Author |
Message |
Paradigm X
Joined: Feb 15, 2011 Posts: 363 Location: Null and void
Audio files: 2
|
Posted: Tue Nov 22, 2011 9:45 am Post subject:
Re: A Collection of LED Drivers |
|
|
Tim Servo wrote: | ...snip...
Of course, here’s the classic. An NPN transistor actually supplies the current to the LED, and the amount of current required from driving circuit is minimal. This circuit works well with digital or analog inputs, and is dirt simple. R1 sets the current to the base of the transistor - I’ve used a wide range of values from 10K to 150K. R2 sets the current through the transistor and into the LED. Most LEDs can take up to 20mA (0.020A), but you can usually get away with MUCH less than this and still have a very visible indicator. Use Ohm’s Law (I = V/R) to figure out how much current you’re putting through your transistor and LED. ...snip
Tim (Captain Blinky) Servo |
Hi Tim / Anyone
I breadboarded this (the first) circuit last night, and it worked a treat. But to get anything like reasonable brightness, i had to use a 500ohm resistor for R2, giving about 24mA on 12v supply. The LFO im monitoring has a positive only waveform (tri) from about 1v to about 9v (its a very basic opamp LFO), which is arguably (?) similar to +/-4v, so not far off whats drawn on the schematic you provided. (EDIT for clarity; what i mean is i think its expecting a similar P-p input voltage. I think!)
R1 was 100k. Although its positive only, its also got a diode to ensure this, as detailed in the TH Drum cookbook.
The LEDs seemed to run fine for about an hour or so, but just checking if ive either misunderstood something, or if the 24MA is far too high and theyll blow eventually.
The Blue LEDs required something like 50k (!) to get anything vaguely sensible (viewable!) and i could never get them to match so sacked them off.
Many thanks, ben |
|
Back to top
|
|
|
Paradigm X
Joined: Feb 15, 2011 Posts: 363 Location: Null and void
Audio files: 2
|
Posted: Tue Nov 22, 2011 4:04 pm Post subject:
|
|
|
Hmm actually using 100 ohm; that's 120ma that can't be right ? Somethings wrong here. |
|
Back to top
|
|
|
JingleJoe
Joined: Nov 10, 2011 Posts: 878 Location: Lancashire, England
Audio files: 14
|
Posted: Wed Nov 23, 2011 3:56 am Post subject:
|
|
|
Thats far too much current for the average LED.
If you are having trouble lighting them with the stated resistor values then I'd suggest it is your transistors that are to blame.
Check thier datasheets and lower the value of R1, it's value will depend on how much current the transistor can handle accross it's base/emitter and emitter/collector.
(I'm sure you know ohms law but here are the apropriate formulas I'd be using anyway; I=V/R, R=V/I)
some suggested resistor values for greater LED brightness:
R1: 10 k
R2: 680 ohms |
|
Back to top
|
|
|
Paradigm X
Joined: Feb 15, 2011 Posts: 363 Location: Null and void
Audio files: 2
|
Posted: Wed Nov 23, 2011 9:24 am Post subject:
|
|
|
Thanks.
I thought the 100k was to prevent the LED getting loaded and vice verca, wont reducing it mean itll be affected? (its so annoying not quite understanding things sometimes )
Although i did try increasing the 100k to 220k as suggested by T Henry elsewhere, which made things worse, so youre probably right. Thanks, ill give it a go tonight. Didnt think to reduce it...
My POS multimeter decided to break last night, the hfe tester 'holders' spring appears to have died so cant test anymore. But can easily swap it out.
Although i can measure the current going thru the LED (i think).
Thanks Joe. Think ill rip it apart and start again. The LFO its attached to is behaving strangely too. Some things to try out anyway.
Cheers |
|
Back to top
|
|
|
Tim Servo
Joined: Jul 16, 2006 Posts: 924 Location: Silicon Valley
Audio files: 11
|
|
Back to top
|
|
|
Paradigm X
Joined: Feb 15, 2011 Posts: 363 Location: Null and void
Audio files: 2
|
Posted: Fri Nov 25, 2011 3:16 am Post subject:
|
|
|
Thanks, sorry limited interwebz access
Ok, well it turns out the LFO schematic i was using is very sensitive to anything connected to it. I used another opamp buffer at the end and everything works perfectly as desribed.
