DIY Audio Effect Pedal Part 3 - Making a low noise supply

[@fredo514]

Thank you for taking the time for sharing this! It’s very useful for learning to start from a real world problem, understand the theory and then observe the fix. It would be really nice if you could go over designing a low noise switching supply.

[@tylermcgrew8335]

Great content, I'm enjoying this series

[@LeonvanBokhorst]

Very helpful. Providing enough inspiration for a few experiments today. I’m going in 🤓

[@FesZElectronics]

Best of luck!

[@elams1894]

Nice vid and well done on the PS as we certainly need more analogue in this world. It astonishes me the amount of digital noise that must be floating around the Ether (can I say that word anymore?). Cheers!

[@24y28]

Thank you for uploading this content, your channel is awesome

[@marctamtonthat]

Is 2 LT3045 a good solution to filter a switching power supply to achieve low noise ?

[@АлександриМаринаСанти]

Hi, thanks for the great video!!! You can share your PCB board

[@p_mouse8676]

In audio applications a perfect stable voltage isn't really mandatory. So I would have gone for a capacitance multiplier with just a zener. The ripple reduction is a lot better (in fact the ripple reduction isn't so great with most voltage regulation, especially for high frequencies) plus you don't have the issue with minimum loads.

[@FesZElectronics]

I would add here that although perfect voltage accuracy is not necessary, you will get some circuits that demand a minimum operating voltage, so you need to make sure that is achieved. But anyway large voltage fluctuations are not really desired. Keeping a well defined voltage will also help in limiting the size of the biggest components - the electrolytic caps - 16V cap is always bigger than a 10V cap. Since this was an external supply (external to the actual device), some noise will still be picked up by the wiring - so the best filter is the one closest to the load - in case of necessity, I would add the cap multiplier right next to the circuit, so the low noise output is also a low noise input. Even if you use a chemical battery with a 1m cable, it will still be noisy.

[@p_mouse8676]

@@FesZElectronics With a zener you will keep everything within ±10% or so? That is less with a couple of capacitors. Especially because the quiescent current of these kind of circuits is bigger than the dynamic current of the signals. (basically a constant current, more or less) So take the higher side of that for your safety margin for the caps. Besides, the voltage of a 9V battery will sag much more significant. 7.5-8V is not uncommon.

[@IvanTarasov]

Would it be practical to try to eliminate 50/100Hz noise with a notch filter?

[@FesZElectronics]

Highly depends on where you want to add the filter - the lowest fundamental frequency of a bass guitar is at ~50Hz and even with normal guitar you can get various effects at low frequency (things like tremolo if I am not mistaken the terms); point is that adding filters in the audio range on the signal line is not a good idea - since this will affect the normal operation. On the other hand if you want to filter the power supply, then you don't need notch filters, but any low pass will do; as long as the cutoff is way below the frequency of interest.

[@IvanTarasov]

@@FesZElectronics yes, I was thinking of the 50/100Hz noise from the low noise power supply that you have. Is there any reason not to include the LPF in it? Or, maybe, on the power input side of the pedal?

[@FesZElectronics]

You don't need a dedicated LPF; I mean, the large electrolytic capacitors are already forming an LPF with the various resistances in the circuit (like the transformer winding resistance) and I also added the active regulator which is providing the same LPF basic function. In the end, you need more filters if you actually have a noise issue, otherwise, there is no point. By the way, the input of the pedal also has some amount of capacitors which are helping with noise filtration.