Guitar LED circuit sucking tone

[MrStab]

the way i see it, if anything superficial were to harm my tone, then it can GTFO! lol.

sorry, been busy the past day or so. built one of them Tube Reamers to take a break (awaiting aural reviews on whether i screwed it up or not!). anyways, back to work!

so restore the buffer's 10uF output cap with a 10k resistor immediately after? there appears to be space on the board made by fate for this very occasion.

next time i'll come up with a schematic on request instead of saying "that one joined to that one!" lol. my bad.

[MrStab]

put 10uf cap between buffer out & 386 in, and unsure exactly what you meant i tried a 10k to ground both after this cap & at the output cap (separately), in both instances the output is still around 7.2V. pin 3 is down at 0.01, though

[artifus]

what does it sound like? put the multimeter away for a moment. 386 is a little power amp designed to drive a load. do you have a speaker you could connect for a quick listen? an old pc speaker perhaps? is it working? what happens to the sound when i swap/remove/add components?

http://www.ti.com/lit/ds/symlink/lm386.pdf

http://tymkrs.tumblr.com/post/13831067585/on-reading-datasheets-lm386-page-1

i realise that your project is not sonic but sometimes just listening can provide some clues and offer ideas.

[MrStab]

okay, so i'm such an unbelievable genius that i had V+ connected to pin 6 when it should've been 7. and i had read the pinout before this! damnit!

output voltage is now 4.4! input voltage nonexistent. LEDs stay dimly lit. tapping front of newly-reinstated 10uF buffer output cap makes them flash & fade.

i audio probed it prior to fixing the voltage pin & putting the 10uF cap back in, and got signal through, and i had tried with a speaker before but burned out a chip and i'm on my last lol.

in any case, the only problem now hopefully lies between the buffer output & 386 input!

[artifus]

pin 6 power. pin 7 bypass cap to ground.

*also* if you need to knock up a quick scheme online: http://www.digikey.com/schemeit - you don't have to register, just try it now and export image.

[Gurner]

Grant apologies, I think I mapped your circuit wrong in my head - I think the cap was already in place (C2 on your board layout) ....this is where a schem really helps debug  (yes I know you're only linking two elementary circuits, but nevertheless...it can get confusing)

[MrStab]

pin 6 power. pin 7 bypass cap to ground.

how'd you mean? so pin 7 would be connected anyway but via. the bypass pin & a cap? not sure i understand the purpose of the bypass pin then! lol.

getting a flash/fade when i tap the positive end of he buffer output cap seems promising, but say it does work if i take it out (voltage pretty much dies there & then), how would i go about fixing the 4.5v pin 3 voltage?

no worries, Gurner - i think maybe this cap is dead or too high a value maybe (?), but im gonna test with it reoved & jumpered after a wee coffee and a smoke

[artifus]

pin 7 bypass power with a large cap. it helps with ground junk. read the above links. the toymaker articles talk you thru the whole datasheet in plain english with comments and replies.

*i didn't intend that as rudely as it reads back, i'm tired - forgive me. the toymaker articles are quite an interesting read as they are written by someone starting from scratch and asking for help online. you may have to search for the other three pages if they aren't linked*

[MrStab]

admittedly i did overlook the other link as i had read the datasheet & thought it'd be much of the same, but i'll have a look just now. cheers!

[artifus]

something i have learned recently is that you have to read the datasheet again. and again... and again... and again... and again... and... that those guys really knew what they were talking about but were definitely not paid by the word!

[MrStab]

i must've screwed up a solder joint when i put the buffer output cap back in - the circuit now works! i haven't put the power cap in yet, still haven't read the article properly.

i was looking for a shop which had the most obvious sign in large letters the other day, and i passed right by it twice yet somehow managed to process all the names of other shops but not this one. so worringly i've had real-world parallels with this! lol

gonna put the recommended cap in & then try seeing if i can double the LEDs in parallel, i'll let you guys know how it goes

[artifus]

yay!

(i really needed a yay today - nice one)

[deadastronaut]

whoooohoooooo.....

@#$%ng hell it works!!!..yay!!.

now draw a schematic of what you have.....so others can do it too!..

[MrStab]

i'll start work on a schematic once i've managed to sort out the bypass/cap thing, verified that the buffer actually prevents tone suck (just been using a patch lead from a guitar for the time being), and tested 4 blues in parallel.

do i have this right: you don't hook up pin 7 TO pin 6, but instead by putting a cap on pin 7 to ground, you put it into bypass "mode" and power goes in via. pin 6 instead?