Bit weird as i kind of thought that the transistor buffer was supposed to do that, but there you go.
Ive now got the thankless task of wiring up 23 for my midi to trigger converter!
Many thanks for your help/replies.
Ben (do i really need trigger leds?) Paradigm X |
|
Back to top
|
|
|
JingleJoe
Joined: Nov 10, 2011 Posts: 878 Location: Lancashire, England
Audio files: 14
|
Posted: Fri Nov 25, 2011 5:28 am Post subject:
|
|
|
Paradigm X wrote: |
I thought the 100k was to prevent the LED getting loaded and vice verca, wont reducing it mean itll be affected? (its so annoying not quite understanding things sometimes ) |
I know exactly what you mean, thats how I am currently with exponential convertors
The 100k resistor (R1) only limits the current going into the base and out via the emitter, so in an npn transistor this will affect only the current amplified by the transistor and will not add to the collector current.
(I don't know if you need this explanation but it will help anyone trying to understand (including me))
For example lets say you have a tranny with a gain of 100 and input 1ma to the base, then 100ma will flow from collector to emitter and any load connected to the collector will only have 100ma flowing through it. One should also get 101ma flowing out of the emitter. (base to emitter + collector to emitter current)
You can saturate the transistor too (have it fully conducting) and add a current limiting resistor to the load on the collector to limit the current to whatever you get from the equation I = V/R |
|
Back to top
|
|
|
roman_f
Joined: Jul 03, 2007 Posts: 63 Location: Moscow
Audio files: 1
|
Posted: Sat Mar 31, 2012 3:47 am Post subject:
|
|
|
should i lower the 100k if my amplitude is less then +/-5V? |
|
Back to top
|
|
|
mubase
Joined: Mar 24, 2011 Posts: 117 Location: London UK
Audio files: 5
|
Posted: Mon May 06, 2013 11:19 am Post subject:
LFO driver for green/red??? |
|
|
Hi Tim. Thanks for the write up on LED drivers. I would like to ask how it would be possible to drive a bi-color LED..
I have a Weird sound Generator for which I have replaced my old simple tri wave LFO for the LFX 8K LFO. ( 8000 times better...
The old LFO had a single red/green bicolor LED which takes positive V for red and negative for green. So it looked really good. However, I tried connecting it to the triangle wave output whilst modulating a wave on the WSG and just like you said, the wave shape got weaker...
I have looked at your drivers and I see one for 2 seperate LEDs but not one for a bi-color red/green....
Any suggestions?
All the best.
Steve. _________________ There's a crack in everything...Thats how glue manufacturers make a living. |
|
Back to top
|
|
|
Tim Servo
Joined: Jul 16, 2006 Posts: 924 Location: Silicon Valley
Audio files: 11
|
Posted: Mon May 06, 2013 6:00 pm Post subject:
A Collection of LED Drivers Subject description: A driver for three-pin LEDs |
|
|
Hi All,
Circuits for three-pin dual LEDs? Well I just happen to have something.
First off, here's a slightly different design using an op-amp this time. The first is essentially the same as the one presented earlier. However, I build on that for the second design, using a couple of 'steering diodes' to route the current properly through a three-pin bi-color LED. Note that since you have different current paths, you can use a separate current limiting resistor for the second LED. Very handy for those red/blue or green/blue pairs where the blue LED always seems too bright. Note that this design is for a common cathode dual LED. To use this with a common anode part, simply reverse the direction of the two diodes. Also, you can use 1N914 or 1N4148 diodes.
Now with the first two designs, we get back to the same old problem I originally wanted to address with the transistor based versions: I want something that shows the full range of the input waveform, including activity around the 0 volt region. To get around this, I created the third design. This one uses A1 as the standard buffer, and A2 is set up as a comparator. The signal from A2 is added to the signal to bias it up for the positive half of the input, and down for the negative half. This makes the LEDs come on a little earlier (normally they don't come on until you've reached the forward voltage drop - anywhere from 1.7 V for red up to a little over 3 V for blues). Play with the values of R1 and R2 to get the results you like (one example, 4.7 K for R1 gives a smaller region around 0 V where the LEDs are off, although the on/off transition is a little less smooth). Of course, you can use this design with two-pin duals or even two separate LEDs.
