it....

[moid]

Hello everyone

Well I got enthusiastic today and decided to add a CLR to my circuit (instead of making a 9v to 5v power board) and so added a 2k2 resistor as shown below (resistor in red):

The good news - my LED lights up and doesn't explode (yay!)
The bad news - it doesn't fluctuate at all, regardless of what the pots are set to (boo!!!!). There is one position I found where the LED flickers very rapidly but barely perceptibly... which gave me some hope until I noticed the LED for the main board of the circuit (not the above) flickered as well in time with it... so something is rotten in the state of (electronic) Denmark, so to speak.

So my guesswork analysis is by adding the 2k2 resistor I've managed to divert the current directly to the LED and have bypassed the 7555 and the 2N2222... now before I get keen and start making holes in the stripboard and soldering a variety of resistors in places where they probably shouldn't go, I thought, aha! I should check the output of the IC! Surely one of the pins on the IC should be fluctuating to generate the varying amount of current need to dim/brighten the LED... so out comes the DMM and with a cursory glance at the pinout of the 7555 chip I assumed that pin5 (control voltage) would be my friend... but not according to the schematic I used... so that threw me. I'm really not sure which pin is the important one, so I measured all of them, but none are fluctuating at all!

1. 0.002mV
2. 5.95V
3. 0.003mV
4. 6.9V
5. 4.395V
6. 6.16V
7. 13.3mV
8. 6.76V

I'm running this off a 9V battery, but it's not brand new so the voltage may be a little lower than expected - I think it puts out 7.8V at the moment which should still be fine.

I guess I need to remove the 2k2 resistor... I have thought of running the 2K2 from hole J11 to J14 (with a cut in the stripboard so the current goes through the resistor and not along the strip) but if I can't find any fluctuating current anywhere, so that means something more serious is wrong that should be fixed first? Maybe?

Any ideas folks? I have tried swearing at it, by the way. That hasn't worked, but I'm willing to try more exotic and impressive barrages of rude words if anyone has any suggestions.

[duck_arse]

pin 3, the one marked output, on your board - it has nothing connected to it, is this correct? can we again see the circuit diagram you think you've built, pleeze?

your 2k2 to the led, as shown on the stripboard, is connecting the led to the supply, makes all the rest of the circuit ignorable. probably not the best outcome.

[ElectricDruid]

+1 request for a proper circuit diagram. I also agree with DA that you've got the 2K2 and the LED across the supply rails, so the rest of it effectively does nothing.

Working out what the circuit diagram looks like will help you get clear in your head what you actually want to do. It's not just a useful thing to do, but an important discipline to have to keep your thinking straight.

[moid]

Thanks chaps, your help is greatly appreciated! I will remove the 2K2, but I'll need to put it back somewhere to stop the LED from frying. I've socketed the LED so I can try different colours out.

The schematic (not mine) is here:


It too doesn't attach pin 3 of the IC to anything... so maybe I screwed up elsewhere? The only difference I can see is that I'm using 9V and the original schematic calls for 5V, but surely 5V would still be enough to make an LED go bang?

I might not be able to look at the soldering iron until the weekend sadly, too much work on at the moment

[ElectricDruid]

Your strip board has got both the 47R and the 2K2. I thought the 2K2 was supposed to be a better value for R6 to limit the current?

There's lots of things I don't understand about that circuit. What's T2 actually *for*? Why not just put the LED in series with either the emitter or collector of T1? Why do we need two transistors to drive one LED?
Also what's that diode D2 in series with the 100K do? It drops 0.6V, but I don't see the point. The transistor base is controlled by a current, and the triangle output will run from 3V to 6V roughly, so it's still plenty high enough to switch the transistor on.

[moid]

Hi Tom

I'm sorry I don't know how the circuit works, I followed the schematic and made a layout from it and expected it work. It is supposed to enable the LED to have two separate rates - for fade up and fade down to create interesting patterns. There is a video from the chap who designed the schematic here:

I'll remove the 2k2, but I'm still stuck. I've just bought some breadboard, so I'll have a go at making it on breadboard and see if I get some results.

