smallest 5 led mono vu meter circuit diy?

[Gurner]

edit: tried it with an  072  and using 1n4148, and 1-3 leds for now...didn' get any fluctuation in light when playing guitar though!..

correct diode?

When you say no fluctuation....you mean they're all lit or all unlit?

LED colour will be important (especially if using in series)..what colour LEDs are you using?

[deadastronaut]

yep light up...i'm using 3mm yellow brights...

i'll try some 5mm standard leds...


ive had it  setup with just the one led and then 5 in parallel too......i thought i'd be able to see a glimmer of reaction  but they are full on......

i tried using a 072, 5532, 358.....same..

[Gurner]

The size of LED shouldn't  matter...and yellow LEDS should be fine for in series.

Can you confirm which circuit you are using (it should be one in the java applet)

What happens with no input to the circuit? (with no input, the ouput of the opamp should be 0V, therefore the LEDS should not be lit) ...you might want to put a high value cap in series with the input (before the pot) becuase if there's DC on the input signal, then that would cause the results you're seeing.

[deadastronaut]

hi G:

yep using the java applet circuit...

no difference with input connected or not...leds on all the time..using 072

just tried with 5mm standard yellows, same....no fluctuation..

072 readings

1  2.25
2  2.25
3  0.75
4  0.00
5  3.08
6  3.31
7  5.75
8  9.08

with no input the voltage is the same 2.25...

just to confirm how i have it wired at the mo...i know i haven't the other diodes on there but i thought i'd still get a glimmer of reaction as is..

note, when i put my finger on the input (trim) it does dim...

[Gurner]

which of the two opamps within the TL072 are you using here? (is pin 1 your output to the LEDS or pin 7?)

I'll assume it's pin 1 that's your output.....

Pin 3 should be at 0V with no input....so there's one problem

pin 2 & pin 3 should be exactly the same voltage  as pin 3....so there's another problem. I reckon that something else could be affecting this ....so disconnect any wires feeding the LEDs from pin 1   (so the only connection that pin 3 has is back to pin 2) & re-measure.

Edit: You updated you post with your layout after I asked the above....  

Re pin 3 not being 0V....what voltage reading have you on either side of that input diode?

Also lift that long black wire from pin 1 (to the LEDs) & retake your reading for pins 1,2 & 3.

[deadastronaut]

diode =   0.00- 0.75

lifted wire:

1  2.32
2  2.32
3  0.78

[Gurner]

diode =   0.00- 0.75

lifted wire:

1  2.32
2  2.32
3  0.78

Ok, I've just had a  doh moment  - the TL072 is not rail to rail (I'm so used to working with rail to rail opamps, I overlooked this). the TL072's pin 1 output can't get down as low as 0V for a single supply configuration (i.e. when you're using 0V on pin 4) . The fwd voltage of a yellow LED is about 2V.....so therefore since the TL072 ouput can swing below 2.32V,  the LEDs will always be on.

I don't think there should be 0.78V on pin 3 though (though this is probably due to the TL072 not being 'rail to rai'l again)....if you pull the green wire into pin 3 & it's still 0.78V, then that's the problem

....you need a rail to rail opamp - both on the input & ouput side, (or a bipolar supply...to get you away for now, you could connect -9V to pin 4)

One other option is to use blue LEDs...they need 3.3V, which means they'll be biased off & might work (though then I'd be suspiscious about the other end of the swing ...ie the TL072 can't swing to 9V either!)

Like I said a few posts ago...some opamps might have a few hurdles driving the LEDs here.

[deadastronaut]

pulled the green wire, measures 0.78...hmmmm...

erm...isn't -9v ground?....

[Gurner]

pulled the green wire, measures 0.78...hmmmm...

Then the problem here is solely the TL072 not being rail to rail.

With respect to a quick n' dirty workaround (bipolar supply)....

erm...isn't -9v ground?....

No. Ground is 0V    and -9V is erhm -9V   ...to get a quick -9V, find yourself another 9V battery, connect the RED wire to your circuit 0V line (your blue line on your breadboard) & the black wire to the TL072's pin 4 (but remember to remove the link you have to 0V on pin 4 at present!) ....voila instant -9V (& your opamp pinout voltages will be likely be correct then)

[deadastronaut]

.ok

edit:  your editing is too quick..

tried that, led goes off now...no fluctuation...

[Gurner]

.ok

done it, no difference..

if you mean you have put -9V on pin 4 (vs the former 0V you had on pin 4)  ...can you now post all the pin voltage readings (just pins 1,2,3,4 & 8 )

tried that, led goes off now...no fluctuation...