Of course, these are good drivers if you don't mind throwing an op-amp or two at the design. Most of the time that's not a problem, but I am working on a dual LED driver using transistors. With two transistors, it should be fairly easy, but I'm going to see if I can do it with one. Stay tuned (also, I usually post these on the Magic Smoke Facebook page first, so you can see them earlier there if you're really in a hurry).
Tim (I'm a good driver a very good driver) Servo
Description: |
Various op-amp based LED drivers |
|
Filesize: |
421.66 KB |
Viewed: |
1844 Time(s) |
This image has been reduced to fit the page. Click on it to enlarge. |
|
|
|
Back to top
|
|
|
promocode
Joined: Sep 03, 2015 Posts: 5 Location: usa
|
Posted: Thu Sep 03, 2015 3:32 pm Post subject:
|
|
|
hi -
i came to this thread to look for a circuit that can drive an LED with a bipolar input. i see what seems to be two solutions: 1. a half-wave rectifying LED driver that only uses one polarity of the input voltage and 2. a circuit that splits up a bipolar input across two LEDs.
is it possible to use a +/-5v signal to drive a single LED from off-to-on?
thanks.. |
|
Back to top
|
|
|
blue hell
Site Admin
Joined: Apr 03, 2004 Posts: 24075 Location: The Netherlands, Enschede
Audio files: 277
G2 patch files: 320
|
Posted: Fri Sep 04, 2015 10:14 am Post subject:
|
|
|
promocode wrote: | is it possible to use a +/-5v signal to drive a single LED from off-to-on? |
Sure, maybe elaborate a bit on the application .. but hen again .. what is wrong with a half wave rectifier? _________________ Jan
also .. could someone please turn down the thermostat a bit.
|
|
Back to top
|
|
|
promocode
Joined: Sep 03, 2015 Posts: 5 Location: usa
|
Posted: Fri Sep 04, 2015 2:17 pm Post subject:
|
|
|
I tend to use bipolar waveforms in my electronics, so with only the positive-going voltages, I'm cutting off a lot of what is happening.
Ultimately, I'm hoping to use this driver for homemade vactrols, which I imagine is one of the main reasons people use them here.. |
|
Back to top
|
|
|
promocode
Joined: Sep 03, 2015 Posts: 5 Location: usa
|
Posted: Sun Sep 06, 2015 10:37 am Post subject:
|
|
|
Is that enough information? I'm still having trouble figuring this out. |
|
Back to top
|
|
|
wackelpeter
Joined: May 05, 2013 Posts: 461 Location: germany
Audio files: 10
|
Posted: Sun Sep 06, 2015 11:54 am Post subject:
|
|
|
promocode wrote: | I tend to use bipolar waveforms in my electronics, so with only the positive-going voltages, I'm cutting off a lot of what is happening.
Ultimately, I'm hoping to use this driver for homemade vactrols, which I imagine is one of the main reasons people use them here.. |
So aren't Vactrols not just an LED attached to an LDR? i've built a Buchla LPG with some selfmade "Vactrols" and it works... the only critical Point was using the right LDR with the needed range... which i picked up comparing the resistor values of the LDR with those of the data sheet from the Vactrols...
it's working for me...
and the vactrols i think are also only working in the positive region...
otherwise some offset calibration could be included i think, but you should take care that you didn't have too much as the total voltage raises not too high... depends on where you would like to use a vactrol as cv Controller... _________________ https://soundcloud.com/bastian-j |
|
Back to top
|
|
|
promocode
Joined: Sep 03, 2015 Posts: 5 Location: usa
|
Posted: Sun Sep 06, 2015 12:34 pm Post subject:
|
|
|
Yeah, my understanding is that a vactrol is an LED and a LDR. General vactrols mainly work for a single polarity and most LED driver designs are for a single polarity (or split between two opposite-going LEDs).
I'm trying to figure out how to drive an LED, with -15v being "totally off" to +15v being "as bright as possible".
