so i built this project i found on a french site...

[pinkjimiphoton]

this is the project i built, a simple, i mean SIMPLE tremolo.

http://www.techniguitare.com/HOTPROJETS/Tremolo2Transistors.pdf

i figured, what the hell, and built it according to this vero i worked up, which i believe to be right



and here's the schemo:



it lights up the led,  and if i cover the circuit with my hand, i can hear the volume go up, but i'm not getting any oscillation, i don't think.

i know NADA about this stuff, i'm a fuzz guy.. anybody got any ideas?

or should i just flush this project to the bottom of the "oh well" pile?

all the resistors /caps are specced values, i don't have bc 547's, so tried various npn's, no luck tho.

thanks!!

yah, i added reverse polarity, some filtering, that's about it, so the part numbers may be wack. this was just something to work from. wasn't meant for anyone to use, so i didn't clean it up at all.

[digi2t]

From page 4 of the forum thread on this project;

il suffit de changer la 220k par une 100k, celle qui est a coté de on/off , pour faire démarrer l'oscillation !

Change the 220K resistor to 100K to get the oscillation going.

on peut changer la 1k5 qui est a coté du 100uf pour une 2k2

Change the 1K5 next to the 100uF to 2K2.

plus de patate a l'oscillation avec un bc549c

Better oscillation beat with a BC549C.

et si la led eclaire trop changer la 1k5 par une 3k voir 4k

If the LED is too bright, change the 1K5 to 3K or 4K.

voila ca fonctionne , pas encore tester sur du son !

Voila, it works. Not tested with sound yet.

[pinkjimiphoton]

From page 4 of the forum thread on this project;

il suffit de changer la 220k par une 100k, celle qui est a coté de on/off , pour faire démarrer l'oscillation !

Change the 220K resistor to 100K to get the oscillation going.

on peut changer la 1k5 qui est a coté du 100uf pour une 2k2

Change the 1K5 next to the 100uF to 2K2.

plus de patate a l'oscillation avec un bc549c

Better oscillation beat with a BC549C.

et si la led eclaire trop changer la 1k5 par une 3k voir 4k

If the LED is too bright, change the 1K5 to 3K or 4K.

voila ca fonctionne , pas encore tester sur du son !

Voila, it works. Not tested with sound yet.

COOL!!
thanks dino, i'll give all that stuff a try, and report back. appreciate it bro!!

[pinkjimiphoton]

nope. no diff at all.
voltages look like they SHOULD be ok to me, but then, i don't know about oscillators..

batt: 9.08
after 220r "noise" resistor : 8.83

q1
c  4.17
b  1.33
e  .75

q2
c  8.83
b  5.14
e  4.55

led + 1.85

pretty sure this thing would work if i could get the led to flash!

[markeebee]

Posting from my phone, and can't see the whole of the schematic, but the bit I can see looks quite similar to the RG inspired oscillator that Frequency Central used in the Vibracaster.

The V'caster is VERY picky regarding transistors.  It needs a hfe of at least 800 to start wiggling, and even then can be quite fussy.  I've never had much luck with anything other than 547c (the 'c' is important).

Can't swear this is relevant though.......  

[~arph]

It's a phase shift oscillator.. a very tricky pony

[Mark Hammer]

P3 and the LDR form a voltage divider.  How much change in attenuation it provides will depend on the light/dark value of the LDR, whichin turn, will depend on the illumination it receives from the LED.  If the LFO and LED driver circuit are only making your LDR wiggle between 1M and 100k, you're not going to hear very much variation in volume level.  You'll want the LDR to be above the value of P3 at full darkness, but drop below P3's value (47k) when illuminated.

T'were I , I start out by sticking an LED of respectable brightness/efficiency in the circuit, as is, and see just how much illumination, and change in illumination, I'm getting from the stock circuit.

As an aside, it reminds me of the What compressor, that has this whole big sidechain apparatus, and the audio path also consists of a 2-component passive voltage divider.  In theory, that should be clean, clean clean, with negligible hiss.

[pinkjimiphoton]

it makes no background noise, really clean, a bit of compression. the problem is, i can get the leds to light no problem, but i can't get the oscillator to roll. they just stay lit.

when i do manage to get it to fire for a millisecond or three, it seems to sound great!

i used google translate to try and communicate, they seem to be trying to help...i'll check it out over there, and see what they suggest.

thanks for the suggestions... i'm  thinking i may have to get the proper transistors... they mention nte123, which i have. bloody crummy resistors for fuzz, but maybe for this..

[Mark Hammer]

Ah, well that's completely different.  Never mind.


Are you sure you have the pinouts right?

[pinkjimiphoton]

yah mark, i check any oddball things i don't use all the time (and which i keep accumulating for some reason) with my meter to see what it is and what the pinout is.

i have a gazillion transistors but NONE seem to actually work, unless i'm missing something.

i subbed out the 220k resistor for a 100k as advised, no dice. tried using a pot... at certain ranges, the led will go out for a split second, then back up it comes, loud and proud.

it looked like a pretty cool simple circuit, anyways.

