555 Based Tremolo Snapping Sound

[Box_Stuffer]

So I put together a tremolo effect that uses the 555 chip to flash the LED and it is picked up by the LDR in the audio section. I got everything put together and all seems to function as it should, but there is a snapping sound that is directly associated with the rate of the 555. This is strange because the 555 section is separate from the audio path although they do share power and ground.

It was making this sound on the breadboard, but it would usually stop if I poked at it with my finger so I thought it would go away when it was hard wired. Not the case.

I know some of you are probably tired of seeing this circuit, but I would like to get it to work correctly.

[antonis]

the 555 section is separate from the audio path although they do share power and ground.

You mean physical or electrical seperation..??

'Cause the later needs more items than shown above..

[ElectricDruid]

the 555 section is separate from the audio path although they do share power and ground.

This is a contradiction in terms!

*Most* problems with noise getting from one bit of a circuit to another happen via the power rails. Not always, but commonly. The 555 is absolutely famous for this.

You could try adding another large cap (although the 1000uF they show is overkill) next to the 555's 9V power, and then feed the 9V to the cap through a 100R resistor. That acts as a power filter for the 555 and should help keep it apart from the audio a bit more. That still does nothing for the ground though, so you'll have to be careful how you route that - take the ground wires for the two sections separately back to the power supply input. Don't share ground wires between the two sections.

[antonis]

I'd also add a capacitor from BJT Collector to GND..

[bluebunny]

Also try using a CMOS 555 so it tugs less current?

[amptramp]

People don't seem to have the same problem with a phase-shift oscillator.

A CMOS 555 driving a small power MOSFET or a darlington output may be the best compromise because it doesn't have the drive capability of the bipolar 555.

[antonis]

I think OP dumped this thread https://www.diystompboxes.com/smfforum/index.php?topic=131005.msg1272067#msg1272067 and is trying to get it from the beginning..

[sinthmart]

I, too, once suffered with a click from the operation of the 555 counter in the tremolo. Couldn't fix anything. And I put the filter in the power supply, and the ground (I made the mass without a circuit (correct distribution of the earth bus. Without a loop)), nothing helped and I gave up.
I also made a tremolo generator on another logic chip (4049 with LEDs and different waveforms). Also clicks. Like a tractor ((Could not fix it either. - https://i.imgur.com/nPUG0qH.jpg

Recently I made another circuit, on a temporary, solderless board, and also with an optocoupler from 555. And no clicks and threshold operation of 555 !! Absolutely. And again I don't understand why.
Maybe you need to put a "crib" (landing connector) under 555 on the board, in case different manufacturers make different 555. And try to stick different ones.

[GibsonGM]

My favorite combination....555 with a 386, ha ha.   I've had luck isolating the LFO from whatever it is driving, and keeping leads to the LDR coming in flat, direct, and away from anything in the signal path.  (shrug)  The low power version as stated may help.

[duck_arse]

....
Maybe you need to put a "crib" (landing connector) under 555 on the board,
.....

IC socket?

[antonis]

IC socket?

Good quality IC sockets are more expensive than 555 timers..

[duck_arse]

IC socket?

Good quality IC sockets are more expensive than 555 timers..

indeed; you know what I always say.

[sinthmart]

IC socket?
[/quote]
Yes, crib


But this is only my assumption that 555 different manufacturers work differently.
Most likely I'm wrong.
This is advice in a hopeless situation)). At least something.

[ElectricDruid]

As others said, you want the CMOS version, usually known as the "7555". It draws a lot less current and doesn't drag the power rails about so much. Less clicks in the first place makes the job of getting rid of them much easier.

Having a look now, there seems to be a NXP part called the ICM7555, and a CMOS version of the Texas Instruments LM555 (which is the part given in the schematic) called the LMC555. Either of those would be a good start.

[amptramp]

If you stay with a 555, you may need a completely separate power supply for it.  Either a different wall wart or a different battery if opto coupling is used and any circuit connection should be only the ground, possibly through a resistor, although no connection should be necessary.