Is there a way to do this? |
|
Back to top
|
|
|
wackelpeter
Joined: May 05, 2013 Posts: 461 Location: germany
Audio files: 10
|
Posted: Sun Sep 06, 2015 12:58 pm Post subject:
|
|
|
maybe with an opamp amplifier with negative gain or negative amplification or whatever it is called.. reducing the voltage swing from the +/15 on the Input to maybe +/- 2,5V at the Output... and then some positive Offset to the Output Signal so that it swing effectively between gnd and +5V...
but in that case you would have when you say put an LFO with +/-5V Amplitude into this only an output range of 0V to 1,8V that would drive the LED i think... _________________ https://soundcloud.com/bastian-j |
|
Back to top
|
|
|
promocode
Joined: Sep 03, 2015 Posts: 5 Location: usa
|
Posted: Mon Sep 07, 2015 4:51 am Post subject:
|
|
|
Very smart solution, wackelpeter - it hadn't crossed my mind to do that. I was a little unclear about the last sentence, though. |
|
Back to top
|
|
|
wackelpeter
Joined: May 05, 2013 Posts: 461 Location: germany
Audio files: 10
|
Posted: Mon Sep 07, 2015 6:04 am Post subject:
|
|
|
nope isn't that smart... think i forgot something or took something wrong...
because when you use an fixed Offset the voltage wouldn't swing between GND and some positive when you have less than those +/-15V on the Input... it would swing somewhere between 0V and 5V ...somethinglike +1V on the bottom and +3V on top for example, i think...
this would mean it was glowing at least a bit when it should be turned off...
i would suggest to take a look at the schematics of Thomas White's Buchla LPG Clone or some other Projects including vactrols...
better way would be to use an manual offset via pot at the summing node of the opamp...
i believe there are plenty of other People who can give some better advice than i can do... well, i'm not an expert and only have very small experience in electronics...
if you want o try some circuits you can download an Java based Programm called Jar Circuit or so and can experience with some Solutions...
don't have the link right here at hand and can't look myself how it'S called because my regular Notebook where i had this programm, stopped working last week... _________________ https://soundcloud.com/bastian-j |
|
Back to top
|
|
|
dougcl
Joined: Feb 08, 2008 Posts: 33 Location: Portland OR
|
Posted: Sat Dec 05, 2015 1:07 pm Post subject:
Re: A Collection of LED Drivers |
|
|
Tim Servo wrote: |
Of course, here’s the classic. An NPN transistor actually supplies the current to the LED, and the amount of current required from driving circuit is minimal. This circuit works well with digital or analog inputs, and is dirt simple. R1 sets the current to the base of the transistor - I’ve used a wide range of values from 10K to 150K. R2 sets the current through the transistor and into the LED. Most LEDs can take up to 20mA (0.020A), but you can usually get away with MUCH less than this and still have a very visible indicator. Use Ohm’s Law (I = V/R) to figure out how much current you’re putting through your transistor and LED. For example, in the circuit below with a +15V supply and a 2K resistor, we’re running about 7.5mA (15 / 2,000 = 0.0075A) through the green LED. The protection diode D2 is recommended if the input signal goes more negative than -6V (this is the base – emitter “breakdown voltage” for the 2N3904 and negative voltages greater than this will damage the transistor). |
Doesn't this circuit require that the input voltage at the base exceed both the .7 Vbe drop in the transistor and the forward voltage of the LED to pass current? I was thinking of putting the LED on the collector in series with the resistor, and tying the emitter directly to ground. At least then the LED will start coming on at .7V. Just asking to get feedback on the idea.
Edit: I just bread-boarded this. With the circuit as shown above, the LED starts coming on around 2.6V (of course depends on the LED).
On the other hand, with the LED placed in series with the resistor on the collector, and with the emitter tied directly to ground, the LED starts coming on at about .4V (I am using a 2N3904). Seems to me this latter config is superior for LFO applications, unless I am missing something (quite possible). |
|
Back to top
|
|
|
abelovesfun
Joined: Dec 17, 2012 Posts: 40 Location: Portland, OR
|
Posted: Thu May 12, 2016 4:50 pm Post subject:
|
|
|
How would one utilize a circuit coming off a BC547 and going to a bi color led.
Trying to use a kingston bi color:
_________________ Experimental and modular music at: https://soundcloud.com/she-suburban-heroin-electronics
I make Eurorack DIY Kits (pcb and or panels also available): aisynthesis.com |
|
Back to top
|
|
|
|