Is it possible to alter these vero layouts for lower noise?

[TOPLEL]

I had some problems with vero my layouts in the past where i had some different circuits on the same vero. (like every time one of my comparators made a gate signal you could hear a pop on my unity gain splitter outputs, simply higher noise levels for the same circuit than in my guitar pedals and similar stuff)

Is it possible to alter these layout for lower noise? (like moving chips around, putting some filter caps here and there etc.)

Of course there might wont be problems at all because i have not built them yet but there always could be interesting ideas on this forum about lowering noise (posts like make the ground path shorter, you must separate them into different circuits, put some caps here and there, make the supply lines wider etc.)

They are mostly simple opamp circuits, i want them to live on the same vero but i made the layout on 3 different days that is why they are separated.



first one is a yusynth white noise\pink noise\random cv generator, i would not want this one to be less noisy ( ) but you can see the input circuitry for the power supply.  (a 10ohm resistor into at least a 100uf cap, maybe 470uf) also it would be cool to if i wouldn hear my gate pulses on the outputs of this one



this one is a lot of simple circuits put together by me, a yusynth glide, a mfos level shifter, an inverter and a audio to cv circuit (basically a rectifier with variable gain)



this is a dual glide comparator based on the lm2903 with one of the inputs splitted with a buffer into 4 outputs

i would just put all those circuits next to each other on a wide veroboard and they would connect to the same power supply line

[TOPLEL]

by some googling i've found the following tips:

- i should not put inputs and outputs next to each other
- i should put holes everywhere to minimize the length the signals run
- i should connect those 2 empty strips next to the ground line to minimize impedance
- maybe i should even ground all unused strips
- i should use that power supply filter again for every new parts of the circuit

[WaveshapeIllusions]

Extra filtering is very important yes. The comparator gate pulses get coupled into the supply because it draws a sudden spike in current when it switches. Bypass caps right next to the supply pins (as close as physically possible) will help. Better would be to use a CMOS comparator if you aren't already. They draw less current and as such will be less noisy.

Random noise is helped by keeping wires short and separating inputs and outputs. A certain amount of separation between separate circuits helps too. Keeping high impedance sections isolated helps too, especially from lower impedances. Noise is easily coupled from low Z to high Z.

Shielding everything is good too. It should be quiter boxed up. If each circuit is on separate boards, shielding them from each other helps. Using shielded wire is a good idea as well. Only ground one end though.

Hope this helps.

[Thecomedian]

Hm. Connecting the empty strips to lower impedance sounds strange. WHere do you find that? I assumed if something isnt in the signal or DC loops it doesn't do anything.

[TOPLEL]

Extra filtering is very important yes. The comparator gate pulses get coupled into the supply because it draws a sudden spike in current when it switches. Bypass caps right next to the supply pins (as close as physically possible) will help. Better would be to use a CMOS comparator if you aren't already. They draw less current and as such will be less noisy.

Random noise is helped by keeping wires short and separating inputs and outputs. A certain amount of separation between separate circuits helps too. Keeping high impedance sections isolated helps too, especially from lower impedances. Noise is easily coupled from low Z to high Z.

Shielding everything is good too. It should be quiter boxed up. If each circuit is on separate boards, shielding them from each other helps. Using shielded wire is a good idea as well. Only ground one end though.

Hope this helps.

i will use a LM2903 http://www.ti.com/lit/ds/symlink/lm393-n.pdf
they seem mention to its low current drain

what i am not sure about is should i put those power supply filters series or parallel? parallel would be the "normal" but that would be harder to make on a vero
i mean series would be practical because then i could do this: (but i am not sure if i can do this or it would cause other problems)



btw i am not really concerned about voltage drops on multiple filter resistors (supply is +-15V and i want max +-5V signals out of it)

Hm. Connecting the empty strips to lower impedance sounds strange. WHere do you find that? I assumed if something isnt in the signal or DC loops it doesn't do anything.

http://www2.ee.ic.ac.uk/t.clarke/projects/Resources/Electronic%20Construction%20Techniques.pdf

page 4

[mistahead]

GROUND UP EVERYTHING NOT CONNECTED AND MAKE IT A SHIELD!!!
Or a GND loop... you know... either way the internet told you to...

Inputs and outs spread out, filter and regulate power, box up and use properley shielded jacks/connections.

Beyond that I've not really heard that vero is "oh so much more" prone to cross talk, RF interference, or noise in general - of course I almost exclusively work with it and less other board types so I would feel that way.

[digi2t]

I had the same dilemma with my Infinitphase vero build. How do I keep signals from different sections from bleeding into the audio. At this point I realized that sometimes you have to sacrifice small board size for isolation. It's the bain of vero, I know, but if you insist on working with vero, then you have to make sacrifices, especially when it comes to cuts and jumpers.

With the Infinitphase, I basically broke the vero into the same sections as the schematic, assigned each section to a portion of board, and then did my damndest to keep the signals from crossing over, under, or bleed next to, each section. This ESPECIALLY applies to the audio signal. The final product looked like this;



The upper section is the LFO/CV section on the upper section, the phase stages on the lower right, and the audio on the lower left portion. The LED driver, and bipolar power supply (not shown) are on separate boards.

And then... even with ALL those precautions, I still needed a small cap in the Rate section to smooth the spikes. So, the lesson here; Plan, plan, plan... and then plan some more. That will go a long way to avoiding any major problems, and most definitely help in solving the minor ones.