[sinthmart]

If I knew how much torment I would have in developing my circuit with a 555 counter, and if I return to that circuit, then if there is a click again, then I will separate the 555 counter by power supply with a stabilizer 7808 ... 7807 ... 7806 ..
Hoping this will work.
But I remember that all the tricks to improve the power supply did not give results, and sometimes (!!) I got the impression that clicks on the audio part of the circuit were transmitted by a logic chip (I'm already talking about 4049) as if "through the air" - a radio wave. And the distance of the logic chip and its audio part mattered. I advise you not to place them side by side.
Including the possibility of such interaction in an optocoupler (I had a homemade one). Pulses from 555 ... 4049 go to the LED, and the photodetector (photoresistor) is located in the audio part of the circuit. And I suspected such a harmful interaction.
And I put the optocoupler in a tin shield. And moreover, I was even advised to put a metal mesh between the LED and the photoresistor, shielding, connected to ground, ground)). To separate the audio part from the "numbers".
But this is all - the assumptions of an amateur)).

[Box_Stuffer]

My favorite combination....555 with a 386, ha ha.   I've had luck isolating the LFO from whatever it is driving, and keeping leads to the LDR coming in flat, direct, and away from anything in the signal path.  (shrug)  The low power version as stated may help.

I considered using a BJT or JFET in the buffer section instead of the 386, but it would be about the same amount of parts and its not really any cheaper. Plus, you have to worry about biasing and such. The 386 is simple to work with.

[Box_Stuffer]

the 555 section is separate from the audio path although they do share power and ground.

This is a contradiction in terms!

*Most* problems with noise getting from one bit of a circuit to another happen via the power rails. Not always, but commonly. The 555 is absolutely famous for this.

You could try adding another large cap (although the 1000uF they show is overkill) next to the 555's 9V power, and then feed the 9V to the cap through a 100R resistor. That acts as a power filter for the 555 and should help keep it apart from the audio a bit more. That still does nothing for the ground though, so you'll have to be careful how you route that - take the ground wires for the two sections separately back to the power supply input. Don't share ground wires between the two sections.

Would it help to use diodes in series where the power branches off to each IC?  Would it act like a gate to keep the 555 functions from affecting the 386?  Also, in the case of separating the grounds - would it matter which branch would be connected to the jacks and footswitch?

[ElectricDruid]

Would it help to use diodes in series where the power branches off to each IC? Would it act like a gate to keep the 555 functions from affecting the 386?

I can't see *how* it would help. I see the logic of where you're coming from, but in practice, there'll be a certain amount of current coming through the diode, and that amount of current will vary, and that variation in current is where the problem comes from. And I don't see anything about the diode that fixes that.

Also, in the case of separating the grounds - would it matter which branch would be connected to the jacks and footswitch?

Yes, it would absolutely matter. The jacks and footswitch should be connected to the "audio ground" which is a way of saying "the quietest one"! In this case, definitely not the same one as the LFO. I'd connect them to the same one as the amp probably. I suppose you could give them their own branch, but that sounds like it's going to cause more problems than it solves to me, since then there'd be (some - minor?) differences between the amp ground and the audio ground, while the amp is producing the audio! Arggh!

[Rob Strand]

The circuit has so many issues.  You should easily be able to write a list of 10 things wrong.

Here's a basic markup:

- Add 22k to 47k in series with input.
  As is, the tremolo effect will not work well with low impedance outputs like a preceding Boss/Ibanez pedal.
- separate power and grounds.  wire power and grounds exactly as shown.
- add decoupling caps to chips.  must be locate on IC pins.  again following wiring shown.
- add 10uF to pin 7 of LM386 to improve supply rejection (less ticks)
- add 470R in series with speed to prevent NE555 pin 7 shorting supply rail at full speed
- pin 4 on NE555 should connect to +V
- changes to input output caps
- Try pulling the 10uF to see if you can still get unity gain. 
   The lower LM386 gain will help reduce noise and susceptibility to "ticks" on the grounds

A lot more changes can be made.  Including those mentioned in the last thread.
https://www.diystompboxes.com/smfforum/index.php?topic=131005.0

As a side point.  I suspect the 0.047uF tant should be 0.47uF tant.  (The video has a tantalum.)