[artifus]

do i have this right: you don't hook up pin 7 TO pin 6, but instead by putting a cap on pin 7 to ground, you put it into bypass "mode" and power goes in via. pin 6 instead?

+9v to pin 6. 10uf (or bigger) cap from pin 7 to ground for good measure but not necessary. look at the internal 386 schematic on the datasheet and reread the notes. there is no bypass mode, the cap on pin 7 bypasses an internal resistor, stabilising a voltage divider.

http://www.electrosmash.com/lm386-analysis

[MrStab]

that's actually the very diagram which made me ask what i did. when i said "bypass", i didn't mean what's usually meant round these parts - i meant bypassing that pin for power supply.

i have it working, albeit without the above changes - for some reason that didn't work. it's all routed inside the guitar but the circuit is easy to get to, works quite well actually! just need to try to squeeze out as much brightness as i feasibly can & fix the power supply issue ofc

[MrStab]

here's my dilemma: while i don't doubt for a second that i should be using pin 6 as outlined above, and it was most likely a small, mis-diagnosed error which made it not work for me when i tried it, it just works on pin 7! as the audio pretty much reaches a dead end, and everything seems in working order, wouldn't it be fine to leave it? so long as it's just me taking the risk? however, i don't wanna go putting a schematic out there with something potentially sketchy on it, so if i switch V+ to the proper config, someone else will have to verify it. i'll try to get a video later.

about switching:

using an SPDT switch (well, half of a DPDT), this is how i managed to keep the lights off when the circuit was off (as predictably they stay on if not grounded):



all good? while sound seems more or less fine for the POS guitar i tested this in (has a minor grounding problem to begin with), i wouldn't be satisfied without a full A/B comparison so i'd need to find a way to sever the power if possible. or i could just pull out the battery when playing...

potential problems (may have been mentioned elsewhere): to what extent could a dying battery impact tone-suck-prevention? two batteries is a solution, but that's just excessive. i think that's even the punchline of a joke i once heard. also, turning on & off makes a little high-pitched squeak, not too obnoxious but it's there. not toooo bright using 1k resistors on the LEDs, but could probably improve that by positioning them better and you probably don't wanna risk retinal damage from this, of all things. lol

can't think of anything else for now.

[artifus]

i'm by no means an expert on the 386, i've just tinkered a bit.

the oddest and only non standard lm386 configuration i've seen so far is here:

i've no idea what's going on with pin 6 here. i get the pin 8 to ground oscillator bit (it was listed as a tremolo) but the input and power arrangement is a mystery to me.

as to your tone suckage - i haven't really looked at your circuit, a schematic would help, but i would look at the buffer and how the signal is split for clues. also, decoupling - maybe try two 10 - 100 ohm resistors between +9v and the power inputs to the two chips with a large cap across each to ground. apologies if this has already been suggested and tried.

[MrStab]

(illogical post, since modified)

i've just built the circuit with just the 386 side, and pin 6 works but pin 7 doesn't. weird! your diagram gives me hope for this Twilight Zone circuit, even if it's not the same kind of weirdness. lol


the tone suckage isn't conclusive as yet, could well be subjective. times like these i wish i had an oscilloscope (i volunteer at a charity music shop which has a bunch of oscs just gathering dust downstairs, but it would be sketchy to ask for a freebie!). duly noted on the decoupling - i don't think it's been suggested, but i'll give it a go if further testing necessitates it.

i'll draw up a schematic asap using the link you provided - assuming the tone suck is a non-issue, i'll draw it out with Pin 7 and put a note explaining the situation.

[MrStab]

i'm finding it easier to learn about the 386 in applications, then going back to the datasheet.

i guess the question isn't why pin 7 accepts power, that seems fairly obvious, but the issue is why are the voltages all off when pin 6 is used... looking at the schematic of the 386 itself, wondering how the buffer stage might cause an issue. i suspect some short on my part, but what's really odd is that the voltages when i try to use pin 7 on the 386-only version of the circuit are similar to when i try to use pin 6 on the full thing.

the datasheet says max input voltage is ±0.4V, but if the output voltage from the buffer exceeds that, could it have any bearing...? maybe worth noting that it didn't work without the buffer output cap.