[EATyourGuitar]

It's a phase shift oscillator.. a very tricky pony

I was going to say that. it is actually a one transistor oscillator when it works right. there is a lot of information on the internet but I never got one to work in the sim. I was messing with different RC filter frequencies and even tried external excitation through a square impulse or filtered square impulse. never got this thing to work but I did get it to resonate for about 2 cycles after the impulse. it clearly had a resonant frequency but never got it to have enough gain.

[EATyourGuitar]

yah mark, i check any oddball things i don't use all the time (and which i keep accumulating for some reason) with my meter to see what it is and what the pinout is.

i have a gazillion transistors but NONE seem to actually work, unless i'm missing something.

i subbed out the 220k resistor for a 100k as advised, no dice. tried using a pot... at certain ranges, the led will go out for a split second, then back up it comes, loud and proud.

it looked like a pretty cool simple circuit, anyways.

I would trouble shoot it in two parts. first you need to find the sweet spot where the LED will turn off with the switch open. make R9 smaller till the Q2 base voltage drops low enough to turn off the LED. then when you connect your phase shift oscillator, any AC voltage greater than Q2 B should pass the cap and blink your LED on and off. if when you close the switch, your LED stays lit continuously, that means you must have a working oscillator since it is AC coupled by 10uf. if the oscillator is blinking the LED fast enough you might not notice it is blinking. maybe a trimmer on R9 would help you find the sweet spot? you can also decrease R10 to compensate for an oscillator that is low in amplitude but you run the risk of blowing the LED and it will only work if R9 is perfect anyway.

[Keppy]

One observation: C3 in the schematic says 470, not 470n. Might be an error, but have you tried that?

EDIT: 470 what I don't know, if not 470n. 470p? Like I said, just an observation.

[psychedelicfish]

One observation: C3 in the schematic says 470, not 470n. Might be an error, but have you tried that?

EDIT: 470 what I don't know, if not 470n. 470p? Like I said, just an observation.

if you look at the parts list at the bottom of the schematic on the link in the first post, you'll find that it says 470nF x3 under the "condensateurs"

[Keppy]

One observation: C3 in the schematic says 470, not 470n. Might be an error, but have you tried that?

EDIT: 470 what I don't know, if not 470n. 470p? Like I said, just an observation.

if you look at the parts list at the bottom of the schematic on the link in the first post, you'll find that it says 470nF x3 under the "condensateurs"

Ah. Thanks for spotting that.

Well, gain is very important in phase shift oscillators. Maybe you don't have enough. I would try 1) inserting a very high gain transistor (sounds like you've already done this) and 2) increasing the value of the R1/P2 combo. More resistance there should increase the gain. If you get too much, the sine wave will distort, but if you get it working you can always turn R1 into a trimpot to dial that out.

Also, many oscillators would not work in a perfect world, because they rely on some small amount of noise to be present, which is then amplified to jumpstart the oscillation. Your power supply filtering just may be working against you here, and since there are no active devices in the signal path, I'd try taking it out just to see what happens. I'd also like to know if you wired up the switch or not, and if so, did you toggle it? The spike of current from C5 (charged up to 9v with switch open, suddenly pulled down to wiper voltage when closed) might be necessary to get the thing started. These both seem like longshot solutions compared to changing the gain, but hey, you never know!

[duck_arse]

I've built this circuit or one like it, seems a million times. it makes an ugly lumpy sine that fades at one end of the range pot, and I've never seen it with the emitter resistor/cap, and always has a collector bias resistor to b. I'd ground the e.

[~arph]

I'd remove R11 and put it with a higher value directly between B an C of T1.. and ground T1's emitter indeed.

[~arph]

I've built this circuit or one like it, seems a million times. it makes an ugly lumpy sine that fades at one end of the range pot, and I've never seen it with the emitter resistor/cap, and always has a collector bias resistor to b. I'd ground the e.

Yes, seems you suggested the same.  That bias resistor is R11.

[Keppy]

The bias thing you guys are suggesting surprises me. Wouldn't bias from the collector effectively bypass the phase-shift network? A high enough R value would minimize this, but it still seems strange to me.

As far as grounding the emitter, the cap effectively grounds it at AC frequencies. It seems like the designer here was aiming at a more stable bias network, using a resistor divider and allowing the emitter's DC voltage to rise a bit. This way the bias current does not fluctuate as it would with bias taken from the collector. Collector biasing/emitter grounding seems worth a try if you've seen it before in these oscillators, but if there's a reason it's likely to work better I haven't understood it yet. Thoughts?

[~arph]

Yes that resistor should be large. Check out the EA Tremolo, the schaller tremolo or the blue warbler by Jon here. They all use pso's and all have this resistor ( ea i'm not sure )