This was unexpected!
I am combining this one with a very simple LF353 dynamic mic preamp. This effect will be put through (by its coming owner) an insane amount of distortion (Effect>RAT>Metal Zone - all at full blast).
Now when I test it out and put several distortion pedals after the effect the preamp, which is physically located far far away from the oscillators on the vero, not only boosts the mic but also picks up on the three oscillators. I traced and checked every possible issue, solder bridges, touching components you name it. After checking everything trice I put my probe to the V+ rails and lo and behold that was where the oscillators where leaking into from. The oscs are not really audible in normal situations but once you start stacking extreme amounts of gain it gets noticeable in the background.
To be honest I am stunned! How can this happen? Can I filter out the AC signals from the V+ rails somehow?
Mic Preamp
(https://i.imgur.com/RMyNFyT.png)
3x Oscillators
(https://2.bp.blogspot.com/-vlWiNhnkGbY/UHn6e5Z_F1I/AAAAAAAAAIk/NXC5OXvlhJI/s1600/DTD+schematic.bmp)
The layout I've incorporated
(https://3.bp.blogspot.com/-Drp9kifBlFE/UHn6mDIbMeI/AAAAAAAAAIs/WalRQToJfHc/s1600/Devils+Triangle+Drone+Synth.bmp)
it doesnt have an power filtering capacitor on there. (which i find extremely bad design to begin with)
i suggest adding a 22uF or larger electrolytic capacitor from the 9V to GND on the board, this should ''clean up'' the power bus a lot.
cheers, Iain
IMHO, it's more a power supply ripple, mainly caused by op-amp out mixer (1nF cap might load op-amp output) rather than oscilators DC leakage into supply rail..
Try to delete than cap or put a series resistor after 100μF cap, to form a LPF..
Tried power filtering and the scenario is unfortunately unchanged.
Quote from: antonis on May 06, 2021, 03:22:51 PM
IMHO, it's more a power supply ripple, mainly caused by op-amp out mixer (1nF cap might load op-amp output) rather than oscilators DC leakage into supply rail..
Try to delete than cap or put a series resistor after 100μF cap, to form a LPF..
I am not sure what and where you mean? The preamp has, for now, a separate output meaning it is not going to the summing op-amp shown in the DTD. Could you explain again?
Oscilators mixer op-amp output is loaded by 1nF capacitor, resulting into weird behavior..
(power supply current consumption been a part of it, hence inducing ripple via current fluctuation..)
edit: Just realized you've left "open" the unused part of mixer op-amp..
Short pin 7 to pin 6 and connect pin 5 to Vref
https://www.ti.com/lit/an/sboa204a/sboa204a.pdf?ts=1620340666114&ref_url=https%253A%252F%252Fwww.google.com%252F (https://www.ti.com/lit/an/sboa204a/sboa204a.pdf?ts=1620340666114&ref_url=https%253A%252F%252Fwww.google.com%252F)
Quote from: antonis on May 06, 2021, 06:28:53 PM
Oscilators mixer op-amp output is loaded by 1nF capacitor, resulting into weird behavior..
(power supply current consumption been a part of it, hence inducing ripple via current fluctuation..)
edit: Just realized you've left "open" the unused part of mixer op-amp..
Short pin 7 to pin 6 and connect pin 5 to Vref
https://www.ti.com/lit/an/sboa204a/sboa204a.pdf?ts=1620340666114&ref_url=https%253A%252F%252Fwww.google.com%252F (https://www.ti.com/lit/an/sboa204a/sboa204a.pdf?ts=1620340666114&ref_url=https%253A%252F%252Fwww.google.com%252F)
The second part of that op-amp is already setup as a buffer for a filter. The oscillators is not hooked up to the summing mixer either right now. I am in the middle of finishing the build but stopped to check for errors and stumbled upon all of this.
iainpunk already mentioned it. You need good filtering to the LFO circuit. If you intend to use this thing with high gain pedals you have to bullet proof the circuit to prevent ticks getting out. They will get out.
I suggest going a bit further than iainpunk. Feed the LFO circuit from it's own supply like this,
- 9V passes through a 100R resistor then that goes to the LFO power.
- 100uF across the LFO power rail.
- do not let the LFO ground current pass through the preamp grounds.
Connect it back to the where 0V enters the board (ie. make a star ground).
(Same for the 9V to the 100R as well.)
You can also get ticks from the wiring to the LFO rate pots. Those wires contain sharp edges which can get into sensitive circuits. You need to make sure those wires are not near the audio circuits.
Also take a look at C26 and R28 on the Boss BF-2 LFO,
https://www.hobby-hour.com/electronics/s/schematics/boss-bf2-flanger-schematic.gif
Those parts can smooth-off a lot of sharp edges both from the wiring and on the power supply rails.
EDIT:
Couple of other things,
Your preamp circuit is not clear how it deals with Vref. It needs a cap on VREF.
The 1n connected to the opamp output isn't a good idea. Put a resistor in series with the wire between the opamp output and the 1n cap. Say 1k to 10k. You may need to increase the 1n cap somewhat.
Quote from: Rob Strand on May 06, 2021, 09:39:07 PM
iainpunk already mentioned it. You need good filtering to the LFO circuit. If you intend to use this thing with high gain pedals you have to bullet proof the circuit to prevent ticks getting out. They will get out.
I suggest going a bit further than iainpunk. Feed the LFO circuit from it's own supply like this,
- 9V passes through a 100R resistor then that goes to the LFO power.
- 100uF across the LFO power rail.
- do not let the LFO ground current pass through the preamp grounds.
Connect it back to the where 0V enters the board (ie. make a star ground).
(Same for the 9V to the 100R as well.)
You can also get ticks from the wiring to the LFO rate pots. Those wires contain sharp edges which can get into sensitive circuits. You need to make sure those wires are not near the audio circuits.
Also take a look at C26 and R28 on the Boss BF-2 LFO,
https://www.hobby-hour.com/electronics/s/schematics/boss-bf2-flanger-schematic.gif
Those parts can smooth-off a lot of sharp edges both from the wiring and on the power supply rails.
EDIT:
Couple of other things,
Your preamp circuit is not clear how it deals with Vref. It needs a cap on VREF.
The 1n connected to the opamp output isn't a good idea. Put a resistor in series with the wire between the opamp output and the 1n cap. Say 1k to 10k. You may need to increase the 1n cap somewhat.
Rob, thank you! It did wonders.