Ok, this is correct. The LEDs are off becuase in the absence of an AC input signal...the output of the TL)72 is 0V. I touched upon this earlier, there'll be a dead region, between the AC signal arriving & the LEDs turning on (for the LEDS to run on the input cap has to charge up from 0V to near 2V...but that'll take a short while in which time the input signal might cease ...end reult - not lightage!). The reason your LEDs are not turning on is probably because you input signal is not large enough...it needs a large signal at the input (& Na low impedance one too!)

[deadastronaut]

okey dokey, with battery attached as per...

1  0.00
2  0.00
3  0.00
4  -5.68
5  2.18
6  2.63
7  1.88
8  9.10

[Gurner]

okey dokey, with battery attached as per...

1  0.00
2  0.00
3  0.00
4  -5.68
5  2.18
6  2.63
7  1.88
8  9.10

Pins 1,2 & 3 are correct now (it's why I don't always trust simulations!) & the reasons your LEDS are off (& not turning on) are mentioned in my post above. Rather than short pin 2 back to pin 3 (which is unity gain), you could arrange it to give you some gain (have a look at Alex's the earlier post ...the one he made immediately before his sim post....that'll amplify your voltage) The other thing that you could do is what I mentioned yesterday, arrange it so that in the absence of a signal, that it's not 0V on pin, but rather just below the LED's fwd voltage...so for yellow LEDs, rather than have 0V on pin 3 in the absence of a signal, have something like 1.7V ...then there'd be no dead region. I'm also not convinced that the brightness steps between your centre & outer LEDs are going to be that distinct, but that's for a later post!

[deadastronaut]

hmmmmm.......   cheers anyway...

i just found a vu circuit for lm324...but if i'm going with that i might as well persevere with the lm3914...hmmmm..

or stick with my faux/fake 386 vu...for tinyness...anyway i have ordered some of those AN6884'S, so i'll guess i'll just have to wait on them for now..

bloody simulators.. .

[deadastronaut]

ooh ooh...what about this...using 1 string of leds?..and npn?

http://www.eleccircuit.com/led-vu-meter-increase/

edit:just simmed it...all come on at the same time..not vu...

[merlinb]

but if i'm going with that i might as well persevere with the lm3914...hmmmm..

You boys sure are making a meal of this...

Anyway, in case it interests you, here is a proven LM3914/3916 circuit for 9V guitar operation:


In this case it uses 10 LEDs, but you could just use the lower three steps, and put your outer LED pairs in series. To get the right sensitivity you might then want to replace R23 with a pot so you can dial down the input level.

BTW: An LED bar graph is just a bunch of LEDs in one plastic block; there's nothing clever inside.

[Gurner]

I'm not making a meal of anything.... it wasn't even my circuit (or suggestions), I was simply helping Rob to establishing why the sim'ed cct that Alex provided gave unexpected results.

There are heaps of ways of cracking this particular nut (I went the way of my coding my own bespoke PIC because I wanted total control over the LEDs sequences & response ), but if you remember Rob's original requirement, it was low part count...& towards that, I've already suggested a solution which has only about 5 components in total..

[deadastronaut]

ha ha...thats nice and tiny... ...its me who makes a meal of things as a confirmed dumbass ha ha ha...

but seriously, cheers merlin, ...but i'm going to wait for my AN6884'S now as they are suited for just 5 leds..it'll be nice and small too...(as suggested by Gurner)


btw OT a bit.., ive just been messing around with this circuit...i breadboarded it and it works fine with 2 parallel leds on each part too...its obviously not a vu...but its a cool and above all easy
led (chaser) effect none the less..i might use this for my project instead for now
till my chips arrive..

i used 22k's and 47uf's and 3904's...to suit my speed requirements...

http://www.youtube.com/watch?v=YDxLG0hWXMQ

edit this is a cool one too..

http://www.elektronik-labor.de/Lernpakete/Osterei.html

[earthtonesaudio]

Sorry it took me so long to respond.  The reason the simulation doesn't work at 9V with a TL07x is because the default power supply is +/- 15V bipolar.  Just like the power supplies in op-amp appnotes are usually implied.

The ground reference in that sim is therefore the midpoint of the supplies, not the lowest point.  If you swap all the GND points for 4.5V it works, even with a TL07x.

But, since there's no gain it needs a whopper signal.  Here's a tweaked version that works with millivolt signals (although still should be driven by a buffered source for best results) - you could use a dual op-amp's other half for that.

sim

[artifus]

http://www.techlib.com/electronics/bargraph.htm you could also use a 4049 but be aware of the different pin out.

this is worth considering if you have a 49 or 69 knocking about. the first stage can be treated as a tube sound fuzz - simply add input cap, swap input r from 1m to 100k and feedback 1m r for sensitivity pot. maybe play with feedback caps to tailor sensitvity too. could even take an audio out.