[moid]

OK stupid disaster time. It looks like I linked the wrong video and have been making something that is possibly halfway between one circuit and another. Bugger. Here is the video that is of the actual circuit.

And the page about it is here - this is where I got the schematic from
http://www.pcbheaven.com/circuitpages/555_Breathing_Pulsing_LED/

[duck_arse]

hmm, he has a strange way. it would make more sense to me to use a straight emitter follower where R4 starts, and then an amplifying transistor following to led drive.

[moid]

Thanks for replying Duck. OK so this afternoon I built the circuit on breadboard (brand spanking new clean breadboard, not clogged with dust and tears of despair breadboard!) and... well... I've just made the most complex circuit for making an LED light up No fading in or out at all, just bright strong light. I even built a 5V circuit from a 78L05 just to fit the original circuit design, but twiddling the pots has almost no effect (one of the pots can turn the LED off if twisted to one extreme). It really has been one of those bloody annoying days. I lost on bidding on a pedal and an instrument on ebay (some people just have damn deep pockets) and then while I wasn't paying attention to what I was doing, managed to knock a plastic tray full of all my resistors, diodes, LEDs and knobs to the floor (from the table they were on) and they exploded everywhere... 30 minutes of swearing and clearing them up and putting them back into some sort of order followed. I think I will make some bad sounds on guitar to let the universe know how I feel tonight! As Bob Dylan once put it "There's no success like failure, and failure's no success at all".

So what next? I am using a 7555 chip, would swapping out for an old 555 be worth a try? It would seem illogical to me, as far as I understand they have the same pinout and do the same thing?

Should I try to build a simpler fade in out circuit and see if that works with the original pedal I want to mash this into? If so, I'm open to any suggestions as to which to try. Something with a slow pulse and a pot to adjust the rate of the pulse would be ideal - I might try the circuit in the other video I posted and substitute the 100K resistor for a 100K pot and make the capacitor larger?

By the way Duck, good slogan underneath your post

[11-90-an]

Maybe try an astable multivibrator with 2 transistors? 

[duck_arse]

" ..... not clogged with dust and tears of despair breadboard!" well, there's yer problem.

no, don't change back to 555. in your circuit (not yours) up there, you get a sawtooth wave at the TRIgger, pin 2, which is the cap charge/discharge curve. if you ditch all that D2 and all to its right, you can voltage follower [opamp] that pin - but - you only get a 1/3Vcc to 2/3Vcc out, I think. you could then amplify this to swing more, but you'll run into supply rail one direction and led Vf the other direction.

or - if you just want a triangle-y led drive, look the oscillator circuit in a phase45 fer inst. a dual oppie can do the osc and the led drive no probs. well, we say no probs ......

[kraal]

I've made this layout from this schematic - would anyone mind just checking I did it right? Thanks!

This may look like a stupid question, but why using a voltage regulator IC, with capacitors for this specific usage and not a simple voltage divider (two resistors, 12k and 15k, or any other as long as 9*R1/(R1+R2)=5 ) ?
I know that this doesn't help you much for your root question, but I was wondering.

Regards,

[duck_arse]

if you look the led flash circuit, see R6 at 47R [which I think has to be wrong], when that transistor turns hard on, there will be ~2V across the led, so at 5V Vcc, there will be 63mA at 3V across 47R. now run your resistive divider equation for 63mA, see whether a regulator or the resistors is easier to figure.

someone competent needs to check my theory here - 63mA is too much for that led.


[edit :] HEY MOID! check your leds for function. if your circuit was actually working, and my calc of 63mA was correct, it would take maybe one flash and yer leds ded, maybe. is it possible you think the circuit isn't working when it is actually just destroying leds while you aren't looking? I rekn, at the VERY minimum, that 47R should be 470R, and personally I'd be using bigger still.